Extended Biographical Note

Dr Konstantinos Zagoris Post-doctoral Researcher, Department of Electrical and Computer Engineering,

Democritus University of Thrace.

KONSTANTINOS ZAGORIS, dipl. Eng, Ph.D.

Current Affiliation:

Postdoctoral Researcher, Democritus University of Thrace (D.U.TH.), Department of Electrical and Computer Engineering Contact Information:

Home Address: 6, Ioanninon Str., Xanthi, 67100, Xanthi, Greece Telephone: (+30) 6942983586, (+30) 2541070037 e-mail: kzagoris@ee.duth.gr, kzagoris@gmail.com web-site: http://www.zagoris.gr Personal Information: Date of Birth: December 15, 1978, Alexandroupolis, Greece Military Service: Fulfilled Nationality: Hellenic

EDUCATION AND WORK EXPERIENCE

2012 - now Post-doctoral Researcher, Department of Electrical and Computer Engineering, Democritus University of Thrace

2011 Post-doctoral Researcher, School of Computing, Sci-ence and Engineering, Manchester University of Sal-ford, United Kingdom

2003 - 2009 PhD at the Department of Electrical & Computer En-gineering, Democritus University of Thrace (Greece).

1996 - 2003 Graduated from the Faculty of Engineering of Democritus University of Thrace, in Xanthi (Greece) – Electrical & Computer Engineer, Degree Grade: 7.3/10.

1993 - 1996 Student at the 1st Lyceum of Alexandroupolis (Greece)

RESEARCHES – PUBLICATIONS

Publications in International Scientific Journals:

J[8] K. Zagoris, I. Pratikakis, A. Antonacopoulos, B. Gatos and N. Papamarkos «Distinction between handwritten and machine-printed text based on the bag of visual words model», Pattern Recognition , 47 (3), (1051 - 1062), 2014.

J[7] Zagoris, K. and Pratikakis, I. «Text detection on natural im-ages using mnemonic cellular automata», Journal of Cellular Automata, 9 (2-3), (183-194), 2014.

J[6] A. Arampatzis, K. Zagoris and S. A. Chatzichristofis, «Dy-namic Two-Stage Image Retrieval from Large Multimedia Da-tabases», Information Processing and Management, Vol. 49, No. 1, ppp 274-285, 2013. J[5] K. Zagoris , S. A. Chatzichristofis, N. Papamarkos , «Text Lo-

calization using Standard Deviation Analysis of Structure Ele-ments and Support Vector Machines», EURASIP Journal on Ad-vances in Signal Processing, Vol. 2011, No. 1, pp. 47, 2011. J[4] K. Zagoris, Ε. Kavallieratou and N. Papamarkos, «Image Retrieval Systems Based On Compact Shape Descriptor and Relevance Feedback Information», Journal of Visual Communi-cation and Image Representation, Vol. 22, pp. 378-390, 2011. J[3] K. Zagoris, Ε. Kavallieratou and N. Papamarkos, «A Docu-ment Image Retrieval System», Engineering Applications of Ar-tificial Intelligence, Vol. 23, No. 6, pp. 872-879,2010. J[2] S. A. Chatzichristofis, K. Zagoris, Y. S. Boutalis and N. Papa-markos, «Accurate image retrieval based on compact compo-site descriptors and relevance feedback information», Inter-national Journal of Pattern Recognition and Artificial Intelli-gence, Vol. 24, No. 2, pp. 207 – 244, 2010. J[1] K. Zagoris, N. Papamarkos, I. Koustoudis, «Color Reduction using the Combination of the Kohonen Self-Organized Feature Map and the Gustafson-Kessel Fuzzy Algorithm», Transactions on Machine Learning and Data Mining Vol. 1, No.1,pp. 31-46, 2008.

Publications in Scientific Conferences:

C[25] I. Pratikakis, K. Zagoris, B. Gatos, G. Louloudis and N. Stamatopoulos «ICFHR 2014 Competition on Handwritten KeyWord Spotting (H-KWS 2014)», 14th International Confer-ence on Frontiers in Handwriting Recognition (ICFHR’14), (814-819), 2014

C[24] K. Zagoris, I. Pratikakis and B. Gatos «Segmentation-based Historical Handwritten Word Spotting using Document-Specific Local Features», 14th International Conference on Frontiers in Handwriting Recognition (ICFHR’14), (9-14), 2014

C[23] L. Tsochatzidis, K. Zagoris, M. Savelonas and I. Pratikakis «SVM-based CBIR of breast masses on mammograms», Pro-ceedings of the 3rd International Workshop on Artificial Intelli-gence and Assistive Medicine co-located with the 21th Europe-an Conference on Artificial Intelligence (ECAI 2014), (26 - 30), 2014

C[22] Tsochatzidis, Lazaros and Zagoris, Konstantinos and Save-lonas, Michalis and Papamarkos, Nikos and Pratikakis, Ioannis and Arikidis, Nikolaos and Costaridou, Lena «Microcalcification oriented content-based mammogram retrieval for breast can-cer diagnosis», Imaging Systems and Techniques (IST), 2014 IEEE International Conference on, (257--262), 2014 C[21] K. Zagoris and I. Pratikakis, «Text Detection in Natural Images Using Bio-inspired Models», Document Analysis and

Recognition (ICDAR), 2013 12th International Conference on, (1370-1374), 2013 C[20] V. Virvilis, S. Perantonis, K. Zagoris and I. Pratikakis «A multimodal e-learning environment for orthopedics: design and implementation challenges», System Theory, Control and Computing (ICSTCC), 2013 17th International Conference, (728 - 732), 2013 C[19] K. Zagoris, I. Pratikakis, V. Virvilis and P. Perantonis «ORTHO-eMAN: A Web-based e-Training Platform for Ortho-pedics», NHIBE – 8th International Conference on New Hori-zons in Industry, Business and Education, (97-104), 2013 C[18] K. Zagoris and I. Pratikakis, «Scene Text Detection on Im-ages using Cellular Automata», ACRI 2012, Cellular Automata for Research and Industry”, Santorini, Greece, September 24 - 27, 2012. C[17] K. Zagoris, I. Pratikakis, A. Antonacopoulos, B. Gatos , N. Papamarkos, «Handwritten and Machine Printed Text Separa-tion in Document Images using the Bag of Visual Words Para-digm», 2012 International Conference on Frontiers in Hand-writing Recognition (ICFHR-2012), Bari, Italy, September 18-20, 2012. C[16] L. T. Tsochatzidis, A. Ch. Kapoutsis, N. I. Dourvas, S. A. Chatzichristofis, Y. S. Boutalis and K. Zagoris, «TSOKADO: An Image Search Engine Performing Recursive Query Recom-mendation Based on Visual Information», The Fifth Interna-tional Conference on Advances in Computer-Human Interac-tions (ACHI 2012), pp. 106-111, Valencia, Spain, January 30 – February 2, 2012. C[15] S. Α. Chatzichristofis, K. Zagoris, Y. Boutalis and Avi Ar-ampatzis, «A Fuzzy Rank-Based Late Fusion Method for Image Retrieval», The 18th International MultiMedia Modeling Con-ference (IMMM), Klagenfurt, Austria, January 4-6, 2012. C[14] A. Arampatzis, K. Zagoris and S. A. Chatzichristofis, «DUTH at ImageCLEF 2011 Wikipedia Retrieval», In Working Notes, Conference on Multilingual and Multimodal Information Access Evaluation (CLEF 2011), ImageClef, Wikipedia Retrieval Task, September 19-22, Amsterdam, Netherland, 2011. C[13] S. A. Chatzichristofis, K. Zagoris and A. Arampatzis, «The TREC Files: the (ground) truth is out there», Special Interest Group on Information Retrieval (SIGIR 2011), July 24-28, Bei-jing, China, 2011. C[12] K. Zagoris, S. A. Chatzichristofis and A. Arampatzis, «Bag-of-Visual-Words vs Global Image Descriptors on Two-Stage Multimodal Retrieval», Special Interest Group on Information Retrieval (SIGIR 2011) , July 24-28, Beijing, China, 2011.

C[11]. E. Chatzistavros, S. Α. Chatzichristofis, G. Stamatellos and K. Zagoris, «Comparative Performance Evaluation of Image Descriptors over IEEE 802.11B Noisy Wireless Networks», The 5th IEEE/FTRA International Conference on Multimedia and Ubiquitous Engineering (MUE 2011)», IEEE Computer Society, pp. 128-133, Loutraki, Greece, June 28-30, 2011. C[10] A. Arampatzis, K. Zagoris and S. A. Chatzichristofis, «Dy-namic Two-Stage Image Retrieval from Large Multimodal Da-tabases», European Conference on Information Retrieval (ECIR 2010)Dublin, Ireland, 18-21 April, 2011. C[9] A. Arampatzis, K. Zagoris and S. A. Chatzichristofis, «Fusion vs Two-Stage for Multimodal Retrieval», European Conference on Information Retrieval (ECIR 2010), Dublin, Ireland, 18-21 April, 2011. C[8] A. Arampatzis, S. A. Chatzichristofis, K. Zagoris, «Multime-dia Search with Noisy Modalities: Fusion and Multistage Re-trieval», CLEF (Notebook Papers/LABs/Workshops), 22-23 Sep-tember, Padua, Italy, 2010. C[7] K. Zagoris, A. Arampatzis and S. A. Chatzichristofis, «www.MMRetrieval.net: A Multimodal Search Engine», Pro-ceedings of the 3rd International Conference on SImilarity Search and APplications, SISAP 2010, Istanbul, Turkey, Septem-ber 18-19, 2010. C[6] K.Zagoris, S. Chatzichristofis, N. Papamarkos and Y. S. Bou-talis, «Automatic Image Annotation and Retrieval Using the Joint Composite Descriptor», 14th Panhellenic Conference on Informatics, pp. 143-147, Tripoli, Greece , 10-12 September, 2010. C[5] K. Zagoris and N. Papamarkos, «Text Extraction using Document Structure Features and Support Vector Machines», Computer Graphics and Imaging, pp. 9 – 48, 2010. C[4] K. Zagoris, S. A. Chatzichristofis, N. Papamarkos and Y. S. Boutalis, «img(Anaktisi): A Web Content Based Image Retriev-al System», International Workshop on Similarity Search and Applications, pp. 154-155, 2009. C[3] K. Zagoris, E. Kavallieratou and N. Papamarkos, «Develop-ing Document Image Retrieval System», IADIS International Conference CGV 2008 (part of MCCSIS 2008),pp. 119-126, 2008. C[2] K. Zagoris, N. Papamarkos and I. Koustoudis, «Color Re-duction using the combination of the Kohonen Self-Organized Feature Map and the Gustafson-Kessel fuzzy algorithm», In-ternational Conference on Machine Learning and Data Mining, pp. 703-715, 2007. C[1] K. Zagoris, N. Papamarkos, C. Chamzas, «Web Document Image Retrieval System Based on Word Spotting», IEEE Inter-

national Conference on Image Processing, pp. 477-480, 2006.

Other: D[1]. PhD Thesis titled “Content and Metadata Based Image Document Retrieval” under the supervision of the Professor Nikos Papamarkos, Image Processing Lab, University of Thrace - Greece.

EXPERIENCE IN RESEARCH

01 January 2013 – 31 December 2015: “TranScriptorium: Recogni-tion of Handwritten documents”, FP7-ICT-2011.8.2 – Collaborative project, EU. 01 March 2013 – 30 June 2015: “Content-Based Image Retrieval Services for Breast Cancer Computer - Aided Diagnosis”, CBIR-4-CAD – SINERGASIA, National Fund. 01 January 2012 – 31 December 2013: “A web-based e-training platform for Extended Human Motion Investigation in Orthope-dics (ORTHO-eMAN)”, Project nr. 2011-1-RO1-LEO05-15321. Leo-nardo da Vinci - Transfer of Innovation, EU. 01 January 2006 – 31 June 2009: Research Program “P.E.N.E.D - 03ΕΔ679” for my thesis: “Content and Metadata Based Image Doc-ument Retrieval”. 14 October 2004 – 31 August 2005: Research Program “P.E.S.P” for the collection and processing data concerning the inner evaluation of the postgraduate program of Democritus University of Thrace. 01 March 2004 – 30 November 2005: Research Program “PYTHAGORAS” for the development new color reductions tech-niques. 01 January 2004 – 31 December 2004: Research Program “OPERATIONAL PROGRAM SOCIETY OF INFORMATION” for the development advanced telematiques services for educational pur-poses.

REVIEWER AND SESSION CHAIR JOURNALS CONFERENCES

Session Chair:

The 14th PanHellenic Conference on Informatics, Tripoli, Greece, 10-12 September, 2010

Reviewer:

IEEE Transactions on Evolutionary Computation

Optics & Laser Technology

Journal of Visual Communication and Image Representa-tion

Signal Image and Video Processing

The 6th International Multi-Conference on Computing in the Global Information Technology «ICCGI 2011», June 19-24, 2011 – Luxembourg

The 5th International Conference on Multimedia and Ubiq-

uitous Engineering «MUE 2011» – Track 4: Multimedia Ser-vices and Applications, June 28-30, 2011, Crete, Greece

The 4th International Conference on Multimedia and Ubiq-uitous Engineering «MUE 2010» – Track 4: Multimedia Ser-vices and Applications. Organized by University of Alabama at Birmingham (UAB) Florida International University (FIU) June 28-30, 2010, Sofitel Hotel, Redwood City, San Francis-co Bay, California, USA.

34rd International ACM Conference on Research and De-velopment in Information Retrieval , «SIGIR 2011», July 24-28, 2011, Beijing, China.

19th ACM International Conference on Information and Knowledge Management «CIKM 2010», October 26-30, 2010, Toronto, Canada.

ADDITIONAL KNOWLEDGE / INTERESTS

Operating Systems: Microsoft Windows (2003, 2008, XP, Vista, 7), Linux (Ubuntu, Debian, RedHad, Suse), Mac OSX, OS/2 (eComsta-tion)

Programming Languages: C#, C, C++, Visual Basic, Fortran, JAVA, Delphi

Databases: Oracle, Microsoft SQL Server, MySQL, Access and LINQ/ADO.NET protocols.

Numerical Computing Environments: Mathematica, Matlab

Web Development: ASP.ΝΕΤ, ASP MVC, Coldfusion, PHP, WPF (XBAP) , Silverlight (BAP), Flash.

DEVELOPER/AUTHOR 1. Word Spotting as Recommender System http://vc.ee.duth.gr/ws

2. img(aktisi) on-line image retrieval engine: http://www.anaktisi.net

3. Document Image Retrieval Engine: http://orpheus.ee.duth.gr/irs2_5

4. Shape Retrieval Engine: http://orpheus.ee.duth.gr/cspd

5. MultiModal Retrieval Engine http://www.mmretrieval.net

6. TREC Files Evaluator http://cocytus.nonrelevant.net/thetrecfiles

7. Text Finder http://orpheus.ee.duth.gr/download/TextFinder_1.0.9.zip

8. Color Reduction Program http://orpheus.ee.duth.gr/download/pythagoras.zip

9. The Trec Files Evaluator http://thetrecfiles.nonrelevant.net/

FELLOWSHIPS 1. Member in Greece Technical Chamber (T.E.E.) (record number 98960) since March 2004.

2. Member of the IEEE 3. Member of ACM

SCIENTIFIC INTERESTS

1. Pattern Recognition. 2. Document Analysis and Retrieval. 3. Image Processing and Retrieval. 4. Color Analysis. 5. Fuzzy and Neural Algorithms for Classification. 6. Computer Vision 7. Computer Science, Programming 8. Databases: Distributed Architectures, Key-Values Storage,

Hashing Algorithms, Tree-Based Data Structures, Large Scale Data Management.

9. Computer Architectures and Microprocessors, Application on Digital Image Processing, Machine Vision and Pattern Recognition.

TEACHING EXPERIENCE

1. October 2008 – September 2009: Provide educational Ser-vices for the seminars of the company INFOLAB (http://www.infolab.gr/)

2. Spring Semester 2004-2005: Assistance in conducting the laboratory experiments of the undergraduate course “Elec-tric Circuits II”.

3. Spring Semester 2004 - 2005: Assistance in conducting the undergraduate course “Pattern Recognition”.

4. Fall Semester 2005 - 2006: Assistance in conducting the la-boratory experiments of the undergraduate course “Elec-tric Circuits I”.

5. Spring Semester 2005 - 2006: Assistance in conducting the laboratory experiments of the undergraduate course “Elec-tric Circuits II”.

6. Fall Semester 2006-2007: Assistance in conducting the la-boratory experiments of the undergraduate course “Elec-tric Circuits I”.

7. Fall Semester 2006-2007: Assistance in conducting the la-boratory experiments of the undergraduate course “Digital Image Analysis”.

8. Spring Semester 2006 - 2007: Assistance in conducting un-dergraduate course “Electric Circuits ΙΙ”.

9. Spring Semester 2006 - 2007: Assistance in conducting the undergraduate course “Pattern Recognition”.

10. Fall Semester 2007 - 2008: Assistance in conducting the la-boratory experiments of the undergraduate course “Elec-tric Circuits I”.

11. Spring Semester 2007 - 2008: Assistance in conducting the laboratory experiments of the undergraduate course “Elec-tric Circuits II”.

APPENDIX A BOOK OF ABSTRACTS

J[1] Color Reduction using the Combination of the Kohonen Self-Organized Feature Map and the Gustafson-Kessel Fuzzy Algorithm Authors: K. Zagoris, N. Papamarkos, I. Koustoudis Abstract: The color reduction in digital images is an active re-search area in digital image processing. In many applications such as image segmentation, analysis, compression and trans-mission, it is preferable to have images with a limited number of colors. In this paper, a color clustering technique which is a combination of a Kohonen Self Organized Featured Map (KSOFM) and a fuzzy clustering algorithm is proposed. Initially, we reduce the number of image’s colors by using a KSOFM. Then, using the KSOFM color clustering results as starting val-ues, we obtain the final colors by a Gustafson-Kessel Fuzzy Classifier (GKFC). Doing this, we lead to better color classifica-tion results because the final color classes obtained are not spherical.

J[2] Accurate image retrieval based on compact composite descriptors and relevance feedback information Authors: S. A. Chatzichristofis, K. Zagoris, Y. S. Boutalis and N. Papamarkos Abstract: In this paper a new set of descriptors appropriate for image indexing and retrieval is proposed. The proposed de-scriptors address the tremendously increased need for efficient content-based image retrieval (CBIR) in many application areas such as the Internet, biomedicine, commerce and education. These applications commonly store image information in large image databases where the image information cannot be ac-cessed or used unless the database is organized to allow effi-cient storage, browsing and retrieval. To be applicable in the design of large image databases, the proposed descriptors are compact, with the smallest requiring only 23 bytes per image. The proposed descriptors' structure combines color and tex-ture information which are extracted using fuzzy approaches. To evaluate the performance of the proposed descriptors, the objective Average Normalized Modified Retrieval Rank (ANMRR) is used. Experiments conducted on five benchmark-ing image databases demonstrate the effectiveness of the pro-posed descriptors in outperforming other state-of-the-art de-scriptors. Also, a Auto Relevance Feedback (ARF) technique is introduced which is based on the proposed descriptors. This technique readjusts the initial retrieval results based on user preferences improving the retrieval score significantly. An online demo of the image retrieval system img(Anaktisi) that

implements the proposed descriptors can be found at http://www.anaktisi.net.

J[3] A Document Image Retrieval System», Engineering Appli-cations of Artificial Intelligence Authors: K. Zagoris, Ε. Kavallieratou and N. Papamarkos Abstract: In this paper, a system is presented that locates words in document image archives. This technique performs the word matching directly in the document images bypassing character recognition and using word images as queries. First, it makes use of document image processing techniques, in or-der to extract powerful features for the description of the word images. The features used for the comparison are capable of capturing the general shape of the query, and escape details due to noise or different fonts. In order to demonstrate the ef-fectiveness of our system, we used a collection of noisy docu-ments and we compared our results with those of a commer-cial optical character recognition (OCR) package.

J[4] Image Retrieval Systems Based On Compact Shape De-scriptor and Relevance Feedback Information Authors: K. Zagoris, Ε. Kavallieratou and N. Papamarkos Abstract: One of the most important and most used low-level image feature is the shape employed in a variety of systems such as document image retrieval through word spotting. In this paper an MPEG-like descriptor is proposed that contains conventional contour and region shape features with a wide applicability from any arbitrary shape to document retrieval through word spotting. Its size and storage requirements are kept to minimum without limiting its discriminating ability. In addition to that, a relevance feedback technique based on Support Vector Machines is provided that employs the pro-posed descriptor with the purpose to measure how well it per-forms with it. In order to evaluate the proposed descriptor it is compared against different descriptors at the MPEG-7 CE1 Set B database.

J[5] Text Localization using Standard Deviation Analysis of Structure Elements and Support Vector Machines Authors: K. Zagoris , S. A. Chatzichristofis, N. Papamarkos Abstract: A text localization technique is required to success-fully exploit document images such as technical articles and letters. The proposed method detects and extracts text areas from document images. Initially a Connected Components Analysis technique detects blocks of foreground objects. Then,

a descriptor that consists of a set of suitable Document Struc-ture Elements is extracted from the blocks. This is achieved by incorporating an algorithm called Standard Deviation Analysis of Structure Elements (SDASE) which maximizes the separabil-ity between the blocks. Another feature of the SDASE is that its length adapts according to the requirements of the application. Finally, the descriptor of each block is used as input to a trained Support Vector Machines (SVM) that classifies the block as text or not. The proposed technique is also capable of adjusting to the text structure of the documents. A preliminary version of this work has been presented in [1]. Experimental results on benchmarking databases demonstrate the effective-ness of the proposed method.

J[6] Dynamic Two-Stage Image Retrieval from Large Multime-dia Databases Authors: A. Arampatzis, K. Zagoris and S. A. Chatzichristofis Abstract: A text localization technique is required to success-fully exploit document images such as technical articles and letters. The proposed method detects and extracts text areas from document images. Initially a Connected Components Analysis technique detects blocks of foreground objects. Then, a descriptor that consists of a set of suitable Document Struc-ture Elements is extracted from the blocks. This is achieved by incorporating an algorithm called Standard Deviation Analysis of Structure Elements (SDASE) which maximizes the separabil-ity between the blocks. Another feature of the SDASE is that its length adapts according to the requirements of the application. Finally, the descriptor of each block is used as input to a trained Support Vector Machines (SVM) that classifies the block as text or not. The proposed technique is also capable of adjusting to the text structure of the documents. A preliminary version of this work has been presented in [1]. Experimental results on benchmarking databases demonstrate the effective-ness of the proposed method.

C[1] Web Document Image Retrieval System Based on Word Spotting Authors: K. Zagoris, N. Papamarkos, C. Chamzas Abstract: Nowadays, the huge non-indexing quantities of im-age archives (especially document images) require the devel-opment of intelligent tools for their retrieval with convenience comparable of the texts search engines. The proposed tech-nique addresses the document retrieval problem by a word matching procedure. It performs matching directly in the imag-es bypassing OCR and using word-images as queries. It is con-stituted of two different parts: The offline and the online oper-ation. In the offline operation, the archive of document images

is examined and the results are stored in a database. The online operation consists of the web interface, the creation of the word’s image and finally, the matching stage. The proposed matching process it can be described shortly as a two threshold rating system. Finally, the proposed system has been build and it can be found in at the web address: http://orpheus.ee.duth.gr/irs2.

C[2] Color Reduction using the combination of the Kohonen Self-Organized Feature Map and the Gustafson-Kessel fuzzy algorithm Authors: K. Zagoris, N. Papamarkos and I. Koustoudis Abstract: The color of the digital images is one of the most im-portant components of the image processing research area. In many applications such as image segmentation, analysis, com-pression and transition, it is preferable to reduce the colors as much as possible. In this paper, a color clustering technique which is the combination of a neural network and a fuzzy algo-rithm is proposed. Initially, the Kohonen Self Organized Fea-tured Map (KSOFM) is applied to the original image. Then, the KSOFM results are fed to the Gustafson-Kessel (GK) fuzzy clus-tering algorithm as starting values. Finally, the output classes of GK algorithm define the numbers of colors of which the image will be reduced.

C[3] Developing Document Image Retrieval System Authors: K. Zagoris, E. Kavallieratou and N. Papamarkos Abstract: A system was developed able to retrieve specific documents from a document collection. In this system the que-ry is given in text by the user and then transformed into image. Appropriate features were in order to capture the general shape of the query, and ignore details due to noise or different fonts. In order to demonstrate the effectiveness of our system, we used a collection of noisy documents and we compared our results with those of a commercial OCR package.

C[4] img(Anaktisi): A Web Content Based Image Retrieval Sys-tem Authors: K. Zagoris, S. A. Chatzichristofis, N. Papamarkos and Y. S. Boutalis Abstract: img(Anaktisi) is a C#/.NET content base image re-trieval application suitable for the web. It provides efficient re-trieval services for various image databases using as a query a sample image, an image sketched by the user and keywords. The image retrieval engine is powered by innovative compact and effective descriptors. Also, an Auto Relevance Feedback

(ARF) technique is provided to the user. This technique read-justs the initial retrieval results based on user preferences im-proving the retrieval score significantly. img(Anaktisi) can be found at http://www.anaktisi.net.

C[5] Text Extraction using Document Structure Features and Support Vector Machines Authors: K. Zagoris and N. Papamarkos Abstract: In order to successfully locate and retrieve document images such as technical articles and newspapers, a text locali-zation technique must be employed. The proposed method de-tects and extracts homogeneous text areas in document imag-es indifferent to font types and size by using connected com-ponents analysis to detect blocks of foreground objects. Next, a descriptor that consists of a set of structural features is ex-tracted from the merged blocks and used as input to a trained Support Vector Machines (SVM). Finally, the output of the SVM classifies the block as text or not.

C[6] Automatic Image Annotation and Retrieval Using the Joint Composite Descriptor Authors: K. Zagoris, S. Chatzichristofis, N. Papamarkos and Y. S. Boutalis Abstract: Capable tools are needed in order to successfully search and retrieve a suitable image from large image collec-tions. Many content-based image retrieval systems employ low-level image features such as color, texture and shape in order to locate the image. Although the above approaches are successful, they lack the ability to include human perception in the query for retrieval because the query must be an image. In this paper a new image annotation technique and a keyword based image retrieval system are presented, which map the low-level features of the Joint Composite Descriptor to the high-level features constituted by a set of keywords. One set consists of colors-keywords and the other set consists of words. Experiments were performed to demonstrate the effec-tiveness of the proposed technique.

C[7] www.MMRetrieval.net: A Multimodal Search Engine, Authors: K. Zagoris, A. Arampatzis and S. A. Chatzichristofis Abstract: We introduce an experimental search engine for mul-tilingual and multimedia information, employing a holistic web interface and enabling the use of highly distributed indices. Modalities are searched in parallel, and results can be fused via several selectable methods. The engine also provides multi-stage retrieval, as well as a single text index baseline for com-

parison purposes. Initial impressions on its effectiveness are positive, while its efficiency may easily be improved.

C[8] Multimedia Search with Noisy Modalities: Fusion and Multistage Retrieval Authors: A. Arampatzis, S. A. Chatzichristofis, K. Zagoris Abstract: We report our experiences from participating to the controlled experiment of the ImageCLEF 2010Wikipedia Re-trieval task. We built an experimental search engine which combines multilingual and multi-image search, employing a ho-listic web interface and enabling the use of highly distributed indices. Modalities are searched in parallel, and results can be fused via several selectable methods. The engine also provides multistage retrieval, as well as a single text index baselines for comparison purposes. Experiments show that the value added by image modalities is very small when textual annotations ex-ist. The contribution of image modalities is larger when search is performed in a 2-stage fashion, i.e., using image search for re-ranking a smaller set of only the top results retrieved by text. Furthermore, first splitting annotations to many modali-ties with respect to natural language and/or type and then fus-ing results has the potential of achieving better effectiveness than using all textual information as a single modality. Con-cerning fusion, the simple method of linearly combining evi-dence is found to be the most robust, achieving the best effec-tiveness.

C[9] Fusion vs Two-Stage for Multimodal Retrieval Authors: A. Arampatzis, K. Zagoris and S. A. Chatzichristofis Abstract: We compare two methods for retrieval from multi-modal collections. The first is a score-based fusion of results, retrieved visually and textually. The second is a two-stage method that visually re-ranks the top-K results textually re-trieved. We discuss their underlying hypotheses and practical limitations, and contact a comparative evaluation on a stand-ardized snapshot of Wikipedia. Both methods are found to be significantly more effective than single-modality baselines, with no clear winner but with different robustness features. Never-theless, two-stage retrieval provides efficiency benefits over fusion.

C[10] Dynamic Two-Stage Image Retrieval from Large Multi-modal Databases Authors: A. Arampatzis, K. Zagoris and S. A. Chatzichristofis Abstract: Content-based image retrieval (CBIR) with global fea-

tures is notoriously noisy, especially for image queries with low percentages of relevant images in a collection. Moreover, CBIR typically ranks the whole collection, which is inefficient for large databases. We experiment with a method for image re-trieval from multimodal databases, which improves both the effectiveness and efficiency of traditional CBIR by exploring secondary modalities. We perform retrieval in a two-stage fashion: first rank by a secondary modality, and then perform CBIR only on the top-K items. Thus, effectiveness is improved by performing CBIR on a ‘better’ subset. Using a relatively ‘cheap’ first stage, efficiency is also improved via the fewer CBIR operations performed. Our main novelty is that K is dy-namic, i.e. estimated per query to optimize a predefined effec-tiveness measure. We show that such dynamic two-stage set-ups can be significantly more effective and robust than similar setups with static thresholds previously proposed.

C[11] Comparative Performance Evaluation of Image De-scriptors over IEEE 802.11B Noisy Wireless Networks Authors: E. Chatzistavros, S. Α. Chatzichristofis, G. Stamatellos and K. Zagoris Abstract: In this paper we evaluate the image retrieval proce-dure over an IEEE 802.11b Ad Hoc network, operating in 2.4GHz, using IEEE Distributed Coordination Function CSMA/CA as the multiple access scheme. IEEE 802.11 is a widely used network standard, implemented and supported by a variety of devices, such as desktops, laptops, notebooks, mobile phones etc., capable of providing a variety of different services, such as file transfer, internet access et.al. Therefore, we consider IEEE 802.11b being a suitable technology to investigate the case of conducting image retrieval over a wireless noisy channel. The model we use to simulate the noisy environment is based on the scenario in which the wireless network is located in an out-door noisy environment, or in an indoor environment of partial LOS - Line-of-sight power. We used a large number of de-scriptors reported in literature in order to evaluate which one has the best performance in terms of Mean Average Precision - MAP values under those circumstances. Experimental results on known benchmarking database show that the majority of the descriptors appear to have decreased performance when transferred and used in such noisy environments.

C[12] Bag-of-Visual-Words vs Global Image Descriptors on Two-Stage Multimodal Retrieval Authors: K. Zagoris, S. A. Chatzichristofis and A. Arampatzis Abstract: Using Bag-of-Visual Words (BOVW) is fast becoming a widely used representation for content based image retrieval mainly, because of their better retrieval effectiveness over global feature representations on collections with images being

near-duplicate to the test queries. In this experimental study we demonstrate that this advantage of BOVW is diminished when visual diversity is enhanced by using a secondary modali-ty, such as text, to pre-filter images. In detail, the TOP-SURF descriptor is evaluated against Compact Composite Descriptors on a two-stage image retrieval system, which first uses a text modality to rank the collection and then perform CBIR only on the top-K items.

C[13] The TREC Files: the (ground) truth is out there Authors: S. A. Chatzichristofis, K. Zagoris and A. Arampatzis Abstract: Traditional tools for information retrieval (IR) evalua-tion, such as TREC’s trec_eval, have outdated command-line in-terfaces with many unused features, or ‘switches’, accumulat-ed over the years. They are usually seen as cumbersome appli-cations by new IR researchers, steepening the learning curve. We introduce a platform independent application for IR evalu-ation with a graphical easy to use interface: the TREC_Files Evaluator. The application supports most of the standard measures used for evaluation in TREC, CLEF, and elsewhere, such as MAP, P10, P20, and bpref, as well as the Averaged Normalized Modified Retrieval Rank (ANMRR) proposed by MPEG for image retrieval evaluation. Additional features in-clude a batch mode and statistical significance testing of the results against a pre-selected baseline.

C[15] A Fuzzy Rank-Based Late Fusion Method for Image Re-trieval Authors: S. A. Chatzichristofis, K. Zagoris, Y. S. Boutalis and A. Arampatzis Abstract: Rank-based fusion is indispensable in multiple search setups in lack of item retrieval scores, such as in meta-search with non-cooperative engines. We introduce a novel, simple, and efficient method for rank-based late fusion of retrieval re-sult-lists. The approach taken is rule-based, employs a fuzzy system, and does not require training data. We evaluate on an image database by fusing results retrieved by three MPEG-7 descriptors, and find statistically significant improvements in effectiveness over other widely used rank-based fusion meth-ods.

C[16] TSOKADO: An Image Search Engine Performing Recur-sive Query Recommendation Based on Visual Information Authors: L. T. Tsochatzidis, A. Ch. Kapoutsis, N. I. Dourvas, S. A. Chatzichristofis, Y. S. Boutalis and K. Zagoris Abstract: This paper tackles the problem of the user's incapa-

bility to describe exactly the image that he seeks by introducing an innovative image search engine called TsoKaDo. Until now the traditional web image search was based only on the com-parison between metadata of the webpage and the user's tex-tual description, query. In the method proposed, images from various search engines are classified based on visual content and new tags are proposed to the user. Recursively, the results get closer to the user's desire. The aim of this paper is to pre-sent a new way of searching, especially in case with less query generality, giving greater weight in visual content rather than in metadata. An on-line early version of TsoKaDo is available at http://tsokado.nonrelevant.net.

C[17] Handwritten and Machine Printed Text Separation in Document Images using the Bag of Visual Words Paradigm Authors: K. Zagoris, I. Pratikakis, A. Antonacopoulos, Basilis Ga-tos , N. Papamarkos Abstract: In a number of types of documents, ranging from forms to archive documents and books with annotations, ma-chine printed and handwritten text may be present in the same document image, giving rise to significant issues within a digiti-sation and recognition pipeline. It is therefore necessary to separate the two types of text before applying different recog-nition methodologies to each. In this paper, a new approach is proposed which strives towards identifying and separating handwritten from machine printed text using the Bag of Visual Words paradigm (BoVW). Initially, blocks of interest are de-tected in the document image. For each block, a descriptor is calculated based on the BoVW. The final characterization of the blocks as Handwritten, Machine Printed or Noise is made by a Support Vector Machine classifier. The promising perfor-mance of the proposed approach is shown by using a con-sistent evaluation methodology which couples meaningful measures along with a new dataset.

C[18] Scene Text Detection on Images using Cellular Automa-ta Authors: K. Zagoris and I. Pratikakis Abstract: Textual information in images constitutes a very rich source of high-level semantics for retrieval and indexing. In this paper, a new approach is proposed using Cellular Automata (CA) which strives towards identifying scene text on natural images. Initially, a binary edge map is calculated. Then, taking advantage of the CA flexibility, the transition rules are changing and are applied in four consecutive steps resulting in four time steps CA evolution. Finally, a post-processing technique based on edge projection analysis is employed for high density edge images concerning the elimination of possible false positives. Evaluation results indicate considerable performance gains

without sacrificing text detection accuracy.