visualization of land property for the urban environment · 2007. 4. 18. · • donny scholten,...

Visualization of Land Property for the Urban

Environment

Quaye-Ballard, J. A. March 2003

Visualization of Land Property for the Urban Environment

By

Quaye-Ballard, J. A.

Thesis submitted to the International Institute for Geo-information Science and Earth Observation in partial fulfilment of the requirements for the degree of Master of Science in Geo-Informatics. Degree Assessment Board: Chairman Prof. Dr. Menno-Jan Kraak Thesis Advisors Drs. Barend Köbben Drs. Connie Blok External Examiner Dr. Sisi Zlatanova

INTERNATIONAL INSTITUTE FOR GEO-INFORMATION SCIENCE AND EARTH OBSERVATION

ENSCHEDE, THE NETHERLANDS

Disclaimer This document describes work undertaken as part of a programme of study at the International Institute for Geo-information Science and Earth Observation. All views and opinions expressed therein remain the sole responsibility of the author, and do not necessarily represent those of the institute.

Dedication

This book is dedicated to My father, Mr. Samuel Quaye,

My mother Miss Victoria Naa Lamiley Addy, My only sister and my three brothers

For their love, guidance and assistance, which have brought me this far.

Abstract The mapping of information onto graphs, images or models as a means of visualization can be identi-fied as a powerful tool for data presentation. Currently, visually disseminating property information by real estate agents is mostly done through the display of pictures on hardcopy. This mode of presen-tation most often does not portray the surroundings of the property in question. It is also not in digital format. Demands of clients are often not considered, rather the sale of property by getting clients into the field. A system developed for real estate agents to help their clients visualize properties and their environment will introduce 3D visualization techniques to real estate agents. This research aims at integrating the 2D cadastral map and the 3D characteristics of buildings for real estate agents to help their clients visualize properties in 3D. Clients will thus use a system to visualize properties and its surroundings in 3D. To design a prototype, verbal interviews and litera-ture reviews are conducted to determine real estate agents’ requirements in visualizing property. Re-sult from the verbal interviews indicates that: there should be some awareness of the techniques of visualization that could be used by real estate agents to present the properties. To determine the visu-alization technique, modelling technique and visualization tool for design of the prototype, literature reviews was conducted. The visualization techniques adopted are: geometric modelling, map-based, object-based and image-based techniques. GIS modelling technique comprising the extrusion method was adopted for this research. Height of buildings for the extrusions was measured by Suunto Altime-ter PM-5. Photographs of buildings for photo texturing were taken using the Sony Digital Mavica Quick Access FD Drive 2X camera. The research adopts virtual reality as a visualization tool so that, users could interact with the application, as if they are inside the virtual environment presenting the property. Virtual reality, as a presentational medium offers different views of reality. The softwares used for developing the prototype are: ArcGIS 8.2, Internet Space Builder (ISB), Microsoft Photo Edi-tor, and Cortona 4.0 plug-in for web browser (Internet Explorer). In this research, a prototype has been developed and it is subjected to usability testing, using the ‘think aloud’ method on test partici-pants to determine its efficiency, effectiveness and satisfaction. This is a method which intends to cap-ture what the participants are thinking whiles working. Result from the usability testing indicates that, the prototype is suitable for: realistic representation of the environment; informing one about a prop-erty offered for sale; and identifying properties and their surroundings. Keywords: visualization, virtual reality, properties, real estate agents, 2D cadastral maps, 3D charac-teristics of buildings, 3D visualization.

II

Acknowledgement I am grateful to my former head of department, A.B. Agyemang, of Geodetic Engineering Department, Kwame Nkrumah University of Science and Technology (KNUST) for having nominated me to pur-sue a Master of Science (MSc) course at ITC under the sponsorship of the Netherlands Fellowship Programme (NFP). I specially thank my supervisors, Barend Köbben and Connie Blok of ITC, Geo-information Process-ing (GiP) Department, who directed me at every stage of my research work. I owe the success of this thesis to you. Thank you very much for the success. Special thanks go to those who have contributed in anyway to the preparation of this thesis, especially the following persons:

• Real estate agents, Klinker Makelaars, Zwijnenberg Makelaars and Ten Hag Woningmake-laars, all of Enschede, The Netherlands for their time and participatory efforts in the research interview. Further thanks goes to Eric Klinker (real estate agent), for sacrificing his duties to participate in the testing of the prototype;

• Donny Scholten, for preparing forty-four letters (twenty-two for testing and twenty-two for in-terview) on ITC official letterhead to real estate agents for fixing appointments dates for the interview and the testing of the prototype.

• Marijke Smit and Saskia Tempelman, both of ITC for their effort in helping me with laptop and telephone for the interview. Also, thanks goes to Saskia Groenendijk of Dish Hotel for the unmeasured assistances at the Dish Hotel.

• ITC staffs, Johan de Meijere, Connie Blok, Nicoline Emmer, Corné van Elzakker, Richard Knippers, Boudewijn de Smeth, Walid Belal, Dick van der Zee, and Jeroen Verplanke, who voluntarily participated in the testing of the prototype. Also to ITC staffs that volunteered for the testing but due to other issues could not participate. I appreciate the support and advice I received from ITC staffs: Klaus Tempfli, Rob Lemmens, Wim Feringa, Johan de Meijere, Gerrit Huunerman, Arko Lucieer and Etien Koua;

• Mercy, Harry, Issifu, Atu Hayford and Divine for reading and correcting this thesis. Also to fellow students, classmates and friends who one-way or the other helped in this research.

I thank Rev. Josine, the ITC Christian Community Church, and the Ghana Student Association of En-schede. I thank, Emily Magdalene Duplessis of South Africa for her love, guidance, support and en-couragement during my stay in Holland. I will forever be grateful and may the good Lord hold on you tight. I also appreciate the friendship support from Tine Ningal of Papua New Guinea (PNG). Above all, I express my sincere gratitude to the Almighty God for granting me divine protection, care, grace and wisdom to complete this research successfully.

III

Table of Contents Abstract............................................................................................................... i Acknowledgement..............................................................................................ii Table of Contents ..............................................................................................iii List of Tables .................................................................................................... vi List of Figures ..................................................................................................vii

Chapter One : Introduction ................................................................................. 1 1.0 Introduction ............................................................................................................................. 1 1.1 Background ............................................................................................................................. 1 1.2 Research Problem.................................................................................................................... 2 1.3 Prior Work............................................................................................................................... 2 1.4 Research Objectives ................................................................................................................ 4

1.4.1 Main Objective ................................................................................................................ 4 1.4.2 Sub-Objectives ................................................................................................................ 4

1.5 Research Questions ................................................................................................................. 4 1.6 Research Methodologies ......................................................................................................... 5 1.7 Structure of the Thesis............................................................................................................. 5 1.8 Study Area............................................................................................................................... 6

Chapter Two : Land Related Issues .................................................................... 7 2.0 Introduction ............................................................................................................................. 7 2.1 Some Definitions of Land and Land Related Issues................................................................ 7

2.1.1 Cadastre and Cadastral Maps .......................................................................................... 8 2.1.2 Limitations of 2D maps in the property market .............................................................. 9

2.2 Users of land Property........................................................................................................... 10 2.3 Role of Real Estate Agents.................................................................................................... 10 2.4 Requirements of Real Estate Agent in Visualizing Property ................................................ 11

2.4.1 Literature Review.......................................................................................................... 11 2.4.2 Verbal Interview and Questionnaire.............................................................................. 12

2.4.2.1 Method Used........................................................................................................ 12 2.4.2.2 Formulation and Design of the Questionnaire ..................................................... 12 2.4.2.3 Survey.................................................................................................................. 12 2.4.2.4 Analysis of the Results and Conclusions............................................................. 13

2.5 Why The Need For Visualizing Property? ............................................................................ 14 2.6 Summary ............................................................................................................................... 14

Chapter Three : Visualization........................................................................... 15 3.0 Introduction ........................................................................................................................... 15 3.1 Overview of Some Visualization Issues................................................................................ 15 3.2 Some Definitions and Terminologies of Visualization ......................................................... 15 3.3 Techniques Associated With Visualization........................................................................... 17

IV

3.3.1 Input Techniques ........................................................................................................... 17 3.3.1.1 Geometric Modelling........................................................................................... 17 3.3.1.2 Video Imaging ..................................................................................................... 17 3.3.1.3 Geometric Video Imaging ................................................................................... 17 3.3.1.4 Image Draping ..................................................................................................... 18

3.3.2 Output or Appearance Techniques ................................................................................ 18 3.3.2.1 Map-based (or plan view) visualization............................................................... 18 3.3.2.2 Object-based (or model view) visualization ........................................................ 18 3.3.2.3 Image-based (or worldview) visualization .......................................................... 18

3.4 Scene Components in 3D Visualization ................................................................................ 19 3.5 Summary ............................................................................................................................... 21

Chapter Four : Virtual Reality .......................................................................... 23 4.0 Introduction ........................................................................................................................... 23 4.1 3D Maps ................................................................................................................................ 23 4.2 Virtual Reality ....................................................................................................................... 23 4.3 Definition and Description of Virtual Reality ....................................................................... 24 4.4 Characteristics of Virtual Reality .......................................................................................... 25 4.5 Classification of Virtual Reality............................................................................................ 25

4.5.1 Classification of Virtual Reality With Respect to Immersion of the User .................... 25 4.5.2 Classification of Virtual Reality in Computer Graphics using the S-P- I Model .......... 26

4.6 Virtual Reality and Cartography ........................................................................................... 27 4.7 Virtual Reality Modelling language (VRML)....................................................................... 28 4.8 Summary ............................................................................................................................... 29

Chapter Five : Workable System and Prototype Implementation...................... 31 5.0 Introduction ........................................................................................................................... 31 5.1 Conceptual Stage................................................................................................................... 31 5.2 Testing Stage ......................................................................................................................... 34

5.2.1 Preparing The Data........................................................................................................ 34 5.2.2 Field Observations and Measurements.......................................................................... 34 5.2.3 Problems Encountered in the Testing Stage .................................................................. 37 5.2.4 Results of the Testing Stage .......................................................................................... 38

5.3 Implementation Stage............................................................................................................ 38 5.4 Summary ............................................................................................................................... 39

Chapter Six : Usability Testing......................................................................... 41 6.0 Introduction ........................................................................................................................... 41 6.1 The Method Used................................................................................................................... 41 6.2 Design of the ‘Think Aloud’ Test.......................................................................................... 41 6.3 Implementation of the ‘Think Aloud’ Method ...................................................................... 42 6.4 Analysis and Conclusion from Pre-defined Tasks and Questionnaire .................................. 43 6.5 Recommendations from the Test Participants in the ‘Think Aloud’ method........................ 45 6.6 Summary ............................................................................................................................... 48

V

Chapter Seven : Conclusion and Recommendation .......................................... 49 7.0 Introduction ........................................................................................................................... 49 7.1 Conclusion............................................................................................................................. 49 7.2 Recommendation................................................................................................................... 51

BIBLIOGRAPHY

APPENDICES

VI

List of Tables Table 3. 1: Properties of Plan View, Model View and World View .................................................... 21 Table 5. 1: Outline of the software's used at the testing stage............................................................... 36 Table 6. 1: Advantages and disadvantages of ‘think aloud’ method..................................................... 43 Table 6. 2: Number of participants’ answers for the questionnaire....................................................... 45

VII

List of Figures Figure 2. 1: Land Administration . ......................................................................................................... 8 Figure 2. 2: Unique Parcel Reference numbers ...................................................................................... 9 Figure 3. 1: Visualization process . ....................................................................................................... 16 Figure 3. 2: Plan View, Model View and World View of visualizing property obtained at the testing

stage............................................................................................................................................... 20 Figure 3. 3: Components of 3D Scene .................................................................................................. 21 Figure 4. 1: Schematic illustration of VR regarded as a continuum based upon levels of interaction and

the real world ................................................................................................................................ 26 Figure 4. 2: The S-P-I model for classification of Virtual Reality (VR) in computer graphics. ........... 27 Figure 4. 3: Schematic representation of Virtual Reality and Cartography. ......................................... 28 Figure 5. 1: The inputs for the design of the conceptual, testing and implementation stages. .............. 33 Figure 5. 2: The main stages in the creation of the 3D environment for testing stage. ......................... 35 Figure 5. 3: Demonstration of computing height of building using the Suunto Altimeter PM-5.......... 36 Figure 5. 4: Illustration on how obstructions were avoided. ................................................................. 37 Figure 5. 5: Results obtained at the Implementation stage.................................................................... 38 Figure 6. 1: Mechanism of acquiring user feedback using the ‘think aloud’ method .......................... 42 Figure 6. 2: Number of participants’ answers for the questionnaire expressed as percentage and

visualizing the results using the Pie Chart..................................................................................... 47

VISUALIZATION OF LAND PROPERTY FOR THE URBAN ENVIRONMENT

1

Chapter One Introduction

1.0 Introduction

This chapter is introduced to highlight the general idea of the research problem, objectives, questions, and methodology used for the research. In addition, prior work in relation to the research theme is highlighted. Finally, the area where the research was carried out is also introduced to give an overview of the study area.

1.1 Background

Property, which is either the buildings associated with land or more specifically the legal rights at-tached to land for many years till now have been a basic necessity of humans (Dale and McLaughlin, 1999). As population increases, inevitably leads to increase in both complexity of tasks that have to be tackled and the information that has to be processed and presented for visualizing the environment, buildings, roads et cetera (Zlatanova, 2000). In many cases, the need arises for 3D representation of the environment to which stakeholders such as the real estate agent can interact with. In this respect the real estate agent as well as large private and public construction works can use the map as an index to present all the properties of apartments and offices required for sale and maintenance. As noted in the research by Paintsil (1997) on “3D Topographic data by Aerial Photogrammetry”, ex-isting Geographic Information Systems (GISs) which offer a suitable platform for carrying out such operations are usually confined to 2D and in some cases 2½D where the third dimension (height) is stored as an attribute. This, however, does not satisfy the requirements of the various applications that require 3D information about natural and artificial objects (e.g. the buildings, roads etc.) for their activities. This and other related demands could be satisfied by the acquisition and storage of 3D spatial data in the GIS. Moreover, the demand for spatial information is most pressing in urban areas. It is predicted that at the turn of the millennium almost 50 percent of the world’s population will live in cities; in the United States 75 percent of the population does so already (Tempfli, 1998). He further comments on the promises of 3D GIS to cope with the increasing complex analysis tasks in urban areas. This will offer comprehensive thematic descriptions. Among them are utilizing 3D topographically structured data to analyse spatial relationships; promoting realistic visualization by storing textures; and also it can be interfaced with virtual reality systems (via the internet). In this view, 3D GIS and visualization envi-ronment could serve better than other digital descriptions of a wide range of application from envi-ronmental to real estate management. With recent developments in computers (in terms of power and graphics, hardware performance as well as new software technologies) we experience a switch in architectural visualizations from still images and pre-rendered animations to interactive 3D model and virtual reality (Bourdakis, 1997). In this view, visualisation in 3D (including planimetric and height views) portrays the structure of the urban environment such as buildings. Linking images to the map will offer different views of reality. Use of 3D maps is increasingly becoming popular as a quest for more realistic representation of spatial


2

data increase. The visualization of land in a visual environment will help stakeholders (users) develop and share insights into thinking and creativity of analysing data for decision making (Cartwright et al, 2001). The authors continued by saying, the database is consulted when one needs to know where land property is located as well as the surrounding description of the land property on the computer display. Virtual Reality is one of the techniques for representing 3D spatial data. It enables complex details of the real world to be visualized by utilizing the human ability to navigate through familiar environ-ments and fully interact with spatial information of various types (Ogao, 1997). One way the technique is made available is by the Virtual Reality Model Language (VRML). With VRML, the environment can be created for presentation and exploration, for example using virtual reality. Instead of reading a description of the environment it is possible to navigate through it and see the information (Horne et al, 1995). In addition, individual objects such as building models can be created and then manipulated with the motive of producing elevation views of the object. It is as well possible to manipulate the object in real time and for that matter view it from any desired angle. Visualization of the data and model outputs needs to be comprehensible not to only experts but also the general public who are increasingly becoming important in decision making about the future (Tang and Bishop, 2002).

1.2 Research Problem

As illustrated above, 2D maps do not portray all the features of the urban area such as buildings. The third dimension is left out in display. Thus an effective and efficient means is required to portray the third information on 2D maps, especially 2D cadastral maps per say. This research focuses on: • Visualizing land property in the urban environment using the technique of Virtual Reality whereby

the 2D cadastral map and the 3D characteristics of the building are integrated on a computer dis-play and

• The creating of 3D virtual environment for visualizing land property for the real estate agents. In this respect, the linking of images to geometry would facilitate the development of a system for real estate agents to help their clients.

1.3 Prior Work

There have been a number of 3D related researches and projects in the past and recent both at ITC and the research community at large. An attempt is made to review some of the selected research studies: Kraak (1994) in “Interactive modelling environment for 3D maps: Interface function issues” dis-cusses the characteristics of a prototype of a cartographic modelling system that can handle 3D data linked to a digital terrain model. It enables the user to relate 2D maps with corresponding terrain sur-face, thus facilitating discovery and better understanding of relations between spatial datasets. Ogao (1997) in “Visualization of 3D Spatial Data using Virtual Reality Modelling Language (VRML)” studies the suitability of VRML as an implementation tool for Desktop Virtual Reality (DVR) for cartographic visualization purposes. It looks at how VRML satisfies the cartographer’s needs, for instance tools that enable an effective construction of 3D scenes and further utilities to en-able aspects of cartographic visualization to be achieved while at the same time allowing for a wide


3

exchange and dissemination of spatial information. This studies the effective means of interpreting 3D scenes on the web. Bourdakis (1997) in his research “Making Sense of the City” mentions freedom of movement within the data as a major feature of Virtual Reality based system. It is possible to experience the data from different viewpoints, camera angles, and heights creating a more inclusive map of the information in-volved. He notes further that, the fly navigation mode turn into advantages when large amount of ab-stract data are visualised. One can zoom into an area, get a higher resolution representation and still move back and have an overview of the whole area using familiar metaphors (walk, fly and examine). Tempfli (1998) in “3D topographic mapping for urban GIS” referred to the design of mapping from images in the urban environment especially buildings. It involves the extraction of building from im-ages and the use of VRML in visualising the extracted data in 3D. Zlatanova (2000) on “3D GIS for Urban Development” reports on the definition of conceptual model that is capable of handling the variety of objects of interests for urban planners in a way appro-priate to analysis and interactive 3D visualisation employing current technological developments. She employs VRML to develop a front-end user interface. The paper “Automatic Reconstruction and Visualization of a complex Buddha Tower of Bayon, Angkor, Cambodia” by Gruen et al (2001) reports on modern techniques of analytical and digital photogrammetry to device photo realistic 3D model of one of the very complex towers of the famous Bayon temple of the city of Angkor Thom. This high quality model was then subjected to visualisation and animation. The ultimate goal of this activity and investigation was to develop a system that is ca-pable of producing high quality photo realistic 3D models in fully automated mode. The final models can then be connected to VRML for viewing with standard visualization packages. The research by Tang and Bishop (2002), “Interaction Methodologies for Interactive Forest Mod-elling and Visualization” gives an overview of the visualizations currently used in forest management and examines existing systems using those visualization. The research identified the scope of visuali-zation: what can be achieved in terms of realism, interactivity, integration with other spatial and non-spatial data sources, and links to process modelling. They further illustrated on the review options for integration of visualization and modelling with GIS as well as to set criteria for an effective end-user product. Stoter (2002) in “3D Cadastres: State-of-the art” reports; with increasing pressure on land comes a growing interest in using space both beneath and above the surface. That is how properties superim-posed on top of the other should be reflected in the cadastral Database Management System (DBMS); as properties in strata, as a set of 2D layers or as individual 3D parcels? The final solution would be the definition of 3D parcels or 3D legal objects within the current cadastral system. The Cyclorama developed by CycloMedia International is a digital panoramic image which allows literally look around 360 degrees (source: http://www.cyclomedia.nl/uk/index0.html). The Cyclorama, which is obtained by digitising photos, offers a method by means of which large numbers of fish-eye photos can be made of urban landscape in a fast and efficient manner. Although it is a 2D visual repre-sentation of the environment it perfectly matches the human abilities of interpretation, providing a complete picture of the environment. An entire environment could be projected into your office; where


4

one can look around in 360 degrees from every position on the map; and look at various elevations of buildings through administrative applications. In addition, the article “Ons Huis scoort met digitale diensten” by De Twentsche Courant Tubantia (2002), reports on the need for digital computer envi-ronment for the dissemination of property information to buyers who want to buy properties through the web. In the above-mentioned current researches, little attention was focused on the area of land property especially for the real estate agents where 3D data of buildings are added to 2D maps for visualization. This is the area the research seeks to attain. In the paper “Visualizing the world” by Fairall (2002), he did mention, for many people a house plan is just a series of box. To see a landscape or a kitchen is a mental leap they cannot make. In addition science and art off visualization is putting the missing third dimension back into mapping. Also little attention was considered for the real estate agents as an application group in visualizing properties.

1.4 Research Objectives

The theme (or idea) of the research is to use visualization techniques to visualize the third dimension of buildings in the urban environment to assist in the dissemination of building information by real estate agents. The term real estate agent is defined as: ‘a person who is authorized to act as an agent for the sale of land and buildings’ (Online Dictionary.com, 2002). To realise this, the theme is decom-posed into the following main and sub- objectives:

1.4.1 Main Objective

• To integrate the 2D cadastral map and the 3D characteristics of the building for visualization.

1.4.2 Sub-Objectives

• To determine needs or requirements of real estate agents for visualizing geo-spatial data. • To determine the limitations of 2D maps in visualizing land property for the property market. • To find out how to geo-reference the photographic images taken on the field. • To determine how to add 3D building data to 2D maps as well as the best available method for

obtaining 3D virtual environment. • To find out the visualization technique that is available and suitable to visualize the proposed

system (that is the 3D environment) as well as developing a prototype. • To test the suitability of a prototype.

1.5 Research Questions

Based on the theme of the research and the objectives stated above, the following questions are what the research seeks to address:

1. What are the needs or requirements of real estate agents for visualizing geo-spatial data? 2. What are the limitations of 2D maps in visualizing land property for the property market? 3. How to geo-reference photographic images taken in the field? 4. How to add 3D building data to 2D cadastral maps as well as the available method for obtain-

ing 3D virtual environment? 5. What visualization technique is available and suitable to visualize the proposed system? 6. How to conduct usability test for the proposed system?


5

1.6 Research Methodologies

The following tasks or methodologies would be used to help answer the research questions stated above: Task 1. Analysing the requirements. This task would be used to answer questions1 and 2. Comprehensive literature review and verbal in-terview would be used to investigate and analyse the real estate agents requirements of visualizing property. The verbal interview will be conducted by interviewing the target group, the real estate agents. ITC staff at the Department of Urban and Regional Planning and Geo-Information Manage-ment would also be interviewed since they always confront the real estate agents. Also the limitations of 2D maps for the property market would be reviewed. Task 2. Data acquisition and usage The purpose of this task is to answer question 3. The task would consider the direct acquisition of (data on the field) with digital camera bearing in mind, geo-referencing of images and camera stations. Task 3. Method for obtaining 3D virtual environment This task will tackle question 4. That is, comprehensive literature review would be undertaken or car-ried out with the view of finding the best method to use to create the 3D environment. This would also focus on how to link images to geometry. That is linking the acquired photo images to geometry to get the 3D environment. Also how to touch up photos for correcting the image content would also be dealt with. Task 4. Developing and testing the prototype This task would be used to answer questions 5 and 6. The task would involve looking for the best tools and sensibly combining them to get a working system as well as get a prototype. Also the efficiency and effectiveness of the proposed system for the implementation of the final product would be tested. The test would be conducted through usability test where the target group would be staff of ITC and real estate agents.

1.7 Structure of the Thesis

To achieve the goal of the research, the study has been arranged in the following structure: Chapter one: Introduction. This outlines the content of the research with highlights on research prob-lem definition, prior work, objectives, questions and methodology. Also the application group, the real estate agents, would be introduced. Chapter two: Land related issues. This chapter would dwell on land related issues such as properties, Land Administration, cadastres and cadastral maps. In addition, review on the role and needs of the real estate agent would be introduced. The survey interview for real estate agents requirements in visu-alizing property would be introduced. Chapter three: Visualization. This chapter would review visualization and various aspects of visuali-zation such as map-based visualization, object-based visualization, image-based visualization and 3D visualization. Chapter four: Virtual Reality. This chapter would review the basic concepts and overview of Virtual Reality. This is introduced as part of visualization technique for presenting the proposed system. Chapter five: Workable System and Implementation. This chapter would illustrate the concept, acqui-sition and reconstruction behind dataset. This chapter would be concerned with also the implementa-tion and results of the prototype. Chapter Six: Usability testing. This chapter would be concerned with the usability testing of the pro-totype. Feedback from test participants, using the ‘think aloud’ method would be mentioned.


6

Chapter Seven: Conclusion and Recommendation. In this chapter, concluding remarks of all the pre-vious chapters would be summarised and recommendations for further work are presented.

1.8 Study Area

The chosen study area for the research is part of Enschede municipalities in the province of Overijssel of the Netherlands. It lies between latitude 51° 13' N and 53° 13' N and longitude 5° 53' W and 7° 53' W. The area consists of well-structured commercial as well as residential properties. It is predomi-nantly a Central Business District (CBD) and it is noted for its growing business and commercial ac-tivities. It has a strategic location, with proximity to the railroad station where numerous newcomers and town folks disembarked. Many offices of the city and the county are located here. Well-developed and improved building facades, streetscapes, street-side parking lots, expanding commercial facilities such as the Holland Casino and other developments are also part of the area features. The area was chosen for the study based on the above information and more importantly due to easy access for data collection with the intention of visualizing the area in 3D for the real estate agents.


7

Chapter Two Land Related Issues

2.0 Introduction

This chapter tackles the requirements of real estate agents in visualizing geo-spatial data as well as the limitations of two-dimensional (2D) maps in visualizing property for the property market. These were achieved by a literature review and verbal interview. Interviewing practising real estate agents and ITC staff at the Department of Urban and Regional Planning and Geo-Information Management contrib-uted to the achievement of the verbal interview. The chapter starts with discussion on some definitions on land and other land related issues followed by users of land property, role of real estate agents, re-quirements of real estate agents in visualizing property and finally on why there is a need for visualiz-ing land Property.

2.1 Some Definitions of Land and Land Related Issues

Some definitions associated to land are given here to explain the concepts used in this research. These definitions are derived from Dale and McLaughlin (1999): • Land is the physical thing that encompasses the surface of the earth and all things that are attached

to it both above and below. • Property is either the buildings associated with land or more specifically the legal right attached

to land. • The Land Administration involves the processes of regulating land and property development

and the use and conservation of the land, the gathering of revenues from the land through sales, leasing and taxation, and resolving of conflicts concerning the ownership and use of the land.

Adding to the above definitions, the United Nation’s Ad Hoc Group of Experts on Cadastral Surveying and Land Information Systems (1985), defined land as ‘an area of the surface of the earth together with water, soil, rocks, minerals and hydrocarbons beneath or upon it and the air above it’ (see Henssen, 1996). Dale (1994) described land as ‘the source of all material wealth. From it we get many things that we use or value, be it food, clothing, fuel, shelter, metal, or precious stones. We live on the land and from the land, and to the land our bodies or ashes are committed when we die. Land is also an abstract thing that is manifested as a set of rights to its use with a value that can be traded even though the physical object cannot be moved’. On this view, the availability of land is the key to human exis-tence, and its distribution and use are of vital importance. Property signifies dominion or right of use, control, and disposition, which one may lawfully exercise over things, objects, or land. One basic characteristic of property is that of real property and personal property. The term real property refers to land, which in its general usage includes not only the face of the earth but also everything of a permanent nature over or under it (see Legal Information Institute, 2002). It includes for example structures and minerals. The personal properties are properties other than real property consisting of things, which are temporary or movable such as furniture or a car. Based on the preceding discussions, land would be considered or assumed in this research to include buildings.


8

The term Land Administration is applied to refer to the public sector whose activities are required to support the alienation, development, use, valuation and transfer of land (see Figure 2.1). The basic building block in any Land Administration System is the cadastral parcel. Poor Land Administration is an impediment to the growth of an economy. For example banks are hesitant to meet the needs for fi-nancing without security of title because of the higher costs and more significant risks. A good Land Administration contributes to economic development in a number of ways. For instance it provides security to investors and permits real estate to be traded in the market place. Land Administration is introduced in this research because real estate agents need to refer to Land Administration for their activities. Also the Land Administration requires property boundary information for its activities. Ca-dastral maps are further elaborated in the next sub-section.

Figure 2. 1: Land Administration (Source: Dale and McLaughlin, 1999).

2.1.1 Cadastre and Cadastral Maps

The Cadastre is an information system that uses land parcel or the cadastral parcel, which ‘is a uniquely delimited tract of land within which a coherent set of definable interests is recognised’ (Dale and McLaughlin, 1999). The authors further added; Cadastre is a basic unit to register land ownership. According to the International Federation of Surveyors (FIG) statement on Cadastres, ‘a cadastre is usually, and in most countries, a parcel-based and up-to-date land information system containing re-cords of interests in land that is rights, restrictions and responsibilities. It usually includes a geometric description of land parcels, linked to other records describing the nature of the interests, the ownership or control of those interests and often the value of the parcel and any improvements that have been made to it’ (see Stoter, 2002). The cadastre could be seen to represent the Land Administration system. Cadastral map plays an important role in the Land Administration system as it shows the boundaries that define the land property or ownership (Kraak and Ormeling, 1996). Cadastral maps are large-scale maps normally at a scale between 1:100 and 1: 2500. It consists of outlines of buildings and parcel identification numbers or unique parcel reference number (UPRN) used to identify each land or prop-erty unit. Figure 2.2 shows the organisations involved in the use of the unique parcel reference num-ber for financing, surveying, physical planning as well as seeking information about landowners. In this research, the cadastral map forms the base data for the 3D visualization of the land property in the

Development

Utilization

Transfer

Alienation

Valuation

Land Administration

Land and Real Property


9

computer environment. Further discussions on 3D cadastres and limitations of 2D cadastral maps are introduced in the next sub-sections.

Figure 2. 2: Unique Parcel Reference numbers (Source: Dale and McLaughlin 1999).

2.1.2 Limitations of 2D maps in the property market

This section discusses some limitations of 2D maps in the property market as mentioned in the re-search question. • According to Stoter (2002), 2D maps do not portray vertical dimensions of the earth surface, for

example constructions built on top of each other, apartments, subterranean railways and roads, as well as cables and pipelines. Legal boundaries, which are usually fixed in 2D space although, seem sufficient for current land registration. But may prove incapable of ensuring legal security in the future. For instance, buildings have a 3D component, which needs to reflect in the Land Ad-ministration system. This research does not consider solving this limitation. Current 2D cadastral system does not reflect the vertical dimensions of buildings. This research does consider solving this limitation.

• For the property market 2D cadastral map do not portray the looks of buildings for the real estate agents in the sale of property to their clients. 2D maps portrays surface data, cadastral maps do not have height data assigned to it and as a result it is assumed that parcel boundaries are all located on surface level. Hence, only 2D cadastral maps will not properly visualize the environment for the real estate agents. In addition, paper maps do not possess the possibility of interacting with objects on display. The limitation concerning the looks of building is what the research aims to solve.

Summarizing, 2D cadastral maps do not portray or represent the realistic view of buildings in the true environment. 2D cadastral maps are, however, easy to comprehend when the amount of data presented is reasonably less, but as the need for detail grows, the 2D cadastral map becomes difficult to under-stand. It is of this view that a workshop was organised on ‘3D Cadastres’, in November 2001,Delft, The Netherlands, to consider the 3D issue of cadastral registration in an international context. To rem-edy such limitations of 2D systems, researchers are exploring the use of 3D Virtual Reality. It is of this view that the research seeks to integrate the 2D cadastral map and 3D characteristic of buildings for

Unique Parcel Reference Number

(UPRN)

Queries about Land and Property

Ministry of Physical Planning

Surveys and Mapping

Ministry of Justice

Ministry of Finance


10

realistic views and interaction for the real estate agents. Users of land property are an important topic to discuss. This is discussed in the next section.

2.2 Users of land Property

According to Koehne and Howard (2001), the ease of access to property information provides benefits to the general public and business community including real estate agents, conveyancers, valuers, sur-veyors and others by placing information at users fingertips. This simply means the users of land prop-erty are the real estate agents, conveyancers, valuers, surveyors as well as the landowners of the gen-eral public. On this view, visualization of land property would help the public and private set-ups such as the real estate agents. The role of the real estate agent is introduced in the next section.

2.3 Role of Real Estate Agents

This section is based on the findings on Employment Development Department (2002) and Bureau of Labour Statistics (2002). The real estate agent is the person who is authorized to act as an agent for the sale of land and buildings (Online Dictionary.com, 2002). Real estate agents assist in the process of buying and selling real estate for their clients. Other terms used in place of real estate agents are real estate broker, realtor, estate agent, land agent and house agent. The real estate agent performs an es-sential service for both the buyer and seller by providing an orderly transfer of the property involved. Real estate agents perform several tasks to complete transactions of properties. They help buyers find the appropriate property that meets their needs and financial resources. For instance, they obtain prop-erty listings and make preliminary estimates to determine the selling price of a property. To anticipate prospective buyers questions, real estate agents must thoroughly be familiar with the physical condi-tion as well as other characteristics of the property involved. Once a buyer is found for a property, the real estate agent prepares a purchase agreement. Real estate agents may also refer clients to lawyers and tax consultants to resolve legal and tax issues that may arise in sales transactions. Real estate agents can specialize in selling apartment buildings, residential, recreational, commercial, industrial, or farm property. Whatever the speciality, real estate agents must have knowledge of the state of property concerned. Real estate agents’ current role in visualizing property is through the means of hardcopy pictures. These pictures of properties are shown when clients visit their offices. The computers are mostly used to generate lists of properties for sale, their location and description. It is also used as a computer net-work to link digital images they have in their offices to the Web and to other organizing bodies. In the Netherlands, these organizing bodies collect all the images and information’s of properties from regis-tered real estate agents, and then publish these images and information’s on their magazines. An ex-ample is the NVM Makelaar magazine. The computers are not used in visualizing properties real estate agents have for sale in their offices. It is of this view that the research aims at developing a system for the real estate agents to help their clients visualize properties and its surroundings on the computer using virtual reality. Investigation of real estate agents’ requirements for visualizing property on the computer in their offices is one of the objectives in this research. These requirements are elaborated in the next section.


11

2.4 Requirements of Real Estate Agent in Visualizing Property

To tackle the requirements of real estate agents in visualizing property, literature review and verbal interviews were conducted. These are summarized in the next sub-sections.

2.4.1 Literature Review

Literature on the requirements of real estate agents in visualizing property is rather sparse, as a result literature review was made on users requirements in visualizing scientific data. Conclusions were then drawn for real estate agents’ requirements in visualizing property. This difficulty could be attributed to the fact that, researches on visualization requirements are mostly centred on: visualizing scientific datasets and city models, visualization tools and visualization interfaces. The reviews made are as fol-lows: With reference to Andrienko et al. (2002) in “Testing the Usability of Interactive Maps in Com-monGIS”, for visualization tools to be effective, users expects: to know the purpose of the visualiza-tion tool; to have training; education of the visualization tool; and a simple interface. Adding, users expect to understand the purpose of the visualization tool and learn more on how to use them. According to Ogao (2001) in “Collaborative visualization of geospatial data in the WWW”, users expect to have direct and interactive interface to their geographical data in the use of visualization tools. That is, users expect direct and simple interface when visualizing their data. Rather than many complex interfaces. According to Slocum et al. (2000) in “Map time: Software for exploring spatio-temporal data as-sociated with point locations”, users expect interactive visualization techniques to be easy to apply. The author further added; ‘users may find new interactive techniques difficult to apply and will not derive the full benefit from them, or simply not utilize them if cognitive and usability issues are not considered’. Zlatanova (2000) in “3D GIS for Urban Development” added; realism (that is, the rep-resentation of physical objects, such as roof material and street surfaces) is required in visualization. With reference to Kraak and Ormeling (1996) in “ Cartography: Visualization of spatial data”, there are some general questions that could be asked, which in the long run would help determine the user requirements in visualization. These questions are: Usefulness (that is, is the application or soft-ware compatible with other software packages to allow for data exchange, for example on the Web) and User friendliness (that is, how easily will the new user be able to work with it? and what level of training is required?) From the literature review, it can be deduced that with respect to visualization, users such as real estate agents would expect: visualization tools to be easy to use; simple visualization interface; the ability to plug in their own software; and to visualize their data in their own office and on the Web. To supple-ment this general overview of user requirements, verbal interviews were conducted on practicing real estate agents. The verbal interviews were conducted such that a face-to-face talk could be established between the researcher and real estate agents to acquire real estate agents requirements in visualizing property. This procedure and results of the verbal interviews are illustrated in the next sub-section.


12

2.4.2 Verbal Interview and Questionnaire

Adding to the literature review mentioned above, verbal interviews were also conducted with few real estate agents and ITC staff at the Department of Urban and Regional Planning and Geo-Information Management. The designs of the questionnaire for the interviews were based on the objective (that is, real estate agent requirements of visualizing property) of the interview, determination of the target group and developing well-structured questions. Although there exist many designs, closed questions were formulated. This design was adopted for easy analysis of the results. The method used, the im-plementation of the survey, formulation and design of the questionnaire and analyses of the survey results are as described below.

2.4.2.1 Method Used

Interviews and questionnaires were used for the sample survey. The interview method was chosen be-cause the compilation of data is easy (Krathwohl, 1993). The author further added to the critics of in-dividual interviewing survey that, it ensures that questions are correctly understood; it reduces ‘I don’t know’ responses; and it provides depth explanations to the pre-coded questions. The questionnaire used exists of closed-ended questions in which the questions are ‘structured’ and ‘totally structured’ types. ‘Structured’ types consist of questions that are determined and the interviewer codes responses as they are given. ‘Totally structured’ types consist of questions that the coding is predetermined and the respondent presented with alternatives for the questions so that the phrasing of the responses is structured. These types of questions were chosen for easy coding and easy analysis of respondents’ feedback. In this method of survey, it was realised that more relevant information were obtained from the interview although it was labour intensive, slow and long sitting in questioning and coding. This technique requires training. The design of the questionnaire is described in the next sub-section.

2.4.2.2 Formulation and Design of the Questionnaire

Although the prime aim of the survey interview is based on finding the requirements for visualizing property in the urban environment for the general public in times of sale of properties, it was also cen-tred on gathering information to help build the prototype for the real estate agents. It also served as a pre-test questionnaire. The formulated questions were classified under the following headings (see also Appendix G): • Some Statements, which are statement-type questions for determining how images of properties

are presented to real estate agents’ potential clients. • Current System, which are questions related to the current way real estate agents visualize prop-

erty for sale. • Wishes and Needs, which are questions posed to determine the wishes and needs of the real estate

agents. • Final Question, which is a question posed to determine whether real estate agents are interested in

participating in3D visualization. The way the survey was carried out is described in the next section

2.4.2.3 Survey

To investigate practicing real estate agents visualization requirements, formal letters on ITC template were initially posted to twenty-two real estate agents asking their permission for an interview. Out of the twenty-two letters sent, two responded. These letters were sent to allow the real estate agents show their interest before the interview. After they have showed a positive interest, a telephone call was made to them to fix dates and time for the interview, as they are very busy people. Unlike the real es-


13

tate agents, letters were not sent to ITC staff. During the interview a copy of the questionnaire was given to the respondents (real estate agents and ITC staff) as I read through a copy of the question-naire, at the same time recording and coding the real estate agents feedback and relevant comments. Also, I demonstrated an early prototype to the respondents during the survey to introduce them to vir-tual reality and the intended objective of the research. In addition, one real estate agent participated by answering the questionnaire without been interviewed. The results and conclusions from the survey interview are highlighted in the next sub-section.

2.4.2.4 Analysis of the Results and Conclusions

The analysis is based on the results from the formulated questions used in the survey interview. The results were obtained from interviewing one ITC staff and two real estate agents. Results were also obtained from one real estate agent who answered the questions without being interviewed. This was due to limited time he had. Result and conclusions obtained from the results of the survey are as fol-lows. The following results were obtained from one ITC staff: • The opinion is that the real estate agents are satisfied with their current system although the 3D

approach will one time come to play to supplement existing on. All that they want to present is the pictures of the building nothing more. Real estate agents always wish to go to the field to show the properties they have for sale.

• Current 3D representations of buildings are not true pictures of the existing buildings but rather artificial images on most web sites. They would prefer a true picture of the buildings rather than artificial ones. An alternative to this application would be the oblique aerial photograph, which would portray a true picture of the environment. However, the photograph would lack immersive interaction.

• Real estate agents will prefer a system that is cheap and affordable. The following results were obtained from three real estate agents: • Real estate agents agreed that potential clients want to first see images of properties at their offices

before they visit the property. Buyers want to see what real estate agents have for sale before visit-ing the property itself. This mechanism according to real estate agents is an effective way of offer-ing property for sale to their clients. According to real estate agents, a 2D image of property does not convey enough information about the property’s site and surroundings although it is their nor-mal procedure of presenting property information to their clients. As a result, some real estate agents uses video coverage’s shown on video cassette recorder (VCR) at their offices as an addi-tional tool for presenting property.

• Real estate agents use a hardcopy picture of the property they have for sale for their clients when they visit their offices. Use of a softcopy picture on the computer screen to visualize properties is not part of their normal duties. The scanned images or downloaded images from digital cameras are put into the computer for other purposes, not for their clients. However, a form of image en-hancement is applied to images for re-sizing and also to correct image sharpness.

• Although real estate agents are content with their current system, they wish to link images of prop-erties to cadastral or real world location as well as to link images of property to representation of the geometry of the property site and its surroundings. They wish to present properties in three-dimension although they have little idea about visualization. In addition, real estate agents who have little idea about virtual reality think an interactive 3D visualization of property would be an ideal and interesting way for presenting their properties for sale.


14

Concluding, real estate agents wish to use interactive 3D visualization system, as they claim it would add another taste and functionality to their current system. This shows the necessity for 3D visualiza-tion of property for the real estate agents. A conclusion drawn was that, real estate agents require a very attractive presentational medium of conveying property information to their clients. They wish to present property surroundings to their clients when they visit their offices. Real estate agents have vague idea about the importance of visualization. It was realised during the interview that, discour-agements could happen since the real estate agents are business-oriented, as a result do not see the sig-nificance of visualization. From the survey, I could conclude that there is opportunity to raise the im-portance of visualization to real estate agents. That is to say, there should be awareness of some of the techniques of visualization that can be used by real estate agents to present the properties they have for sale to their clients. User (client of real estate agents) demands in visualizing property is not consid-ered much by real estate agents rather the sale of properties by using ‘feelings’ of the client. Intended system would be technically promising in the ongoing research into 3D Cadastres in various countries.

2.5 Why The Need For Visualizing Property?

Visualization is ‘a science and art of putting the missing third dimension back into mapping’ (Fairall, 2002). In this research, visualization of property would focus on the exterior looks of the property and it’s surrounding in time of sale of the property whereby the user can interact with the medium display-ing the property. Thus, the interactive medium of presentation and realistic environment would offer aesthetic views of the property. Users of maps are now more demand focused and expect linking maps to real world structures on a computer display. In addition, 2D cadastral maps portray surface data for which 3D features of properties cannot be seen. Height data needs to be assigned to a surface data for realistic visualization of the environment as surface data without it gives the impression that parcel boundaries are all located on the surface level. This calls for the need of visualizing property in 3D for which the research is based. Adding 3D features of properties to the 2D cadastral maps for visualiza-tion would help real estate agents present properties to their clients.

2.6 Summary

Summarizing, the chapter highlighted the importance of 3D visualization of properties in the urban environment. The chapter also highlighted issues relating to land. The chapter also raises the need for 3D visualization of the real estate agents for their clients. Verbal interview conducted helped to de-termine real estate agent requirements in visualizing property. These requirements as well as visualiza-tion, virtual reality and the approaches for generating the 3D model (as would be introduced in the subsequent chapters) helped in developing the prototype for the real estate agents for their clients in visualizing properties.


15

Chapter Three Visualization 3.0 Introduction

In this chapter, a review is made on visualization techniques, which would be used in achieving the visualization technique necessary for the research. In other words, tackling the research objective of finding the visualization technique that is available and suitable to visualize the propose system as in-troduced in the previous chapter. The chapter starts by introducing an overview of some visualization issues, then discussions on some definitions and terminologies as used in visualization. Visualization techniques such as map-based visualization, object-based visualization and image-based visualization are addressed. Level of detail and scene components in 3D visualization are also highlighted.

3.1 Overview of Some Visualization Issues

According to Löffelmann (1998), ‘for many years up to now maps are used as visualization of geo-graphic data. Techniques like colour coding, height fields, iso-lines, and icons, are used to show topog-raphic information such as mountains and rivers. In science large collections of data that has to be processed requires a medium of presentation. Usually it is not suitable for human researchers to inves-tigate such data sets by reading lists of numbers or other textual representations. The mapping of in-formation onto graphs or images as a means of visualization can be identified as a powerful tool for data investigation. In a nutshell, maps serve as means to make visible or present spatial phenomena for the end user’. Also, as mentioned in chapter two 2D cadastral maps are used to present or display property boundaries such as buildings. This research mainly addresses 3D visualization of land prop-erty in the urban environment employing the 2D cadastral map and virtual reality, which will be dis-cussed in depth in next chapter. In 3D visualization, the notion is mainly the introduction of a third-dimension (that is, height information) incorporated in the visualization process to achieve realism of the real world in the computer-environment (Fairall, 2002). Adding to the above discussion, the research is based on achieving a goal of representing buildings in 3D for the real estate agents to help their clients. Nevertheless, the most interesting aspect is about the techniques of visualization and how best to address its needs in visualizing buildings. In this respect 3D visualization will be of very important use, as it will allow the development of models that will show very realistically how a building in the real world will look like in the computer-environment. The advantage of this way of doing things is that, it would help as well as make it possible to place buildings in its surrounding for presentation behind the desktop computers. Understanding visualiza-tion requires some basic knowledge of the definitions and terminologies associated with visualization. These are outlined in the next section.

3.2 Some Definitions and Terminologies of Visualization

Before introducing definitions and terminologies of visualization, it is worth noting that the aim of visualization technology is to take information and give users much better idea about what the world described by data would really look like. Many definitions and descriptions exist in defining and de-scribing visualization. For instance, Fairall (2002) defines visualization as the science and art of put-


16

ting the missing third dimension back into mapping. Zlatanova (2000) describes visualization as a general term to denote the process of extracting data from the model and representing them on the screen. Kraak and Ormeling (1996) defined visualization as the use of computer technology to create visual displays, the goal of which to facilitate thinking and problem solving. Jiang (1996) describes visualization as mental process for obtaining spatial information, with the aim of finding patterns using visual thinking. Adding to the definitions and descriptions, Visvalingam (1994) tried assigning specific meanings to Visualisation, Visualisation in Scientific Computation (ViSC) and Visualization by avoid-ing confusion in their usage. The author explains that Visualisation is primarily a mental process, which serves a variety of purposes, including visual analysis. The visual analysis refers to the use of visualisation as a distinct method of inquiry for provoking insight and for concept refinement; ViSC is the discipline concerned with developing the tools, techniques and systems for computer-assisted visu-alisation. ViSC ‘studies the mechanisms in humans and computers, which allows them to perceive, use and communicate visual information’ (McCormick et al., 1987); and Visualization is simply a process. The process involves ‘a series of transformation that convert raw simulation data into a dis-playable image. The goal of the transformations is to convert the information into a format amendable to understanding by human perceptual system’ (Haber and McNabb, 1990). In summarizing the above definitions and descriptions, the basic idea or goal of visualization is that the interpretation of the results should effectively and accurately represent the original data set. The research adopts the definition of visualization by Fairall (2002), as the research seeks to integrate the 3D characteristics of buildings to 2D cadastral map for the real estate agents to help their clients. One primary issue that needs to be considered to achieve the goal of designing visualization for a system is the process used in creating the visualization. Brodlie (1994) diagrammatically illustrated the visuali-zation process as depicted in Figure 3.1 below.

DATA IMAGEGEOMETRYMODEL

��

��

��

��

Map to abstractvisualization object

��

DisplayedImage

Figure 3. 1: Visualization process (source: Brodlie, 1994).

The input data is basically from measurements, which is the initial step; followed by the model which represents a construct of the underlying entity attributed to the input data; followed by the geometry which involves the logical step required to represent the object geometrically where a particular visu-alization is chosen for the representation of the input data; and lastly rendering an image from the geometrical representation that is giving a pictorial interpretation of the input data. Zlatanova (2000) also added the process of visualization and interaction can be completed in two steps, which are known as pipelines, that is the input and output pipelines. The input pipeline comprises the detection of users actions and the corresponding post-processing of the model. The output pipeline comprises the process of sending information onto the screen. Some of the techniques required for the visualiza-tion process are illustrated in the next section.


17

3.3 Techniques Associated With Visualization

To get a clear and general idea of the techniques in visualization a link was made to the approaches of forest visualization to help in reviewing and identifying the visualization techniques. The section will be based on the findings of Tang and Bishop (2002), McGaughey (1982) and Verbree et. al (1999). Before introducing the techniques, it is interesting to note that modern visualization tools range from digital maps, simple 3D perspective diagrams to complete virtual realities. Also, the role of visualiza-tion in spatial applications is not only limited to the representation of tangible objects such as proper-ties as in this research but extends to intangible objects such as numeric and statistical analysis. In re-spect of the above-mentioned link, various approaches of visualization can be identified and these are: geometric modelling, video imaging, geometric video imaging and image draping, which could gener-ally be applied to any application. In addition, within the 3D modelling procedures of visualization it is worth distinguishing what is map-based (or plan view) visualization, object-based (or model view) visualization, and image-based (or worldview) visualization. These techniques and presentational modes in visualization are discussed in depth in the next sub-sections.

3.3.1 Input Techniques

This section reviews on the input techniques required for the visualization process. These techniques are as follows:

3.3.1.1 Geometric Modelling

Geometric modelling methods build geometric models of individual components such as ground sur-face, trees and structures. In the forest perspective, it assembles the component models to create an image of a forest stand or landscape. In addition, in the built or urban environment it assembles the component models needed to create a model of buildings. Scenes depicting the complete model are rendered from a variety of viewpoints as indicated in section 3.4. In its simplest form, this technique can be used to generate perspective drawings showing typical GIS data coverage such as roads, build-ings, streams, and polygon data overlaid onto the ground surface.

3.3.1.2 Video Imaging

Video imaging uses computer programs to modify scanned full-colour video or photographic images to represent changes to the landscape conditions. Video images produce television-quality or better full-colour visual representations that depict current and future conditions. Video imaging typically requires a library of images representing different conditions to replace portions of an original image. Also direct manipulation of images is possible.

3.3.1.3 Geometric Video Imaging

This approach combines geometric modelling and video imaging techniques to produce very realistic images that accurately represent the data, describing the effects of the application involved such as buildings or forest management activities. Operators use geometric modelling to produce scenes that specify the location, arrangement, and scale of proposed landscape changes. Video imaging is then used to modify a digitised image to reflect these changes. The technique can be extended to include geometric modelling to determine the locations for digitised images. Hybrid approaches result in im-ages that accurately reflect the data describing proposed changes.


18

3.3.1.4 Image Draping

Image draping mathematically ‘drapes’ an image over a digital terrain model and then renders the re-sulting scene from a variety of viewpoints. Operators usually obtain the image from a satellite scene, aerial photograph, orthophoto, or map sheet and use techniques common to video imaging to modify the original image to reflect management activities. Several GIS and image processing applications provide draping capabilities. Most include rectification procedures to properly orient and align a digi-tal image to the ground surface. Simple applications utilize orthophoto images that have already been registered to the ground surface and corrected for elevation, or relief displacement.

3.3.2 Output or Appearance Techniques

This section reviews on the output or appearance of visualizations that Verbree et al (1999) named as views and describes them as the modes for modelling and visualization. With reference to results ob-tained at testing stage in this research these views are demonstrated in Figure 3.2. These view modes are achieved by interaction, which is associated with different visualization techniques (see Table 3.1). These interactions could be distinguished into orientation and navigation, selection and query, and manipulation and analysis.

3.3.2.1 Map-based (or plan view) visualization

In this visualization, data are displayed thematically and a two-dimensional map is the main display medium. That is a 2D GIS interface where the GIS data are displayed in a map format. The user is able to create and manipulate the objects as 2D symbols and simple 2D vector or raster representations. Users could also access the standard GIS functionality through graphical interaction and alphanumeric queries. Navigation is achieved through panning and scrolling. See Figure 3.2 (a) for illustration on map-base visualization.

3.3.2.2 Object-based (or model view) visualization

This visualization involves the use of computer-generated symbols to represent actual objects, for ex-ample the use of pre-defined objects represent roof of buildings or trees. This view is typical in many GIS applications, such as Arc GIS and Arc View software’s. A typical example is the 2½D view gen-erated with the Arc GIS 3D Analyst extensions in this research. Here the view manipulation involves moving the object and not the camera making it difficult to get inside the environment. Visualization is achieved through a bird’s eye view; such that is the user looks down the model as if it is a 3D model (see Figure 3.2 (b)).

3.3.2.3 Image-based (or worldview) visualization

This is also known as photo-realistic visualization. The term ‘photo-realistic’ describes a still image created by rendering a 3D geometric model, which is of high quality, and comparable to an equivalent photograph captured by a conventional camera (Sithole, 1997). In this view the user sees the model from a certain position within the model. This strategy of visualization involves a texture mapping of which photo images are draped onto a 3D geometry (see Figure 3.2 (c)). The various techniques identified help in tackling the research objective of finding out the visualiza-tion technique that is available and suitable to visualize the proposed system. The identified techniques as applied in forest visualization could also be applicable in this research. This research is confined to possibilities of geometric modelling obtained from the extrusions of the 2D cadastral boundaries of individual parcels resulting in full three-dimensional buildings. The technique adopted involves mixed


19

approaches of both geometric modelling and image draping as photo images were draped (textured) on the extruded 3D models. These techniques were chosen because an aim was to use the images of prop-erties on the extruded model by photo texturing. Video imaging and geometric-video imaging were outside the scope of this research objective. Multiple view approach (comprising the plan, model and world views) was adopted due to stages involved in transforming the 2D data via 2½ D to 3D model. All the views incorporated in this research are illustrated in Figure 3.2. In this research, the plan view helped as a base data to work on, which were displayed on the computer and on paper. The model view is the extrusions from the footprints of the 2D view. The world-view consists of the images tex-tured onto the 3D model, viewed in the virtual environment. Thus, the sequence of representation in achieving the objective of integrating the 2D cadastral map and the 3D characteristics of buildings shifted from 2D (plan view) to 2½D (model view) to 3D (model view). From Figure 3.2, figure (a) is the 2D view as presented in ArcMap, figure (b) is the extruded 2D view into 2½ D view presented in ArcScene that has to be exported to 3D view (figure (c)) for rendering in Cortona plug-in for web browser (Internet Explorer). In the model view (that is, the 2½ D), the 2D polygons of buildings were extruded using the computed height measurements (see Appendix D). To achieve an interactive photo-realistic 3D, level of detail is worth to be introduced. This is because; the scene (that is, properties and the environment) has to be rendered without significantly degrading the quality of the rendered scene. Also, too many details would tend to hide the information one is in-terested to display. The concept of level of detail makes highly complex scenes to be displayed feasi-ble (Zlatanova, 2000). The level of detail was, however, not introduced in this research since a small working area was considered in developing the prototype.

3.4 Scene Components in 3D Visualization

In 3D visualization, there is the need to mention the scene components as they combine to give the final 3D scene. Interestingly, the readability or legibility of the scene depends on the components used to render the model or part of it on the screen. According to Zlatanova (2000), scene components are required in 3D visualization and interaction for rendering of images on the computer screen or in other words to create realistic scene. The scene denotes the rendered image on the screen. The author further explained the components necessary to achieve the readable 3D images, which are shown below (see Figure 3.3): • Geometry: The geometry refers to the representation of geometric characteristics of objects. In this

research the geometry is the block-like extrusions representing the building. • Illumination and shading models: These control the lighting from artificial sources and

corresponding reflections on the surface. • Colours and Textures: This represents the surface properties of the object. • Camera Position: This specifies the location of the user inside the model and orients the model

accordingly.


20

Figure 3. 2: Plan View, Model View and World View of visualizing property obtained at the testing stage.

(a) Plan View presented in ArcGIS

(c) World view presented in Cortona 4.0 (Web Browser)

(b) Model view presented in ArcScene (ArcGIS)


21

Figure 3. 3: Components of 3D Scene (Adapted from Zlatanova, 2000)

Table 3. 1: Properties of Plan View, Model View and World View (Source: Verbree et al, 1999)

Model Visualizations Navigation Selection Manipulation Analysis

Plan View (2D)

2D geometry attribute val-ues

Cartographic Pan and zoom Specify po-sition

Pointing Query Distance

Create Remove Translate Rotate

Buffer Overlay Network Proximity

Model View (2½D)

2D geometry extrusions by attribute val-ues. Multi-TIN surfaces

Extruded 2D geometry. Cartographic 3D

View-point Center of interest Zoom Fly to

Pointing Query Relation

Translate Rotate Scale Define rela-tions

Line-of-sight Volumes Proximity

(3D)

World View (3D)

3D CAD level-of-detail

Realistic Textures Video

Walk through. Virtual guide

Pointing Scenarios Sound Sight Shadow

3.5 Summary

The chapter outlined the various visualizing techniques, which helped to identify the working visuali-zation technique for the research work. Level of detail was also introduced, as it is factor in displaying feasible the scenes in 3D visualization. Lastly, scene components was also mentioned to highlight the on the over all constituent elements required to give the final 3D scene.


22


23

Chapter Four Virtual Reality

4.0 Introduction

This chapter is introduced to highlight the presentational medium required to visualize the proposed system. The chapter first starts by introducing 3D maps and then gives a general overview and basic concepts in Virtual Reality (VR). As will be introduced in chapter 5 (section 5.2.1), the 3D virtual en-vironment was achieved by exporting the extruded 2½ D cube-like shapes of the cadastral parcel boundaries into VRML. The chapter then introduces Virtual Reality followed by definitions and char-acteristics of VR, VR and cartography and lastly Virtual Reality Modelling Language (VRML).

4.1 3D Maps

The final output, which is a 3D digital map to be visualised on the computer screen calls for discussion on 3D maps. Maps are no longer only final products they used to be, for instance the paper maps, which functioned and function as a medium for storage and presentation of spatial data (Kraak and Ormeling, 1996). On-screen maps and its corresponding database have changed these functionalities. Adding, the knowledge of database technology and computer graphic techniques has resulted in a new and alternative presentation options such as the 3D maps and animated maps. Definition of 3D maps can be coined from Köbben (2002) on “3D Perception: How do we ‘see’ 3D?” According to the au-thor, ‘a map is said to be three-dimensional when it contains stimuli which make the map user perceive its contents as three-dimensional’. The author further added: to see in 3D involves physiological (that is, involving human visual stimuli to see real depth; for example, focussing of the eyes and turning of the eye towards the object) and psychological depth cues (that is, involving graphical techniques to ‘fool the eye’; for example, shading and perspective). In this research, the visual stimuli are the ex-truded footprints of 2D maps into 3D geometry with images of the properties textured onto the 3D ge-ometry, which was then visualized in VRML to give a realistic virtual 3D view. Virtual reality with its immersive capabilities is elaborated in the next section.

4.2 Virtual Reality

Virtual reality has several terminologies such as virtual world and virtual environment but that to which the research adopts is virtual reality. Virtual reality could be said to be the result from interac-tion between the cognitive1 level of human beings, usually designated as mental maps, and the visual and audible images produced by computers (Jacobson, 1994). Adding, virtual reality is a space delib-erately designed by man, representing real or abstract spaces in which objects exists. Virtual reality is a computer environment designed by man to represent the actual or real world.

1Cognitive: The mental process of knowing, including aspects such as awareness, perception, reasoning, and judgment (source: Online Dictionary.com, 2002)


24

Burdea and Coiffet (1994) also added, virtual reality facilitate human-computer interaction by use of three-dimensional representation and direct manipulation of virtual objects. These representations can be seen as plane images, as 2½ D models in a conventional monitor or in a truly 3D space in an im-mersive environment. Many aspects exist when one talks about virtual reality, as whether it is being designed or constructed for a single user or multiple users and how the environment can be interacted by the user (see Figure 4.1). Adding to this section, calls for a definition of virtual reality, which is introduced in the next section

4.3 Definition and Description of Virtual Reality

Virtual Reality can be defined from different perspectives. For instance, considering the situation where the user is immersed in the computer-generated environment. Also virtual reality could be de-fined from the view of the technological tools being employed for example use of head mounted dis-play units and motion-tracking gloves. Virtual reality could also be defined from the psychological perspective where technology does not come into play rather the state produced in the users minds that could occupy their awareness in a way similar to that of real environments (Keppell and Macpherson,�1997). According to Fällman et al (1999) the best way to define virtual reality is to centre on the user and look at the style of interaction that takes place between the user and the computer generated-environment. The author further illustrated, users manipulate what is perceived to be real objects in the same manner, as they would manipulate in the real world. A concise and acceptable definition of virtual reality which this research fits into is that by Fisher and Unwin (2002), defined as ‘the ability of the user of a constructed view of a limited digitally-encoded information domain to change their view in three dimension causing update of the view presented to any viewer (the user)’. The definition gives a brief summary of the essence of all visualizations that can be called virtual reality. Thus, applying the definition, the user such as clients of real estate agents would have the capability of changing the viewing positions of the property on the computer. Online Dictionary.com (2002) defined virtual reality as ‘a computer simulation of a real or imaginary system that enables a user to perform operations on the simulated system and shows the effects in real time’. The computer simulation involves hypothetical three-dimensional visual world created by the com-puter. Virtual reality can be viewed from the perspective of human-computer interface (Brodlie et al, 2002). These authors described virtual reality as a tool or approach that can be used to undertake in-vestigations or present information in a new and challenging ways. The question now is, why the importance of virtual reality or why use virtual reality for this research? The reason is: ‘the use of visualization methods in the analysis of geo-referenced data is based on static models, which restricts the visual analysis capabilities. Thus, the use of virtual reality, which gives the user the ability to change viewpoints and models dynamically, can overcome such limita-tions’ (Diotin and Kooy, 1995). Fairbairn and Parsey (1996) also added, virtual reality as a technique enables users to move towards a finer emulation of the complexities of the real world. In addition, Baker and Wickens (1992) contributed that; virtual reality provides an interactive environment, which is well suited to supporting the interactive presentation of scientific datasets. Virtual reality has the potentiality in navigation and object manipulation as well as immersion in the dataset. That is, virtual reality facilitates egocentric views in which the display is constructed from the point of view of the user who may be inside the data space as compared to desktop tools, which generally restricts the user to outside looking (termed exocentric view). It is to these interactive presentational capabilities of vir-tual reality that this research adopted virtual reality as a presentational medium. In other words, virtual


25

reality brings the promise of a much more comprehensive way of interacting with the data. The user is within the middle of the plan and can directly interact with the 3D data (Verbree et al, 1999).

4.4 Characteristics of Virtual Reality

Base on the above chosen definition of virtual reality, virtual reality can be characterized by a number of features. These characteristics can be attributed to the nature of problem at hand, as in designing an effective virtual environment for analysing scientific data, demands a full understanding of the nature of the problem. Baker and Wickens (1992) characterized virtual reality according to a number of fea-tures, which contributes to the sense of reality. These are as illustrated below: • Virtual views are in three-dimensional perspective. That is, often in stereoscopic, unlike the planar

two-dimensional views commonly used in desktop environments. • Virtual point-of-view is egocentric, whereby the user is provided with a view of virtual space that

matches their physical position and direction of view within the space, allowing the immersion of the user in the virtual world.

• The virtual environment is dynamic rather than static. These characteristics of virtual reality can be used to classify virtual reality under desktop, semi-immersive and fully immersive virtual reality. These are discussed in the next section

4.5 Classification of Virtual Reality

This section discusses on the classification of virtual reality based on the immersion of the user and the S-P-I model (Simulation, Presentation and Interaction). Two possible methods of classifying are illus-trated in the sub-sections below.

4.5.1 Classification of Virtual Reality With Respect to Immersion of the User

With reference to Cronin (1999), there are basically three different kinds of virtual reality classified by the type of immersion that is being provided. These are desktop, semi-immersive and fully immersive virtual realities. Ogao (1997) also contributed that; these classifications can also be based on hardware and interface capabilities. These classifications of virtual reality can also be regarded as a continuum based on the levels of interactions and the real world used to facilitate transformation (Brodlie et al, 2002; See Figure 4.1). These classifications are as follows: • Desktop Virtual Reality (DVR), which is by far the most common and least expensive form of

virtual reality. It typically consists of a standard desktop computer. It is a form of virtual reality, which lacks any feelings of immersion on the part of the user.

• Semi-Immersive Virtual Reality, which attempts to give the user a feeling of being at least slightly immersed by the virtual environment. This is often achieved by workbenches and reach-in dis-plays. Ogao (1997) classified this as Transparent Virtual Reality, which uses the real world as backdrop and is seen through the device presenting the spatial information.

• Fully Immersive Virtual Reality, which consists of head mounted visual display units that allow users to be completely isolated from the physical world. Such an application operates in a spe-cially constructed environment or laboratory, which makes it more expensive and time-consuming to construct. Aside this setback, it fully filters out interference from outside world as well as allowing oneself to focus entirely on the virtual environment.


26

Figure 4. 1: Schematic illustration of VR regarded as a continuum based upon levels of interaction and the real world (Adapted from Brodlie et al, 2002).

Based on the above classifications, this research adopts that of DVR for the presentation of property information that can be used by real estate agents to help their clients. From Figure 4.1 (a) illustrates how visualization takes the advantage of interactive visualization techniques where the user can inter-act with the data to vary selection, classification, exaggeration and symbolisation for transforming the dataset. In this research, the user would interact with the data to vary selection in visualizing proper-ties. Such an illustration can be classified under DVR; (b) shows how the interface upon real world spatial information as well as taking the advantage of strong sense of immersion. This can also be clas-sified under Semi-Immersive Virtual Reality; and (c) shows how the user is fully and physically im-mersed in the model and responds as if operating in the real world. Such an illustration can be classi-fied under Fully Immersive Virtual Reality.

6.1.1 Classification of Virtual Reality in Computer Graphics using the S-P- I Model

Virtual Reality can be placed in the S-P-I model system of classification to determine its Simulation (S), Presentation (P) and Interaction (I) in computer graphics. In the S-P-I model, Virtual Reality has the highest requirements of simulation, presentation and interaction (see Figure 4.2). The S-P-I con-cept is purposefully designed for system classification in computer graphics (Encarnação et al, 1993). The Simulation ranges from geometry to higher level dynamic entities like buildings; the Presentation also ranges from single events to real time events; and Interaction ranges from non-interactions through interactive to immersive interactions.

Increase in level of Virtual Reality

Geographical Environment



Recognised Geographical Information



Representation Representation Representation

User

User

User

(a)(b)

(c)


27

Figure 4. 2: The S-P-I model for classification of Virtual Reality (VR) in computer graphics (Adapted from Göbel and Neugebauer, 1993).

In virtual reality, three essential components can be identified that is the simulation, presentation and interactions as indicated in the S-P-I model above. Simulation is the process whereby the behaviour of people and objects within the world are modelled; Presentation is the process of displaying the syn-thetic environment to the user either with immersive technology or simply on the desktop computer; and Interaction is simply the human interaction (Brodlie and El-Khalili, 2002). The authors explained; there is always a communication within these three components, which results in a continuous process. That is, Simulation results the state of which the environment is fed into the presentation component and the interaction from the user is returned from the presentation component to the simulation com-ponent. Aside the classification of virtual reality, a distinction could be made between virtual reality and cartography. This is discussed in the next section.

4.6 Virtual Reality and Cartography

Virtual Reality could be said to be a subset of cartography (Brodlie et al, 2002). Virtual reality capa-bilities could be used to increase the cartographic functionality in the computer environment (Berger et al., 1996). What makes virtual reality distinct or different from traditional cartography are the trans-formations in representing the world in each category (see Figure 4.3). This can be seen in the nature of the relationship between the representation/user and map/map image in virtual reality and tradi-tional cartography respectively as shown in figure 4.3.

Dynamic symantics Static semantics Geometry

None Interactive Immersive

Realtime events

Event sequences

Single events

Interaction

Presentation

Simulation

VR


28

Figure 4. 3: Schematic representation of Virtual Reality and Cartography (Source: Brodlie et. al, 2002).

From Figure 4.3, diagram a) depicts virtual reality identified as the creation of construct from reality when producing a virtual reality simulation. Diagram b) illustrates an extended form of diagram a), which shows the series of transformations involved in producing virtual reality. In this diagram, if rep-resentation and user are interchanged with map and map image respectively, the framework for virtual reality would correspond to traditional cartographic process as shown in diagram c), which illustrates a transformational view of cartography.

4.7 Virtual Reality Modelling language (VRML)

Based on the above discussions on virtual reality, it is interesting to mention that, virtual reality is not a method to imitate reality but rather to simulate aspects of reality as a sensual form of computation, which has called for major developments like VRML (Gillings and Goodrick, 1996). Desktop virtual reality, which offers view of a particular environment, uses VRML to create and manipulate these dis-play environments (Rhyne, 1999). The original ideas and concepts behind VRML were not towards a common format for virtual reality, but rather as an attempt to create three-dimensional interface to ab-stract information on the World Wide Web (Pesce et. al, 1994). The interesting question to ask is, what is VRML? VRML is a scene description language that describes the geometry and behaviour of a 3D scene or world. VRML is not a programming language; rather it is a scene description language (Cris-pen, 1998). To view a VRML world one needs a VRML browser, which is a program that reads a VRML file and displays the geometry, lighting, and animation as a 3D world. A VRML browser could be: a stand-alone application program, which can compile stand-alone applications that can view and manipulate VRML worlds. Examples are the Open World and the Open Inventor; a helper application, which has all the capabilities needed to view and manipulate VRML worlds. In this browser, whenever the web browser receives a link to a VRML world it will launch the helper application. An example is


Reality

Construct


Representation

User

Map


Map image


Census Ground Survey Remote Sensing Compilation

Selection Classification Simplification Exaggeration Symbolization

Reading Analysis Interpretation

Transform 1

Transform 2

Transform 3

a) VR (basic model)

b) VR (extended model)

c) A transformational view of cartography

Some Specific Purpose


29

the VRWave; a plugin, which uses the facilities provided by web browsers to display VRML worlds in web browsers. Examples are Microsoft VRML for Microsoft Internet Explorer, the Cosmo Player for Netscape Navigator and Communicator and the Parallel Graphic’s Cortona2 4.0 plug-in. A VRML world consists of one or more files conventionally with a ‘wrl’ suffix that together describe the geome-try and attributes of objects in 3D scene (Lovett et al, 2002). These files are used to define shapes, tex-tures, viewpoints and lights (Carey and Bell, 1997; Hartman and Wernecke, 1996). A choice of VRML was made for this research because; with VRML it is possible to explore the vir-tual environment in a flexible and an interactive manner. With VRML, it is possible to define specific viewpoints using the predefined tools on the control panel interface. Also, VRML has the capabilities of ease of viewing with a standard World Wide Web (WWW) browser, limited software costs and po-tential for dissemination through the WWW (Lovett et al, 2002).

4.8 Summary

In summary, coupling visualization discussed in the previous chapter and virtual reality discussed in this chapter contributed to the design of the proposed system. Processing the geographical data, as would be introduced in the next chapter, into a transferable file format (wrl format in VRML) and then passing the information to the virtual reality package for visualization contributed to the development of the system that would be used by real estate agents to help their clients in visualizing properties.

2Cortona: Cortona is a fast and highly interactive Web 3D viewer that is ideal for viewing 3D models on the Web. Cortona works as a VRML plug-in for Internet browsers such as the Internet Explorer (Source: Parallel Graphics, 2002)


30


31

Chapter Five Workable System and Prototype Implementation

5.0 Introduction

This chapter will illustrate the concept behind the dataset as well as the acquisition of the dataset. Also, highlights will be raised on the method used for the reconstruction of the 3D model from the 2D model. The chapter is introduced to tackle tasks 2, 3 and 4 raised under the research methodology. That is, referencing of the images in the computer, referencing camera station during photo exposures, generating the geometric model, linking photo images to the generated geometry, touching-up of pho-tos, developing the working system and lastly implementing the prototype. To achieve meaningful re-sults, the tasks were broken down into stages. That is, the conceptual, the testing and the implementa-tion stages. The conceptual and the testing stages involve the design and the development of workable system for the research. At the conceptual stage, literature review was made to obtain an idea about the approach to use in creating the 3D model. It involves also, selecting the kind of data acquisition ap-proach and softwares to use. The reviews made in chapters 2, 3, and 4 also contributed to the achieve-ment of the conceptual design for this research. The testing stage is the stage where the conceptual ideas were used on existing datasets as well as collecting height measurements, implementing the se-lected software’s and combining all these to get a working system. For the testing state, a small area of the Top 10 data of Enschede was used to develop the workable system. The implementation stage, involves the application of the working system developed at the testing stage on a larger dataset (2D cadastral map of Enschede). See Figure 5.1 for the inputs of conceptual, testing and implementation stages.

5.1 Conceptual Stage

The absence of a third dimension in commercial geographic information system (GIS) can be attrib-uted to the complexity of implementation (Pfund, 2002). This often makes visualization tools for users to be a very challenging issue, especially where 3D is needed. In 3D modelling, there is the need for adequate data acquisition methods for the 3D modelling construction. For data acquisition methods for 3D modelling construction, three modelling approaches could be distinguished (Pfund, 2002). These approaches are Photogrammetry, CAD and GIS. The photogrammetric technique, which is quite accu-rate, is able to generate the roof shape of buildings (see Paintsil, I997 and Tempfli, 1998 for more reading). CAD systems are normally used for refining existing 3D data (see Pfund, 2002 for more reading). The photogrammetric and CAD approaches are not within the scope of this research. The GIS approach follows a totally different approach. Starting with huge quantities of existing 2D data, 3D objects are computed using adaptive data structures. That is, a 2D GIS data is combined with a DTM, for the altitude, and linked to the information about the height of an object or more detailed data about 3D shape. The advantage of this method is that, the 3D model corresponds with its original 2D shape. That is, the base shape of the 3D model is the same as the 2D shape. The approach is conven-ient for simple objects that are stored in 2D. The approach, however, limits the possibilities for han-dling more uniquely shaped objects, for example roofs of buildings.


32

For either the photogrammetric approach or the GIS approach, there are two ways of creating the 3D model for building reconstruction. According to Zlatanova (2000), these two ways of creating a 3D model for building reconstruction are; the top-down and extrusions approaches. For the top-down ap-proach, the measured elements are the upper part of buildings (that is, roof outlines). Example is the photogrammetric approach. The top-down approach allows for more details about the roofs to be col-lected, but requires longer and complex processing of the data. The extrusions approaches starts from footprints of buildings. The extrusions approach could be applied on cadastral maps (Lammi, 1997). In this approach, the reconstruction is simple and allows fast implementation. Example is the GIS ap-proach. The application of any one of the above two approaches is dependent on: • Available data sources, that is 2D GIS and/or aerial images, • Hardware and software, This research adopts the GIS approach of modelling the 3D environment. That is, the designing of the 3D model for the virtual environment for this research is dependent on the GIS and extrusion ap-proaches. These approaches were chosen because; the main objective of the research was to integrate the 2D cadastral map with the 3D characteristics of buildings, where there was the need to use GIS approach of extruding the boundaries of buildings on the 2D cadastral map. For this research, the heights of buildings for the extrusion were determined from field measurements as described in section 5.2.2. The reasons for the choice of softwares and survey equipment for the 3D model for this research are illustrated below: Selecting Software: The demands to software led to the choice of certain softwares. Hence, different criteria were set for the different software applications. The first criterion was to select property boundary and its surroundings from the dataset. ArcMap software was used for this selection. The sec-ond criterion was to extrude the 2D footprints of the property boundary from the 2D map. The 3D Analyst extension in ArcScene was used to extrude the footprints of the 2D map into 2½ D cube-like shape. This converted the 2D to 2½ D and also exporting the 2½ D to 3D geometry in VRML with ‘wrl’ file format. The third criterion was to texture images onto the 3D geometry. The Parallel Graphic’s Internet Space Builder (ISB) software was used for the texturing. Lastly, the fourth criterion was to display the created model on the computer. The Cortona 4.0 plug-in for web browser (Internet Explorer) was chosen for the display of the 3D model. See Table 5.1 for the softwares used. Selecting Survey Equipment for Height Measurement: The criterion used here was a way of meas-uring height of buildings without much precision or accuracy as well as use of non-bulky instruments to study areas. Many sophisticated equipments exist such as the Theodolite and Total Stations, but the Suunto Altimeter PM-5 was chosen for determining the height of buildings. The Suunto Altimeter PM-5 is very handy and light in weight (see Appendix H). The other alternatives give a more precise measurement but that of bulkiness was a limiting factor and more also the aim was not to obtain a very precise measurement of the height of buildings for this research. All the ideas acquired at the conceptual stage were then used to develop working system at the testing stage. The testing stage is elaborated in the next section.


33

Figure 5. 1: The inputs for the design of the conceptual, testing and implementation stages.

Research Methodology

Verbal Interview (Real Estate Agents & Staff at ITC)

Letter sent for fixing Interview dates

LR on Virtual reality and why such a chosen method for the proposed system. LR on VRML

Conceptual Stage: Comprising literature review for the platform to use

LR on determining visualization techniques and choosing a type of visualization technique for the proposed sy stem.

Prepared questionnaire for Field Survey to determine requirements of real estate agents in visualizing property

Literature Review (LR) on Land Related Issues • Role of Real Estate

Agents • Land Related Issues • Cadastres and

Cadastral maps • Limitations o f 2D

Cadastres in the property market

• Why Need for visualizing property

• Requirements of real estate agents in visualizing property

Data acquisitioned methods

Testing & Implementation Stages Comprising observation & developing a working System

Taking mages with Digital

Camera Measuring Height o f Buildings with Suunto Altimeter & Height Calculations Field

Reconnaissance

Selecting Software & Survey Equip.

Usability Testing Using ‘think aloud’ method to test the prototype

Developing Prototype

Legend

Chapter Two

Chapter Four Chapter Three

Chapter Five Chapter Six

Chapter Seven

Conclusion and Recommendations Conclusion from Developed

System Conclusion

from Survey


34

5.2 Testing Stage

For the testing stage, the Top 10 Enschede ESRI Personal Geodatabase was used to achieve the work-ing system (see Appendix A1). This stage is further divided in sub-sections as illustrated below.

5.2.1 Preparing The Data

The data, which was in ESRI Personal Geodatabase format, was first converted to shape file using the ArcMap software to generate the lines and polygon features. The ArcMap software was also used to select the building, road network and land use areas to work on. The 3D Analyst extension was loaded in ArcMap for generating the 2½ D cube-like shapes of the buildings. The generated 2½ D cube-like shape was then exported to 3D geometry, a VRML file format in ArcScene. The image content was edited using the Microsoft Photo Editor. The ISB software was then used to photo texture the images onto the 3D geometry. These processes are illustrated in Figure 5.2, designed using Figure 3.1 (chap-ter 3). The final product was viewed in the Cortona plug-in for web browser (the Internet Explorer). Field observations discussed in the next section, also contributed in accomplishing the goal of data preparation.

5.2.2 Field Observations and Measurements

Before the field observations, the precautions undertook are as illustrated below: • Before measurements were made, it was ensured that the buildings to be visited could be identified

on the base map. Hence, field reconnaissance was conducted to make sure that the buildings to be photographed and measured were truly on the map.

• During the photograph taking, it was ensured that the direction of the sun was considered since it affects the quality of the images. This could cause blurriness of the images. To overcome this, pic-tures were taken in the mornings between 9am and 10 am.

The field observations are outlined under the following: Photographs: The photographs were taken using Sony Digital Mavica Quick Access FD Drive 2X camera. In achieving the appropriate camera station for good images the following procedures was used: • Without using tripod the camera was held firmly upright to avoid tilt of the image. • The camera position was directly in front of the building to acquire the view of all the sides of the

building. This procedure helped with the zooming in capabilities of the camera when it approaches its full limit of zooming.

The Sony Digital Mavica Quick Access FD Drive 2X camera automatically created names for the im-ages, for example MVC-403F.JPG. This helped to identify the images. The names were changed to suit the respective sides of the building for example HogeLS, HogeRS, HogeFS and HogeBS to mean left, right, front and back sides of Hoge building. This technique of re-naming helped to achieve the identification of the images in the computer for further usage. Images edited with Microsoft Photo Edi-tor software for developing the system shown in Appendix E7. Base Measurements: Due motor vehicles as well as no assistant during field measurements ‘pacing’ was used for the base measurements instead of stretching tapes from base of building to the position where angular measurements was taken. This base measurement was used in calculating the height of the building as illustrated under derived measurements.


35

Angular Measurements: In taking the angular measurements with Suunto Altimeter PM-5 from the location where the top and bottom of the building could be seen, the following were observed (The observations is as shown in Appendix D1): • The top of the building was sighted and the readings were read off from both the angular and the

percentage scale. • The bottom of the building was then sighted and the readings were read off from both the angular

and the percentage scale.

Figure 5. 2: The main stages in the creation of the 3D environment for testing stage.

Raw ESRI GEODATABASE

Selecting 2D working area

Selected Working Area

2½ D (Scene file)

Textured 3 D (VRML file)

ArcGIS (ArcScene)

Generating 2½ D model (Extrusion of 2D)

Internet Space Builder (ISB)

Texturing

Cortona + Internet Explorer

Displaying image in browser

3D virtual environment

Software’s Used Processes Output Stages

ESRI Geodatabase To shapefile

ArcGIS

(ArcMap)

Initial data preparation

3 D (VRML file)

Exporting 2½ D to 3D


36

Table 5. 1: Outline of the software's used at the testing stage

Software Purpose ArcGIS 8.2 ArcMap For selecting working area and displaying the 2D data. For converting ESRI geodatabase to Shapefile (shp). For exporting ArcMap files (mxd) to Shapefile (shp). For editing the lines and polygon features. ArcScene For extruding footprints of building polygons to obtain 2½ D. For exporting ArcScene files (sxd) to VRML file. Internet Space Builder (ISB) For Photo Texturing.

Microsoft Photo Editor For Editing Image content as well as cropping of images. Cortona 4.0 For displaying the prototype.

Derived Measurements: The derived measurements are further outlined under the headings: base cal-culations and height calculations. These are illustrated below.

a. Base Calculations An average pace measurement between my heels were obtained as 0.7 m. Hence subsequent pace measurements in Meters (m), for example as P Step say, can be obtained by using the proportion be-low: If One Step = 0.7 m, therefore P Step = P * 0.7 m Using this proportion the base length from base of building to the position where angular measure-ments is computed as shown in Appendix D2 and D3.

b. Height Calculations

Figure 5. 3: Demonstration of computing height of building using the Suunto Altimeter PM-5.

From Figure 5.3, the formula for calculating the height of building for the Suunto Altimeter PM-5 is derived as follows:

Height of Building = BD = BC + CD Known Distance = Base length = AD = EC and aº and bº are angles of inclination and depression respectively read on the Suunto Altimeter

Using basic trigonometry: tan aº = BC/AC, implies BC = EC * tan aº

Known distance

Building

Observer

aº

bº

D A

C

B

E


37

tan bº = CD/AC, implies CD = EC * tan bº Therefore, BD = BC + CD = EC * tan aº + EC * tan bº But EC = AD Hence,

BD = EC * (tan aº + tan bº) Therefore, Height of Building = AD * (tan aº + tan bº)

Using this formula the height of a building is computed as shown in Appendix D2 and D3. The height values are then used in the ArcScene environment to extrude the buildings. It must be noted that, to achieve better results for the base and angular measurements, redundant observations were taken for a particular set of readings as well as on different buildings (see Appendix D1).

5.2.3 Problems Encountered in the Testing Stage

This section elaborates on problems encountered during the testing stage. • It was very easy to get lost in the virtual environment of the ISB software during photo texturing. • It was also observed that although the objectives were achieved correctly, the photos did not stick

to the 3D geometry. There were small space or gaps between the textured photo and the geometric model. The cause could be attributed to the capabilities of the software. A solution could be the use of ‘gif’ image file formats instead of ‘jpg’ file format (that is, by loading ‘gif’ images into the ISB texture gallery instead of the ISB photo gallery). There is a trade off to this solution, as the problem with the gabs will be solved but the image quality will be distorted in the photo texturing. Another solution will be the application of a different software, which I have highlighted under the recommendations

• Obtaining sides images of some buildings was very difficult, since the buildings were close to each other. Also in some cases there were obstructions such as tree hedges preventing access to side of buildings. As a result of this I had to position myself at a different position with respect to the building. From Figure 5.4, ‘a’ and ‘c’ are the best positions for acquiring the images of the sides of building ‘A’. However, due to the obstruction imposed by building ‘B’, the image position ‘a’ was shifted to ‘b’. This technique helped in tackling the referencing of camera stations on the field.

Figure 5. 4: Illustration on how obstructions were avoided.

•

•

Building A a Building B

b

• c


38

5.2.4 Results of the Testing Stage

The final results obtained at the testing stage are as illustrated in Appendix B where that with pre-defined buildings of the ISB software is included to portray aesthetic view of the virtual environment. With the ISB software, photo texturing was made on the geometric model with the acquired photos taken in the field. In addition, geo-referencing of the images was done manually by texturing them onto the geometric model in the ISB software. Re-naming the ‘jpeg’ files was done to identify the im-ages in the ISB software. The camera stations were achieved by positioning the held camera at ap-proximately the centre of the side of the building to be imaged as shown in Figure 5.4. A workable system was achieved at the testing stage by adopting the concepts at the conceptual stage. This worka-ble system was then used at the implementation stage described in the next section.

5.3 Implementation Stage

For the implementation stage, the 2D cadastral map of Enschede was used. For this stage all the meth-ods and procedures used at the testing stages were adopted (see Figure 5.5). The problem encountered at this stage was that, the 2D Cadastral map of Enschede is of poly-line features as a result polygon features had to be created by digitizing. The developed prototype was then used in the usability testing to test for its efficiency, effectiveness and satisfaction.

Figure 5. 5: Results obtained at the Implementation stage


39

A rap up on how the prototype was developed is as follows: • The 2D cadastral map, which is in poly-line features had to be first, converted to polygon features

(consisting of buildings and landuse areas) by digitising. These polygons were then used to gener-ate the buildings in the ArcScene software.

• The ArcScene software has the capability of extruding the 2D polygon features of the boundaries of the buildings into 2½D model, using ‘extrusions’ in the ‘layer properties function’. The Arc-Scene software has also the capability of converting the 2½D model to 3D model (a ‘wrl’ format), using the ‘export scene function’.

• With the ‘wrl’ format, the 3D model can then be used in the ISB software. In the ISB software, all the images (‘jpg’ files) were loaded into its photo gallery. These images were then used for the photo texturing. Here, well-labelled (or reference) images are required for proper identification of the images (see section 5.2.2). The main platform for this prototype is the ISB software. After the photo texturing is done and the application ready, the file containing the application had to be pub-lished, using ‘publish function’ in the ISB software. This is a very important, since it enables the application to be viewed in any browser.

• The final environment for visualizing the application was the Cortona plug-in for web browser (Internet Explorer).

5.4 Summary

Summarizing, the conceptual stage helped to develop a framework consisting of tools to use such as the GIS modelling techniques, the software, and survey equipment. The survey interview conducted in chapter two also contributed to the conceptual stage as a form of pre-text questionnaire for determin-ing the requirements of real estate agent requirements as well as the necessary input for developing the intended system. The testing stage is the stage where all the concepts adopted in the conceptual stage were put in operation on a small dataset to achieve a working system for developing the prototype at the implementation stage.


40


41

Chapter Six Usability Testing

6.0 Introduction

As indicated in chapter one (section 1.4.2), one of the objectives of the research was to develop a pro-totype. But for this prototype to be used by the target users, one needs to test for its efficiency, effec-tiveness and satisfaction. As a result, it was considered to have the prototype evaluated by few test par-ticipants in a usability test. This chapter explains how the usability testing for the prototype was car-ried out. The chapter starts with discussions on the usability testing, which comprise of: the method used (that is, the ‘think aloud’ method); design and implementation of the test according to the method; and analysis of the results and conclusion from the tasks and questionnaire. This is then fol-lowed by recommendations from test participants in the ‘think aloud’ method.

6.1 The Method Used

The method adopted for this research for the usability testing is the ‘think aloud’ method. According to Rubin (1994), the ‘think aloud’ method for usability testing is a simple technique intended to capture what the participants (users) are thinking while working. The author added, to implement this tech-nique, one has to let the participant provide a running commentary of their thought process by thinking aloud while performing the tasks of the test. That is, allowing test participants to express their confu-sion, frustrations and even their delight. The author further added, when done well the technique as-sists in ‘reading users minds’. In this research, there is the need to know from users what the map dis-play looks like, whether the map can be used for the tasks given, whether it meets their expectation, and whether the information supplied by the map is what they need. The mixture of performances and preference into that needs to be acquired in the usability test leads to the choices for the ‘think aloud‘ method, because this is an excellent way to unravel cognitive processes taking place in the mind of the users. Further reasons for this choice is as explained in Table 6.1.

6.2 Design of the ‘Think Aloud’ Test

This section illustrates how the ‘think aloud’ test was designed. The design was based on the objective of the usability testing (that is, evaluating the efficiency, effectiveness and satisfaction of the proto-type). Based on this objective, the testing was broken down into tasks and questionnaire. The tasks are comprised of trial and testing sessions. The trial session involves a simple application of the 3D vir-tual environment developed at the testing stage (chapter 5, section 5.2). This is introduced to familiar-ise participants with the user interface before testing the prototype. This was done because; the objec-tive of the usability testing was not to test the software interface, but rather the concepts behind the design of the prototype. The testing session is comprised of pre-defined tasks for the test participants to perform whilst ‘think aloud’, followed by the questionnaire (see Appendix I). The questionnaire is comprised of totally structured questions, which were answered by test participants by ticking their choice of answer on the paper whilst ‘think aloud’, to explain their choice of answer. The structure of the questionnaire is as follows: • Satisfaction, which is the complete fulfilment of a need or want. These questions were used to

determine the test participants’ satisfaction after testing the prototype. For instance, whether the


42

prototype gives enough information about the property one is interested to buy; whether the proto-type appeals to users in buying a property; and how users feel about the application.

• Effectiveness, which is the extent to which goals are achieved. This question was used to find out whether the prototype is easy to use.

• Efficiency, which is the mental effort put into reaching goals. These questions were used to find out if the prototype involves time in visualizing property.

The way the ‘think aloud’ method was implemented is described in the next sub-section.

6.3 Implementation of the ‘Think Aloud’ Test

Implementing the usability testing, twenty-two formal letters on ITC template were sent to real estate agents asking for their participatory interest. Also, e-mail letters were sent to all ITC staff for their par-ticipatory interest. Real estate agents were involved here to determine their interest in the prototype, as the prototype is being designed for them. Also ITC staffs were assumed here to be buyers (that is, cli-ents of real estate agents) of properties, since most of them are in the position of buying a house, or one way or the other, have acquired a house. In all, ten test participants participated, comprising nine ITC staff’s and one real estate agent. The test was conducted in the ITC Cartographic Laboratory (see Figure 6.1 for diagrammatic descrip-tion of the set-up using the ‘think aloud’ method), where there were no disturbances from the outside environment. The test participants were first all introduced to objective of the test and the trial session. This was then followed by explanations on: how to perform the tasks and how to answer the question-naire. Test participants were given a brief and concise demonstration of how the interface works. Later, participants were left alone to perform the test whilst they are being timed. For the questionnaire session, a sample of the traditional way of using pictures by real estate agents to inform clients about property was also made available. This was to allow for comparison to be made between the prototype and the traditional way of visualizing property. Five minutes was allotted for the trial session, for fa-miliarization. Ten minutes for the testing session for which the seven tasks (see Appendix I2) were given to the test participants to perform whilst ‘think aloud’. After, another ten minutes was allotted for answering the questionnaire whilst ‘think aloud’, to determine the satisfaction, efficiency and ef-fectiveness of the prototype’s design.

Figure 6. 1: Mechanism of acquiring user feedback using the ‘think aloud’ method – Set-up in ITC‘s Divi-sion of Geo-Informatics, Cartography and Visualization. (Adapted from: Cusi-Redido, 2002).

Test Participant

Video camera

recorder

Computer monitor

Video quad unit TV

Video cassette recorder

External Wireless Microphone


43

Table 6. 1:Advantages and disadvantages of ‘think aloud’ method (Source: Rubin, 1994 and van Elzakker, 1999).

Advantages of the ‘think aloud’ method Disadvantages of the ‘think aloud’ method

• With the ‘think aloud’ method, participants are able to capture preference and perform-ance information simultaneously, rather than having to remember to ask questions about preferences later.

• With the ‘think aloud’ method, there is no problem of memorising the thought that came up as thoughts are spoken up immediately.

• One receives early clues about misconception and confusion of participants before it mani-fests incorrect behaviour. These early clues help one to anticipate and trace the source of problems easily.

• The method can help participants to focus and concentrate. Participants fall into the rhythm of working and speaking throughout the test.

• ‘Think aloud’ method leads to valid and most complete data on cognitive processes

• With the ‘think aloud’ method, participants would find the technique unnatural and dis-tracting since the thinking aloud is very dif-ferent from their own style of learning. If the participant is not an analytical learner, he or she may severely feel inhibited.

• ‘Think aloud’ slows the taught process, thus increasing mindfulness.

• ‘Think aloud’ is a very time-consuming tech-nique. It is not only the data collection that takes time but particularly the coding and analysis of the verbal protocols. The analysis of the resulting protocols is often difficult and tedious.

• Using the ‘think aloud’ method participants may find it difficult to translate their thoughts into words.

6.4 Analysis and Conclusion from Pre-defined Tasks and Questionnaire

For the tasks, test participants were able to go through all the seven tasks by using the study, plan, pan, turn, roll, restore, align and the fit buttons provided by the Cortona plug-in for web browser (the Inter-net Explorer). Test participants were able to identify properties within the application, even the type of property (that is, properties being built for the old age). In due of this identification, some test partici-pants say the property is not to their liking if they are to buy. The test participants identified the front, the left, the right and the back views of the properties, although there were difficulties in using the but-tons to navigate through the virtual environment. Some participants even claim some sides show no image. Others also claim the speed of the interface was too much, although they could adjust the speed of movement. Adjacent properties were identified at ease. Also, entrances to properties were identi-fied. The road linking the property, parking lots were also identified. Viewing the whole environment in which the properties were located was very easy to be accomplished by the test participant. With the difficulties in using the application, some participants suggested that; the application should in the first place be used by real estate agents to demonstrate properties to their clients rather than allowing par-ticipants or clients to operate it themselves.


44

For the questionnaire, the analysis was based on highest number of participants’ choices made (see Table 6.2). These numbers were further classified as percentages (see Figure 6.2). Based on Figure 6.2, the results obtained from the questionnaire could be analysed as follows: • For Satisfaction: 50 % of test participants partially agree that the look of property using the proto-

type is appealing and very satisfactory; 50 % of test participants totally agree, that the prototype is worth of use for visualizing property; 40 % partially agree, that the application is a good medium of informing property offered for sale; 50% partially agree, that the application gives a realistic view of the environment as compare to traditional way of showing pictures. However, 50 % nei-ther agreed nor disagreed, that the properties represented in the prototype is more appealing as compared with the traditional way of visualizing property. Explaining these higher percentages (see Figure 6.2a) by the protocols from the test participants in the ‘think aloud’ method, the tech-niques behind the application is very pleasing for visualizing properties and their surroundings al-though it does not inform them about the cost of the property. The traditional method contains much textual information about the cost. Adding, there is too much simulation (artificial objects such as tress). Also, the prototype gives a visual impression about the property and its environment one is interested to buy. An example given by a test participant was the purchase of a house with-out even having the chance of seeing the environment in which the property is located. A conclu-sion drawn finally was; the application is satisfactory in terms of use, realistic representation of the environment and informing one about a property offered for sale. But not satisfactory in informing one about the cost of the property.

• For effectiveness: 40 % of the test participants partially disagree to the question that, navigating into the systems for visualizing property and its surroundings is easy. This higher percentage (see Figure 6.2b) can be explained by the protocols from the test participants in the ‘think aloud’ method. That is, most of the test participants are not conversant with virtual reality interface due to the way they interact with the application. Some participants were lost in the virtual environment. Hence, navigating into the system seems very difficult. Some claim with consistent practice it will be very easy to use. Adding, the buttons have too much functionality for use. Concluding, the pro-totype is efficient for test participants who are conversant with virtual interface. But not efficient for test participants who are not conversant with virtual interface.

• For efficiency: 60 % of the participants partially agree that, the application does not involve much time in visualizing property; and 70% of the participants partially agree that, the mental effort in identifying and visualizing properties using the prototype is efficient in terms of time. Explaining these higher percentages (see Figure 6.2c) by the protocols from the test participants in the ‘think aloud’ method, properties were easily identified without difficulty with even more elaborate de-scription of the type of property test participants are visualizing. A conclusion drawn finally was; the application is efficient in terms of the time required to identify property and its surroundings, although the virtual navigation was problematic for test participants.

In all, participants claim the application is good for representing the environment. That is, can have a look in the surrounding as compared to the traditional approach. The application and traditional ap-proach is complementary (that is, both could be used to supplement each other). However, participants were able to grasp the concept of identifying property and it’s surrounding within the application. Real estate agent do think that the prototype is not necessary. That is, the prototype would best use for visu-alizing commercial (bigger) properties rather than domestic properties.


45

6.5 Recommendations from the Test Participants in the ‘think aloud’ method

The following is a summary of improvements recommended for the prototype by the test participants during testing and questionnaire sessions: 1. The applications’ functionality should be reduced. This is because; a button (for example, turn) of

the interface has too much functionality. Participants claim this problem makes the application dif-ficult to use. Well descriptive buttons such as pan and zoom should be used. Need for ‘go into’ the property functionality.

2. The need for click-able functions to display textual information (that is, an access to attribute data).

3. Use of joysticks in replace of the mouse to allow for effective use and navigation within the appli-cation.

4. A map of the area showing where the property is located should also be made available. Also, linking the application to cadastral map such that, a click on the parcel boundary on display will activate the application to view the environment in which the property could be identified.

5. The real estate agent claims the clients need information about the cost of the property. Thus, there is a need to provide a mechanism such that a click on the property would display cost information.

6. An interesting area posed by the real estate agent is that: the application could be developed for newly developed area and large commercial properties not only domestic properties.

For item 1, it is a problem with the browser one employs and also the familiarity one has in using desktop virtual reality. The items raised by the test participants were not implemented because of time constraints.

Table 6. 2: Number of participants’ answers for the questionnaire.

1 2 3 4 5

Totally Agree

Partially Agree

Neither Agree nor Disagree

Partially Disagree

Totally Disagree

On Satisfaction

1. The look of property identified is appealing. 1 5 3 1 nothing2. The way in which the property is currently represented in the system is very satisfactory (i.e. visualization of property is realistic and detailed enough). 2 5 2 1 nothing3. The way property is presented in this application is worth of use for visualizing property you wish to be informed about. 5 3 1 1 nothing

4.The application is a good medium of informing you about a property offered for sale 1 4 3 2 nothing

5. Comparing the two systems of visualizing properties, this application looks more appealing. 2 2 5 1 nothing6. Comparing the two systems of visualizing properties, this application gives realistic view of the environment. 2 5 nothing 1 2

On Effectiveness1. Navigating into the system for visualizing property and its surroundings using this application is easy to use (i.e. the application does not involve effort). nothing 3 2 4 1

On Efficiency1. Visualizing property and its surroundings is skilful in terms of time (i.e.the application does not involve much time in visualizing property). 3 6 nothing 1 nothing2. The mental effort in identifying and visualizing properties using this application is efficient in terms of time. 1 7 1 1 nothing

Questions


46

Question One

Totally Agree10%

Partially Agree50%


30%

Partially Disagree10%

Totally Disagree0%

Question Two

Totally Agree20%

Partially Agree50%


20%


Totally Disagree0%

Question Three

Totally Agree50%

Partially Agree30%


10%

Partially Disagree10% Totally Disagree

0%

Question Four

Totally Agree10%

Partially Agree40%Neither Agree nor

Disagree30%


Totally Disagree0%

Question Five

Totally Agree20%

Partially Agree20%Neither Agree nor

Disagree50%


Totally Disagree0%

Question Six

Totally Agree20%

Partially Agree50%


0%


Totally Disagree20%

a) Answers on Satisfaction


47

Figure 6. 2: Number of participants’ answers for the questionnaire expressed as percentage and visualiz-ing the results using the Pie Chart – a) Answers on Satisfaction, b) Answers on Effectiveness and c) An-swers on Efficiency.

Question One

Partially Agree30%


20%


Totally Disagree10%

Question Two

Totally Agree10%

Partially Agree70%


10%

Partially Disagree

10%

Totally Disagree0%

Question One

Totally Agree30%

Partially Agree60%


0%

Partially Disagree10% Totally Disagree

0%

c) Answers on Efficiency

b) Answers on Effectiveness


48

6.6 Summary

The implementation of the prototype at the implementation stage using the 2D cadastral map of En-schede proved successful. The ‘think aloud’ method also proved to be a useful usability-testing tool to improve the design of the prototype as well as determining the satisfaction, efficiency and effective-ness of the prototype. Feedback from test participants using ‘think aloud’ proved helpful, for further development of the prototype. The ‘think aloud’ method although tedious to implement, proved suc-cessful in determining the cognitive reasoning of test participants about the prototype.


49

Chapter Seven Conclusion and Recommendation

7.0 Introduction

In this chapter, concluding remarks of all the previous chapters would be summarised and recommen-dations for further work would also be presented. Under the section conclusion, a wrap-up of all the ideas that constituted to the research would be discussed. The chapter will contain a reflection on the objectives as well as what has been achieved or not. The chapter will also contain interpretation of the final results (that is, most striking results, what went wrong and why, and what would be done differ-ently). All these led to the suggestions for further improvement of the prototype under the section of the recommendation.

7.1 Conclusion

The main objective of this research was the integration of 2D cadastral map and 3D characteristics of buildings for the real estate agents to help their clients in visualizing properties they have for sale. This was tackled by finding real estate agents’ requirements in visualizing properties. These requirements were discovered by literature review and the formulation of questionnaires for verbal interviews (as in chapter two, section 2.4). The verbal interview also served as input to the development of the proto-type. The verbal interview conducted shows that there is opportunity to raise the idea of visualization to real estate agents. That is to say, there should be some awareness of techniques of visualization that can be used by real estate agents to present the properties they have for sale to their clients. The limita-tion of 2D maps in visualizing land property for the property market was found by literature reviews (as in chapter two, section 2.1.2). It was discovered that, 2D cadastral systems do not reflect the verti-cal dimensions of buildings. Also, paper maps do not possess the possibility of interacting with objects on display. Literature review was conducted to determine: the visualization techniques, visualization tool and modelling techniques for developing the prototype. The visualization techniques adopted were; geometric modelling, map-based, object-based and image-based techniques. The visualization tool adopted is the virtual reality. GIS modelling technique comprising the extrusion method was adopted for this research. It was discovered that, there was a limitation to this modelling approach. This is because; it limits the possibilities for handling more uniquely shaped objects, for example roofs of buildings. Flat surfaced roofed buildings were considered in this research. As indicated in chapter five, section 5.1, review shows that the Photogrammetric modelling technique could handle the more uniquely roof shape of buildings. All the reviews constituted to the conceptual framework for this re-search. The conceptual framework was then used to develop a testing framework (that is, workable system) at the testing stage. The concepts used for the testing stage was then used at the implementa-tion stage for developing the prototype. Integrating 2D cadastral map and the 3D characteristics of the building was found by: using different softwares for converting the 2D map (model) to the 3D model; collecting heights of building using the Suunto Altimeter; and photo texturing in the ISB software. Geo-referencing of photographic images taken on the field were done manually by texturing them onto the geometric model in the ISB soft-ware. The integration and geo-referencing resulted in the 3D realistic view of building. As indicated in chapter five section 5.2.3, it was discovered that there were small gaps between the textured photo and


50

the geometric model in the ISB software during the photo texturing. The cause could be attributed to the capabilities of the software. A solution could be the use of ‘gif’ image file formats instead of ‘jpg’ file format (that is, by loading ‘gif’ images into the ISB texture gallery instead of the ISB photo gal-lery). There is a trade off to this solution, as the problem with the gaps will be solved but the image quality will be distorted in the photo texturing. Another solution will be the application of different software, which I have highlighted under the recommendations. The conversion from 2D map (model) to 3D model shifted from 2D (plan view) to 2½D (model view) to 3D (model view). That is, moving from to 2D to 3D requires an intermediary 2½D model. Heights of building were measured using the Suunto Altimeter because; the aim of the research was not to obtain very precise measurements of the height of buildings. Other alternatives such as Theodolite and Total Stations could be sorted if more precise measurement is required. It is worth mentioning here that, the main objective of the research of integrating the 2D cadastral map and the 3D characteristics of buildings for visualization was achieved by tackling the mentioned research questions (as indicated in chapter one, section 1.5). The Top 10 and 2D Cadastral map of Enschede were main dataset used for this research. The final step of this research was to obtain feedback on the efficiency, effectiveness and satisfaction of the designed prototype from test participants who participated in the usability testing. Ten tests participants con-ducted the testing of the prototype using the ‘think aloud’ method. These test participants comprising a real estate agent and ITC staffs were assumed to be buyers of properties. This was done because get-ting real buyers was a difficult task and more also it was not within the scope of this research. From the protocols of the ‘think aloud’ method, it was discovered that the real estate agent do think the pro-totype is not necessary. A suggestion was that the prototype would be best use for visualizing com-mercial (bigger) properties rather than domestic properties. This shows that, user (client of real estate agents) demands in visualizing property is not considered much by real estate agents rather the sale of properties by using ‘feelings’ of the clients. A conclusion drawn from the usability testing for the pro-totype is that the application is satisfactory in terms of use, realistic representation of the environment as well as informing one about a property offered for sale; the application is efficient in terms of the time required to identify property and its surroundings; and the application is effective for test partici-pants who are conversant with virtual interface. Although a method was found for the 3D virtual environment, this research did not attempt to compare with other methods for obtaining the best available method for 3D virtual environment (as mentioned in the research objectives). This was due to the time constraints. The concept of level of detail makes highly complex scenes to be displayed feasible but this was not introduced in this research since a small working area was considered in developing the prototype. Indicated in chapter one section 1.1, it was discovered that the need arises for 3D representation of the environment to which stakeholders such as the real estate agent could interact with. In this respect the real estate agent as well as large private and public construction works can use the map as an index to present all the properties of apartments and offices required for sale and maintenance. The interpretation of the results mentioned above led to the suggestions for further improvement of the prototype. This is introduced in the next section.


51

7.2 Recommendation

In terms of visualization, cartographers should explore the venues for visualizing properties for the real estate agents since the real estate agents have vague idea about visualization. More also, real estate agents (apart from the traditional way of using pictures for visualizing properties) do not see the im-portance of other means of visualizing properties to help their clients. Real estate agents’ market ori-ented ideas could be changed if a stress is made on the importance of visualization. Demands of the clients of real estate agent in visualizing properties and their surroundings rather than use of text de-scriptions could be met by such a technique. From the protocols of the usability testing, an example given by a test participant was the purchase of a house without even having the chance of seeing the environment in which the property is located. It is also recommended that, all the items mentioned in the summary of improvements recommended for the prototype by the test participants (see section 6.5) that were not implemented by the researcher should be implemented to enhance the efficiency and ef-fectiveness of the prototype. From the conclusion (section 7.1) of this research, the other recommenda-tions raised by the researcher are: • Testing of the prototype with ‘real’ users or buyers (clients of real estate agents), not real estate

agents and staffs of ITC. • The use map as an index to present all the properties of apartments and offices required for sale

and maintenance. • Level of detail is worth to be introduced, if a large urban area is to be visualized in 3D. • Investigation of a software that could do the photo texturing without gaps between models and

textured images. • An attempt to obtain the best available method for 3D virtual environment by making comparison

with other methods • Use of Photogrammetric modelling technique in handling the more uniquely roof shape of build-

ings.


52

BIBLIOGRAPHY

BIBLIOGRAPHY-1

BIBLIOGRAPHY

Andrienko, N., Andrienko, V. H., Bernardo, F., Hipolito, J. and Kretchmer, U. (2002). “Testing the Usability of Interactive Maps in CommonGIS”. In Cartography and Geographic Information Sci-ence, Vol. 29, No.4, pp 325-342. Baker, M. P. and Wickens, C. D. (1992). “ Human Factors in Virtual Environments for the Visual Analysis of Scientific Data”. http://archive.ncsa.uiuc.edu/Vis/Publications/humanFactors.html (Date accessed 22/10/2002, last modified 08/08/1998). Berger, P., Meysembourg, P., Sales, J. and Johnston, C. (1996). “Toward a virtual reality interface for landscape visualization”. Proceedings of the Third International Conference/Workshop on Integrat-ing GIS and Environmental Modelling, January 21-25, 1996, Santa Fe, New Mexico. Bhuhu, S. T, Ruther,H. and Gain, J. (2002).“Three-Dimensional Virtual Reality in Urban Manage-ment”. Developments and Technology Transfer in Geomatics for Envirommental and Resource Man-agement, ISPRS Volume XXXIV, Proceedings of the Joint ISPRS Commission VI Workshop, 25th to 28th March 2002, UCLAS, Dar es Salaam, Tanzania. Editors; Baltsavias, E.P., Bouloucos, T., Matalo, G.E., Niederost, J. and Rivett, U. pp. 201-204. Electronic Archive: http://www.photogrammetry.ethz.ch/general/persons/jana/daressalaam/papers/bhunu.pdf (Date ac-cessed 13/09/2002, last modified 13/09/2002) Bishop, I. D., and Karadaglis, C. (1997). “Linking modelling and visualization for natural resources management”, Environment and Planning B: Planning and Design, 24,pp345-58. Bourdakis, V. (1997). “Making Sense of the City”. In CAADFutures97, August 4-6. Electronic Ar-chive: http://fos.prd.uth.gr/vas/papers/CAADFutures97/ (Date accessed 26/07/2002, last modified 11/03/1998). Brack, C. (2001). “Suunto Clinometer”. Electronic Archive: http://sres.anu.edu.au/associated/mensuration/suunto.htm (Date accessed 11/10/2002, last modified 14/12/2001). Brodlie, K. (1994). “A typology for Scientific Visualization”. In Hearnshaw, H.M. and Unwin, D. J. (eds.). Visualization in Geographical Information Systems, John Wiley and Sons, Chichester, pp.34-41 Brodlie, K., Dykes, J., Gillings, M., Hakley, M. E., Kitchin, R. and Kraak, M. J. (2002) “Geogra-phy in VR: Context”. In Fisher, P. and Unwin, D. (eds). Virtual Reality in Geography. Taylor and Francis Publishers, London. pp.7-16 Brodlie, K. and El-Khalili (2002). “ Web-based virtual environments”. In Fisher, P. and Unwin, D. (eds). Virtual Reality in Geography, Taylor and Francis Publishers, London. pp.35-46. Burdea, G. and Coiffet, P. (1994). “Virtual Reality Technology”. New York: John Wiley. Bureau of Labour Statistics (2002). “Real Estate Brokers and Sales Agents”. In U.S. Department of Labour, Occupational Outlook Handbook, 2002-03 Edition. Electronic Archive: http://www.bls.gov/oco/ocos120.htm. (Date accessed 15/01/2003, last modified 15/01/2003) California Occupational Guides (2002). “Real Estate Agents” http://www.calmis.cahwnet.gov/htmlfile/subject/guide.htm (Date accessed 07/10/2002, last modified 17/09/2002). Carey, R. and Bell, G. (1997). “The Annotated VRML 2.0 Reference Manual”. New York: Addison-Wiley.

BIBLIOGRAPHY

BIBLIOGRAPHY-3

Cartwright, W., Gartner, G. and Riedl, A. (2001). “GeoMultimedia and Multimedia Cartography”. In Proceedings of CORP 2001, Wien, Austria: Technischen Universität. pp245-251. Electronic Ar-chive: http://212.17.83.251/corp/archiv/papers/2001/CORP2001_Cartwright_DO.pdf (Date accessed 19/11/2002, last modified 19/11/2002). Chen, C. (1999). “Information Visualisation and Virtual Environments”. Springer-Verlag London Limited.pp 175-211 Clark J.H. (1976). “ Hierarchical Geometric Models for Visible Surface Algorithm”. In Communica-tions of the ACM, 19, 10 (October 1976), pp. 547-554. Collinson, A. (1997). “Virtual Worlds”. The Cartographic Journal, Vol.34, No.2, pp117-124, Decem-ber 1997. Crispen, B. (1998). “VRML Works”. http://hiwaay.net/~crispen/vrml/faq/q1 (Date accessed 11/11/2002, last modified 11/06/1998). Cronin, P. (1999). “Report on the Applications of Virtual Reality Technology to Education”. HCRC, University of Edinburgh. http://www.cogsci.ed.ac.uk/~paulus/Work/Vranded/vr.htm (Date accessed 04/11/2002.Last modified 05/05/2000). Cusi-Redido, D. (2002). “Disseminating Philippine Census Data through the Web”. MSc Thesis ITC, Enschede, The Netherlands. CycloMedia International N.V. (2002). http://www.cyclomedia.nl/uk/index0.html (Date accessed 19/11/2002.Last modified 10/07/2002). Dale, P. F. (1994). “Land Information System”. In the book Geographic Information System com-piled by Maguire D. J., Goodchild,M. F., and David, R. W. Longman Group, UK Limited. Dale, P. F. and McLaughlin, J.D.”(1999). “Land Administration”. Oxford University Press Inc., New York. De Twentsche Courant Tubantia (2002). “Ons Huis scoort met digitale diensten”. Enschede, The Netherlands. Diotin, R. and Kooy, J. (1995). “Dynamic visualization of spatial data using virtual reality tech-niques”. In Proceedings of the joint European Conference on Geographical Information. The Hague, Netherlands, pp145-50 Dodge, M., Doyle, S., Smith, A. and Fleetwood, S., (1998). “Towards the Virtual City: VR and Inter-net GIS for Urban Planning”. Virtual Reality and Geographical Information Systems, Birkbeck College, London. Employment Development Department (2002) - Real Estate Agents Http://www.calmis.cahwnet.gov/file/occguide/REALAGNT.HTM. (Date accessed 11/09/2002, last modified 29/04/2002) Encarnação, J. L., Astheimer, W., Felger, T., Frühauf, M., Göbel, M. and Müller, S. (1993). “Graphics and Visualization: The essential features for the classification systems”. In ICCG Proceed-ings. Bombay. Fairall, J. (2002). “Visualising the World”.GIS User, April-May 2002, pp31-35. Fairbairn, D. and Parsey, S. (1996). “ The use of VRML for Cartographic Presentation”. In ICA commission on visualization meeting, August 16-18, Delft, The Netherlands. Fällman, D., Backman, A., and Holmlund, K. (1999). “VR in Education: An Introduction to Mul-tisensory Constructivist Learning Environments”. Conference on University Pedagogy,Umeå Univer-sity, Umeå. February 18-19. Electronic Archive: http://daniel.fallman.org/resources/papers/Fallman_VRIE.pdf (Date accessed 31/10/2002.Last modi-fied 31/10/2002).

BIBLIOGRAPHY

BIBLIOGRAPHY-3

Fisher, P. and Unwin, D. (2002). “ Virtual Reality in Geography: An introduction”. In Virtual Reality in Geography, Taylor and Francis Publishers, London. pp.1-4. Georgia Tech: College of Computing (1997). “Questionnaire Design” http://www.cc.gatech.edu/classes/cs6751_97_winter/Topics/quest-design/(Date accessed 04/11/2002, last modified 03/02/1997). Gillings, M. and Goodrick G. (1996). “ Sensuous and reflective GIS exploring visualizations and VRML”. http://intarch.ac.uk/journal/issue1/gillings_toc.html (Date accessed 11/11/2002.Last modified 11/09/1996). Göbel, M. and Neugebauer, J. (1993). “The Virtual Reality Demonstration Centre”In Computer & Graphics Vol.17, No. 6, pp 627-631. Pergamon Press Ltd. Gruen, A., Zhang, L. and Visnovcova, J. (2001). “Automatic Reconstruction and Visualization of a Complex Buddha Tower of Bayon, Angkor, Cambodia. In the book Optical 3-D Measurement Tech-niques V edited by Grü A. & Kahman H. Copy und Druck,Wien. Haber, R. B. and McNabb, D. A. (1990). “Visualisation idioms: a conceptual model for scientific visualization systems. In Nielson G., M., GIS, Shriver B., Rosenblum L., J., (eds) Visualisation in Sci-entific Computing, pp 74-93. Hartman, J. and Wernecke, J. (1996). “The VRML 2.0 Handbook: Building Moving Worlds on the Wed. New York: Addison-Wiley. Henssen, J. L. G. (1996) “Land Registration/Cadastre: General Aspects, Land Tenure, Land Man-agement”. Lecture Notes for Geo-information Courses. ITC, Enschede, the Netherlands. Horne, G., Pailthorpe, B., Simons, B., Arnison, M. and Bordes, N. (1995). “Scientific Visualization using VRML”. http://www.csu.edu.au/special/conference/apwww95/papers95/ghorne/ghorne.html. (Date accessed 24/6/2002, last modified 04/04/2002). Jacobson, R. (1994). “Virtual worlds capture spatial reality”. GIS WORLD. December 36-9 Jiang, B. (1996). “Cartographic Visualization: Analytical and Communication Tools”. In Cartogra-phy, Vol.25, No.2, December 1996. Jiang, B. and Ormerling, F. J. (1997). “Cybermap: the map for Cyberspace”. The Cartographic Jour-nal Vol.34, No.2, pp111-116, December 2000. Keppell, M. and Macpherson, C. (1997) “Is the Elephant Really There? - Virtual Reality in Educa-tion”.�http://dtls.cqu.edu.au/confsem/derun/vr/contents.html (Date accessed 1/10/2002, last modified 23/04/2002). Köbben, B. (2002). “3D Perception: How do we ‘see’ 3D?” In GFM 2 and 3 elective, module 9 lec-ture notes, ITC, The Netherlands. Koehne, J. and Howard, J. (2001). “PropertyAssist: A New Approach to the Delivery of property Information to South Australia. Cartography, Vol. 30, No. 1. Kofler, M., Rehatschek, H. and Gruber M (2002). “A database for a 3D GIS for urban environ-ments supporting photo-realistic visualization”. http://www.icg.tu-graz.ac.at/isprs96/paper.html (Date accessed 27/09/2002, last modified 27/09/2002). Kraak, M. J (1993). “Three-dimensional Map design”. In the Cartogrphic Journal Vol.30, pp 188-194. Kraak, M. J. (1994). “Interactive modelling environment for 3-d Maps; Interface and function issues. Visualization in Modern Cartography, (eds.) MacEachrea and Taylor. Kraak, M. J. and Ormeling, F. J. (1996), “Cartography: Visualisation of Spatial Data”. Addison Wesley Longman Limited, England.

BIBLIOGRAPHY

BIBLIOGRAPHY-4

Kraak, M. J., Smets, G. and Sidjanin P. (1999). Virtual Reality, the new 3-D Interface for Geo-graphic Information Systems. In Spatial Multimedia and Virtual Reality, (eds.) Câmara A., S. and Raper J. Taylor and Francis Publishers. pp132-146. Kraak, M. J. (2002). “Visual Exploration of virtual environments”. In Fisher, P. and Unwin, D. (eds). Virtual Reality in Geography, Taylor and Francis Publishers, London. pp.58-67 Krathwohl, D., R. (1993). “ Methods of Educational and Social Science Research; An integrated Ap-proach. Longman Publishers, pp 360-400 Kumar, R. (1999). “Research Methodology: A Step-By-Step guide for Beginners”. SAGE Publica-tion. London. Thousand Oaks. New Delhi. Lammi, J. (1997). “ Spatially constrained building extraction using edge finding and model fitting”. In Proceedings of ISPRS, Vol. 32 Part 3-4W2, Stuggart, September 17-19, pp 115-119. Legal Information Institute (2002). Real Property (land law): An Overview. http://www.law.cornell.edu/topics/real_property.html (Date accessed 12/09/2002, last modified 12/09/2002). Löffelmann, H. (1998). “Visualising Local Properties and Characteristics Structures of Dynamical Systems”. PhD Thesis Vienna University of Technology, Vienna, Austria. Electronic Archive: http://www.VRVis.at/vis/resources/diss-HL/diss-lowres.pdf (Date accessed 27/09/2002, last modified 20/11/1998). Lovett, A., Kennaway, R., Sünnenberg, G., Cobb, D., Dolman, P., O’Riordan, T. and Arnold, D. (2002). “Visualizing sustainable agricultural landscapes”. In Fisher, P. and Unwin, D. (eds). Virtual Reality in Geography. Taylor and Francis Publishers, London. pp102-130. MacEachren, A., M., Kraak, M., J. and Verbree, E. (1999). “Cartographic issues in the design and application of geospatial virtual environments”. In CD-ROM Proceedings 19th ICA Conference, Ot-tawa, August 14-21, 1999. McCormick, B., DeFanti, T. A. and Brown, M. D. (eds) (1987). “Visualisation in Scientific Com-puting. Special issue ACM SIGGRAPH Computer Graphics 1892: ppI-41. McGaughey, R. J. (1982). “Techniques for Visualizing the Appearance of Forest Operations”. In Journal of Forestry 96(6), June 1998, pp 9-14. Also In Electronic Archive: http://forsys.cfr.washington.edu/viztools.html (Date accessed 08/10/2002, last modified 14/02/1982). Ogao, P.J. (1997). “Visualization of 3-D Spatial Data Using Virtual Reality Model Language (VRML)”. MSc Thesis ITC, Enschede, The Netherlands. Ogao, P.J. (2001). “ Collaborative visualization of geospatial data in the WWW”. 5th AfricaGIS Con-ference and Exhibition. 5-9 November, Nairobi, Kenya. Ogao, P.J. (2002). “Exploratory Visualization of Temporal Geospatial Data Using Animation”. Phd Thesis ITC, Enschede, The Nethelands. ITC Dissertation No. 89. Online Dictionary.com (2002). “Dictionary.com” http://www.dictionary.com/search?q=real%20estate%20agent (Date accessed 23/08/2002, last modi-fied 23/08/2002) Paintsil, J. (I997) “3D Topographic data by Aerial Photogrammetry”. MSc Thesis ITC, Enschede, The Netherlands. Parallel Graphics (2002). “ Cortona VRML Client”. http://www.parallelgraphics.com/products/cortona/(Date accessed 14/11/2002, last modified 14/11/2002).

BIBLIOGRAPHY

BIBLIOGRAPHY-5

Pesce, M. D., Kennard, P. and Parisi, A. S. (1994). “ Cyberspace. Proceedings of the First Interna-tional Conference on the World Wide Web, May 25-26,1994, CERN, Geneva. Pfund, M. (2002). “3D GIS Architecture”. GIM Interntional Magazine. February 2002, 2, Vol16. Plewe, B. (1995). “Geography of Cyberspace”. Geography Department, University of Buffalo,USA. http://www.geog.buffalo.edu/%7Eplewe/cyberspace/ (Date accessed 13/8/2002 last updated 24/7/2002). Redido-Cusi, D. (2002). “ Disseminating Philippine Census Data through the Web”. MSc Thesis ITC, Enschede, The Netherlands. Rhyne, T. M. (1999). “ A commentary on GeoVRML: A tool for 3D representation of geo-referenced data on the web”. International Journal of Geographic Information Science, 13, 4, pp439-443. Rubin J. (1994). “ Handbook of Usability Testing”. John Wiley & Sons, Inc. USA Sester, M. (1998). “Acquiring Transition Rules Between Multiple Representations in a GIS”. In Pro-ceedings of ISPRS, March-April, Tempa, Florida, USA.CD-ROM. Sithole, G. (1997). “Digital Photogrammetry for Automatic Photo-Texture Extraction”. MSc Thesis ITC, Enschede, The Netherlands. Slocum, T., Yoder, S., Kessler, F. and Sluter, R. (2000). “Map time: Software for exploring spatio-temporal data associated with point locations”. Cartographica, Vol. 37, No.1, pp 15-31. Stoter, J. (2002). “3D Cadastres: State-of-the art”. GIM International Magazine. February 2002, 2, Vol16. Stoter, J. and van Oosterom, P. (2001). “Incorporating 3D Geo-Objects into a 3D Geo-database: an efficient use of geo-data”. In: Proceedings Geo-informatics and DMGIS 2001, the 3rd ISPRS Work-shop on Dynamic & Multi-dimensional GIS, the 10th annual conference of CPGIS on Geoinformatics, Asian Institute of Technology, Bangkok, Thailand, 23-25 May 2001. Electronic Archive: http://www.gdmc.nl/stoter/ (Date accessed 16/09/2002, last modified 25/07/2002). Tang, H. and Bishop, I. D. (2002). “Interaction Methodologies for Interactive Forest Modelling and Visualization”. The Cartographic Journal Vol.39, No.1, pp27-35, June 2002. Tempfli, K. (1998). “Urban Topographic Data and Texture by Digital Photogrammetry”. In Proceed-ings of ISPRS, March-April, Tempa, Florida, USA.CD-ROM. van Elzakker, C. P. J. M. (1999). “Thinking Aloud Method about Exploratory Cartography”. 19th international Cartography Conference and 11th General Assembly of ICA, Ottawa, Ontario, Canada, August 14-21, 1999. Verbree, E., Maren, G. V., Germs, R., Jansen, F. and Kraak, M. J. (1999). ‘ Interaction in virtual world views - linking 3D GIS with VR”. In International Journal of Geographical Information Sci-ence. Vol. 13, No.4, pp 385-396. Visvalingam, M. (1994). “Visualisation in GIS, Cartography and ViSC”. In Hearnshaw and Unwin (eds.). Visualization in Geographical Information Systems, John Wiley and Sons: Chichester, pp 18-25. Zeleznik, M. P. (1998). “3D Visualization: What Does it Mean?” In XII International Conference Proceedings on the Use of Computers in Radiation Therapy, May 27-30, 1997, Salt Lake City, UT, pp. 374-78.Electronic Archive: http://opendx.npaci.edu/publications/medical/vispaper/paper.html (Date accessed 23/09/2002, last modified 08/04/2000) Zlatanova, S. (2000). “3-D GIS for Urban Development”. PhD Thesis ITC, Enschede, The Nether-lands and ICGV, Graz, Austria.

APPENDICES

APPENDIX-1

APPENDICES

Appendix A: 2D Map in ArcGIS (Part of Enschede Top 10 vector data and Cadastral data)

Appendix A1: Part of Enschede Top 10 vector data

Appendix A2: Part of Enschede 2D Cadastral map

APPENDICES

APPENDIX-2

Appendix B: Outputs of the Testing Stage.

Appendix B1: Selected Property in a 2D map

Appendix B2: Selected Property in a 2½D map

APPENDICES

APPENDIX-3

Appendix B3: Side 3D View of the property Pre-Defined Building in the Virtual Environment using Cortona plug-in for web browser (Internet Explorer)

Appendix B4: Plan 3D View of the property Pre-Defined Building in the Virtual Environment using Cortona plug-in for web browser (Internet Explorer)

Pre-Defined Tree in ISB

Created 3D Build-ing using photos

Pre-Defined Building in ISB

Road layout

Land use areas

APPENDICES

APPENDIX-4

Appendix C: Outputs of the Implementation Stage.

Appendix C1: Selected Properties in 2D using the Cadastral map

Appendix C2: Selected Properties in 2½D using the Cadastral map

APPENDICES

APPENDIX-5

Appendix C3: Selected Properties in 3D using the Cadastral map

Appendix C4: Realistic view of the Properties in 3D using the Cadastral map

APPENDICES

APPENDIX-6

Appendix D: Tabulation of Angular, Base and Derived Measurements

Appendix D1: Tabulation of Angular Measurements and Base Measurements from Field of the Test-ing Stage

Name of Building

Number of Steps

Number of Re-cordings Top of Building Bottom of Building

Angular

Scale % Scale Angular

Scale % Scale 1 38 73 1 2

Dish Hotel 70 2 37 74 2 3 3 38 74 1 2 1 27 50 1 2

59 2 27 51 1.5 3 Opposite IT's 3 27 50 1 2

1 16 29 1 2 59 2 16 29 1 2 Opposite Ho-

geschool 3 16 29 1 2 Appendix D2: Tabulation of Derived Measurements from Angular Measurements and Base Meas-urements of the Testing Stage

Name of Building

Number of Obsers.

Top of Build-ing

Bottom of Building

Height= [AC * (tan a0 + tan b0)]

Number of Steps (n)

Angular Scale (a0)

Angular Scale (b0)

Base Length

=[n * 0.7m]

Mean Height in metres

1 38 1 39.1 Dish Hotel 70 2 37 2 49 38.6 39

3 38 1 39.1 1 27 1 21.8

59 2 27 1.5 41.3 22.1 22 Opposite IT's 3 27 1 21.8

1 16 1 12.6 59 2 16 1 41.3 12.9 13 Opposite

Hogeschool 3 16 1 12.6

APPENDICES

APPENDIX-7

Appendix D3: Tabulation of Angular Measurements, Base Measurements and Derived Measurements of the Implementation Stage.

Name of Building

Number of Observations

Top of Building

Bottom of Building

Height= [AC * (tan a0 + tan

b0)]

Number of

Steps (n)

Angular Scale (a0)

Angular Scale (b0)

Base Length

=[n * 0.7m]

Mean Height in metres

1 19 6 5.4 Ramelerbrink 18 2 18 4 12 4.7 5 3 18 4 4.7

APPENDICES

APPENDIX-8

Appendix E: Referenced and Edited Images

Appendix E1: MVC-404F.JPG

Appendix E2: MVC-404F.JPG which was relabelled HogesFS

APPENDICES

APPENDIX-9


Appendix E4: MVC-406F.JPG which was relabelled HogesRS

APPENDICES

APPENDIX-10


Appendix E6: MVC-408F.JPG which was relabelled HogesBS

APPENDICES

APPENDIX-11

Appendix E7: Image Cropping and Image Enhancement

Image before cropping and image enhance-ment

Image after cropping

Image after Cropping and Image enhance-ment

APPENDICES

APPENDIX-12

Appendix F: Letters Appendix F1: Letter Sent to Real Estate Agents for fixing Interview

APPENDICES

APPENDIX-13

APPENDICES

APPENDIX-14

Appendix F2: Letter Sent to Real Estate Agents for Testing Prototype

APPENDICES

APPENDIX-15

APPENDICES

APPENDIX-16

Appendix G: Interview Questions- A survey of user requirements for property visuali-zation

The objective behind the survey is to find real estate agents’ requirements for visualizing property in the urban environment for the general public in times of sale of properties. In other words: find out how images of properties are conveyed to clients.

A. Some statements 1. Statement: “Potential buyers want to first see an image/images of the property on hardcopy

or on the computer screen, then visit the property”. a. Totally Agree b. Partially Agree c. Neither Agree or Disagree d. Partially Disagree e. Totally Disagree

2. Statement: “First presenting image/images of property on the computer screen, then visit the

property is more effective for real estate agents”. a. Totally Agree b. Partially Agree c. Neither Agree or Disagree d. Partially Disagree e. Totally Disagree

3. Statement: “A still (or 2D) image of the property does not convey enough information about

the site and its surroundings to buyers.” a. Totally Agree b. Partially Agree c. Neither Agree or Disagree d. Partially Disagree e. Totally Disagree

B. Current System 4. How is information about a property for sale conveyed to the general public when a buyer

visits the office? a. Hardcopy Picture b. Softcopy Picture on computer c. Visiting the site d. Others?…………………………………………………………………………….

APPENDICES

APPENDIX-17

5. Are there any drawbacks in the your current visualization system?

i) No a. I am satisfied with hardcopy pictures b. I am satisfied with clicking techniques to show several pictures on the computer c. Others?……………………………………………………………………………………

ii) Yes

a. I can only present…………………………………………………………………………… b. I like to apply…………………………………………………………………………… c. Others?………………………………………………………………………………

6. The available resources to produce images for presentation are?

a. Conventional camera b. Digital camera c. Others?……………………………………………………………………………………

7. i) If you do not display images on the computer, what is the main reason? …………………………………………………………………………………………………………. …………………………………………………………………………………………………………. ii) If you use images on the computer, which technique(s) do you use?

a. Scanning Pictures b. Downloading images into Computer from camera c. Others?……………………………………………………………………………………

8. What kind of enhancement do you currently apply on the image (for example to get sharp

images)? a. None b. Others?……………………………………………………………………………………

9. i) Do you include the surroundings of property? a. No b. Yes

ii) If Yes

a. Adjacent roads b. Adjacent buildings c. Others?……………………………………………………………………………

APPENDICES

APPENDIX-18

10. Which software for the presentation of the property do you use? a. PowerPoint Presentations of Pictures b. Adobe Photoshop c. Others?………………………………………………………………………………

11. How do you judge the quality of image/images of property presented to buyer?

a. Image or picture (is/ is not) sharp b. Building (well/not well) depicted in picture. c. Others?…………………………………………………………………………………

C. Wishes and needs 12. What do you require to improve the visualization of property information for buyers?

a. Pictures linked to a representation of the geometry of the property site and its surround-ings.

b. Pictures linked to cadastral or real world location c. Present all sides of building in three-dimension d. Possibilities to rescale the representation e. Others……………………………………………………………………………………….

13. Do you have any experience with virtual environments (VE)? i) No ii) Yes, namely with:

a. Desktop Virtual Reality b. Others?……………………………………………………………………………………

14. Please react to the following statement: “ An interactive 3D visualization of property would

be ideal.” a. Totally Agree b. Partially Agree c. Neither Agree or Disagree d. Partially Disagree e. Totally Disagree

D. Final question 15. Are you interested in participating in the interactive 3D prototype visualization system?

i) No ii) If Yes,

Please provide us with the following details of contacts: Name:………………………………………………………………………………………… Contact address:……………………………………………………………………………….. E-mail address:………………………………………………………………………………… Phone Number:…………………………………………………………………………………

Thank you for your sincere co-operation.

APPENDICES

APPENDIX-19

Appendix H: The Suunto Altimeter (Source: Brack, 2001)

APPENDICES

APPENDIX-20

Appendix I: The questionnaire and the structured tasks used for the ‘think aloud’ re-search method

Appendix I1: The Questionnaire

APPENDICES

APPENDIX-21

Appendix I2: Structured tasks

1. View to any property within the application.

2. Identify the following views: the front view, the left view, the right view, and the back

view.

3. View to the property nearest or adjacent to the chosen property.

4. Can you find the entrance to the chosen property?

5. Can you identify any parking lots?

6. Identify any road linking the property.

7. View the whole environment in which the property is located.

visualization of land property for the urban environment · 2007. 4. 18. · • donny scholten,...

Documents