art neural nw

8/14/2019 Art Neural Nw

http://slidepdf.com/reader/full/art-neural-nw 1/6

Paint Coating Quality Inspection in Bridges Using Image Recognition

and Neural Networks

S.A.Dhivhya1 Aishwarya Murali2

II year Computer Science and Engineering

R.M.D Engineering College,Kavaraipettai. [email protected] [email protected]

Abstract

The most established technique of Visual Inspection of

painted body surfaces of bridges is a very timeconsuming and labor intensive process. Inaccurate or

subjective assessment techniques have been identified as

a critical obstacle to effective infrastructure or

constructed facilities management. This paper illustratesa more objective and reliable assessment method to

improve the conditions of the infrastructures or the

quality of constructed facilities. The proposed system will

automate the coating assessment process by using computers to analyze digital images of the areas to be

assessed based on the Back Propagation Network (BPN)

System in Neural Networks. The paint-Quality can be

inspected in four simple steps.

1. Introduction.

Engineers need accurate, consistent, and reliable

information regarding paint system performance for theeffective management and rehabilitation of bridge.Such

information includes quantitative paint system ratings and

accurate estimates of the useful life of the coating system.The extent of coating failure is most important in

determining when a coating is to be repaired. Currently,

the means available in the evaluation of coating

performance is limited. Visual rating of the coating

condition, which represents the most established

technique, is plagued by a number of significant

problems.

2.Earlier Paint Quality Assessment

Techniques

Conventionally ,a laser beam was made to fall on the

painted panel and on to a projection screen. Since the

light source is a coherent beam (has constant phase

relation), the amount of scatter observed in the reflected

image of the laser provides an indication of the quality of paint in bridges. In the past, inspection of scatter-pattern

was performed primarily by humans, because

conventional computer programming techniques that

could be used to automate the observation and scoring process suffered from a lack of flexibility, and were not

particularly robust.

By using a Back propagation type of neural network to

perform the quality-scoring operation, a system can be

constructed that captures the expertise of humaninspectors, and is relatively easy to maintain and update.

To improve the performance of the system, algorithmictechniques can be coupled with the above approach to

simplify the problem.

The main thrust of that research was the non-destructive

evaluation of bridge coatings using several techniques,

namely: Color Visible Imaging, Electrochemical

Impedance Spectroscopy (EIS), and Infrared Imaging.

The results were not reliable and inconsistent. The

research result presented a generic framework, but was

not developed to the implementation stage.

3.Base Concepts applied in our Coating

Quality Assessment Model

3.1. Artificial neural networksConventional computers concentrate on emulating human

thought processes, rather than actually how they are

achieved by the human brain. Neural computers, however,

take an alternative approach in that they directly model

the biological structure of the human brain and the way it

processes information (although at a much simpler level).

This necessitates a new kind of architecture, which, likethe human brain, consists of a large number of heavily

interconnected processing elements operating in parallel

manner. Such an architecture is now both technically and

commercially feasible to be deployed on a standard

computer (from the laptop and desktop to the mainframe)

and is certain to increase in general usage.

Neural networks are mathematical models, originally

inspired by biological processes in the human brain. They

are constructed from a number of simple processing

mailto:[email protected]






elements interconnected by weighted pathways to form

networks. Each element computes its output as a non-

linear function of its weighted input. When combined into

networks, these processing elements can implement

arbitrarily complex non-linear functions which can beused to solve classification, prediction or optimization

problems .An artificial neural network or simply a neuralnetwork is a system based on the operation of biological

neural networks, in other words, is an emulation of

biological neural system.

A Neural Network will have 3 basic parts which are

1)Input Layer 2)Hidden layer and 3)Output layer.

Although computing these days is truly advanced, thereare certain tasks that a program made for a common

microprocessor is unable to perform; even so a software

implementation of a neural network can be made. An

Artificial Neural Network is an adaptive system thatlearns to perform a function (an input/output map) from

data. Adaptive means that the system parameters are

changed during operation, normally called the training

phase.

Fig 1: Neural Network for Back Propagation Algorithm

After the training phase the Artificial Neural Network

parameters are fixed and the system is deployed to solve

the problem at hand (the testing phase). The Artificial

Neural Network is built with a systematic step-by-step

procedure to optimize a performance criterion or to follow

some implicit internal constraint, which is commonly

referred to as the learning rule. An Artificial NeuralSystem (ANS) that is found to be useful in addressing

problems requiring recognition of complex patterns and

performing nontrivial mapping functions is the Back

Propagation Network.

3.2. Back propagation (BPN)

A BPN network is designed to operate as a multi-layer,

feed forward network, using supervised mode of learning.The basic back propagation model is a three-layeredforward architecture. The first layer is the Input Layer, the

second layer the Hidden Layer, and the third layer the

Output Layer. Each layer contains a group of nodes that

are linked together with nodes from other layers by

connections among the nodes. Layers are connected only

to the adjacent layers. The network is a feed-forward

network, which means a Unit’s output can only originate

from a lower level, and a unit's output can only be passedto a higher level.

The Input Layer receives the features of the data that are

entered into the neural network. If n feature values are to be entered into the Input Layer, then there must be n

nodes; where n is the number of features supplied to the

net. The values are passed to the Hidden Layer through

connections from the Input Layer. The nodes in the firstlayer distribute the individual inputs to all of the nodes in

the Hidden Layer. The main function of the hidden layer

is to process the input data during the network training to

connect to the output, i.e. to do the regression process.

The connections between the layers are weighted to

emphasize or de-emphasize the relative value of the input.

Each node sums the n weighted inputs to the Hidden

Layer. The value of the summation can be different for each node, due to differently weighted connections between the first and second layers. The values that are

summed in the hidden nodes are then passed to the nodes

in the Output Layer via another set of weighted

connections.

Fig 2: BPN System for Paint Coating Quality Assessment

FRAMEGRABBER-DATA

ACQUISITION

SELF ORGANIZING

ALGORITHMUSER

INTERFACE-

RESULT

BACK PROPAGATION

NETWORK-

PROCESSING

STAGE



Step 4: ASSESSMENT RESULTS

The Assessment Results stage, where quantitative

measures of defects are obtained from the output of the previous stage. From mapped output of the neural

network, the whole image is represented as 0's or 1's. The

1 values represent the defect; hence, defects can be

identified and quantitatively measured. The system can be

trained to identify different types of defects according to

the specific application.

Once, the network was constructed (and trained), 10

sample images were taken from the snapshot using 2

different sampling techniques. In the first test, the samples

were selected randomly from the image (in the sense that

their position on the beam image was random); in the

second test, 10 sequential samples were taken, so as to

ensure that the entire beam was examined. In both cases,the input sample was propagated through the trained

BPN, and the score produced as output by the network was averaged across the 10 trials. The average score, as

well as the range of scores produced, were then provided

to the user for comparison and interpretation.

6. Examining Rust on a Steel Bridge-

An Example

The following example illustrates how a simple 3 pixel x

3 pixel mage could be used to illustrate the methodology.

STEP 1: Data Acquisition:

A digital image of a steel bridge coating area is obtained

and transferred to the computer.

STEP 2:Preprocessing

Image analysis software is used to enhance the image and

to represent the image in numerical format. Consider the

figure to be an image of a part of a rusted steel bridge beam with a coating on. The darker areas represent defect

object (rust) and the lighter areas represent background.

1 2 3

1 245 21 90

2 260 18 45

3 54 56 57

Table 1: Image Gray Levels

Pattern recognition is done and the numerical values are

classified into two clusters: rust or no rust. K-Means

algorithm is used for classification because it best fits thisspecific application in which the number of clusters to

classify must be predetermined.Given the data setobtained from the image numerical presentation ,the K-

means Algorithm is applied.

V= {245,21,90,260,18,45,54,56,57}

Any two values are selected randomly as centre.Let us assume that the initial centers selected are R1=245

and R2=56

Next, we assign each sample to the class with the closest

center, i.e. to class 1 if it is closer to R1=245 and to class

2 if it is closer to R2=56

.The following table shows the difference between each

pixel value and the initial two centers and hence to which

class it is assigned. The difference is calculated based onthe gray level, not based on the location of the pixel.

Pixel

Value

Difference

from R1

Difference

from R2

Class 1 Class 2

245 0 189 X

21 224 35 X

90 155 34 X

260 15 204 X18 227 38 X

45 200 11 X

54 191 2 X

56 189 0 X

57 188 1 X

Table 2: First Iteration of K-Means Algorithm

C1={245,260}

C2={21,90,18,45,54,56,57}

Fig:4: Graph Representing Initial Clustering

Initial Clusters with centers R1=245 and R2=56

X

Y

245

45260

9021

575654

18

2 31

1

2

3



learning algorithm. The learning algorithm changes the

functionality of the network to suit the problem by

modifying the values of the connection weights between

processing elements.

The time needed to develop a neural application is often

less than that in a conventional approach, since theinteraction between the analyst and the expert isminimized-there are no algorithms or rules to define.

The scope and accuracy of the finished application isimproved since the neural computer can be exposed to

many more examples than can be assimilated by a singlehuman. This system will make the assessment process

more objective, quantitative, and reliable.

8.Conclusion

The basic concept of this paper is to automate the

assessment process by taking digital images of the rustareas and analyzing the images to identify and measure

coating defects. Neural Networks do not require explicit

programming by an expert and are robust to noisy,

imprecise or incomplete data.

Furthermore, knowledge is encapsulated in a compact,

efficient way that can easily be adapted to changes in business environment. It saves time and reduces human

burden. Hence, the application of Neural Network andBPN technique is very crucial and of extremely

significant value. . The system will make the assessment

process more objective, quantitative, and reliable. This

vital technology can never be overlooked

9.References

[1] Neural Networks and their Applications in Industry

Dhruba J Sarma and Susmita C Sarma-DESlDOC

Bulletin of Information Technology, January & March2000

[2]Using Images Pattern and Neural Networks for

Coating Quality Assessment-L.M. Chang and Y.A. Abdel Razig

[3] Freeman, James A. & David, M. Neural networks:

Algorithms, applications, and programming techniques.

Skapura Addison-Wesley Longman, Inc.

[4] De La Blanca, N. (1992). Pattern recognition and

Image Analysis. World Scientific.

[5] Wikipedia Online Encyclopedia

http://en.wikipedia.org/wiki/Neural_network

http://en.wikipedia.org/wiki/Back-propagation





art neural nw

Documents