1 www.ssijmar.in

SHIV SHAKTI

International Journal in Multidisciplinary and

Academic Research (SSIJMAR)

Vol. 6, No. 2, April 2017 (ISSN 2278 – 5973)

Estimation of Age Group using Histogram of Oriented

gradients and Neural Network

Arpit Kumar Sharma

Mobile No. 9414342366

Impact Factor = 3.133 (Scientific Journal Impact Factor Value for 2012 by Inno Space

Scientific Journal Impact Factor)

Global Impact Factor (2013)= 0.326 (By GIF)

Indexing:

2 www.ssijmar.in

ABSTRACT

Requirement of Project: Face images are being increasingly used as additional means of

authentication in applications of high security zone. But with age progression the facial

features changes and the database needs to be updated regularly which is a tedious task.

So we need to address the issue of facial aging and come up with a mechanism that identifies

a person in spite of aging. In my project, effective age group estimation using face features

like texture and shape from human face image are proposed[2]. For better performance, the

geometric features of facial image like wrinkle geography, face angle, left to right eye

distance, eye to nose distance, eye to chin distance and eye to lip distance are calculated.

Based on the texture and shape information, age classification is done using KNN & SVM

algorithm (Best algorithm according to many research paper during my research)."

Proposed System: "In this report, few classification and feature extraction techniques used

for age group classification. In this report first we attempt to combining two type of face

features using haar features extraction (Wrinkle features and Geometrical Features) also used

viola Jones for face detection. Age estimation based on the graphical model structure is

proposed. Three popular features, PCA (Principal Component analysis), HOG and

Haarfeatures, are exploited in our work, and three different graphical model structures

considering spatial information and hidden topics are proposed and implemented. The

experimental results showed that our model performs classification techniques like SVM

(support vector machine), KNN and Neural network and the comparisons between features

extraction algorithm and classification techniques in order to obtain best output. features are

also presented and discussed. Until now, the model we proposed hasn’t been well-tuned, and

we’ll try to improve it for the future works." Research in those areas has been conducted for

more than 30 years. "Traditionally, face recognition uses for identification of documents such

as land registration, passports, driver’s licenses, and recognition of a human in a security

area. [22]

1.1 GENERAL INTRODUCTION

Most of the facial variants such as identity, expression, emotions and gender have been

widely studied, in the case of research of recognition. Automatic age estimation is one such

area that has been rarely explored by the researchers. With the evolution of a human, the

features of the face keep on changing with age. This project is providing a new combine

approach of feature selection for age group classification algorithms. Further this process is

further classified into three main stages: first one is Pre-processing, second is Feature

Extraction (Haar feature extraction), and the last stage is classification. For feature extraction

3 www.ssijmar.in

phase we used two techniques 1) Wrinkle features and 2) Geometrical features for the face

pattern recognition [11]. We know that Wrinkle features are well enough to differentiate

between the adult and senior, Geometrical features is good to create difference between child

and adult/senior. That is why we used a combine technique of wrinkle and geometrical so that

they can solve each other problems and provide the best output. These two approaches are

defined below:

1.1.1) Geometrical features

This include face angle, left eye to right eye distance, eyeball, eye to nose distance, eye to

chin distance and eye to lip distance that is further calculated by making use of the best

feature selection algorithm

1.1.2) Wrinkle features

Age classification based on the texture and shape information is done by using suggested

hybrid algorithm which includes Fuzzy logic and Neural Network. Depending on a number of

groups age ranges are then classified dynamically using hybridization algorithms

individually.

In our research, most facial features like identity, expression, emotions and gender has

been majorly focused. Automatic age estimation is one area that has been rarely

explored till date. As age increases, the feature of the face keeps on changing. This

project provides a comparison study of classification techniques (SVM, KNN

algorithm) and these falls under the category of the best classification algorithms.

Entire process is divided into three stages: Pre-processing, Feature Extraction (Haar

feature extraction) [75], classification (above mentioned algorithm). Machine learning

phase uses different classification algorithm approach in order to provide the best

solution for pattern recognition. That is why we are using one hybrid technique of

wrinkle and geometrical by which they can solve each other problems and provide the

best results. Here we make use of two important features of the face which are

responsible for age identification. Personal identification and verification has evolved

as an active area of research these days. As biometric characteristics of the individual

are unique person to person, biometric authentication techniques have a great

advantage over traditional authentication techniques. Recognition of face is one of the

widely used biometric methods which are used to identify individuals by their face

features. Face, voice, fingerprint, iris, ear, retina are the most commonly used for

authentication purpose. Research in those areas has been conducted for more than 30

4 www.ssijmar.in

years. Face recognition is beneficial for identification of documents such as for land

registration, passports, driver’s licenses, and recognition of a human in a security area

[84]. Face images are highly used as additional means of authentication in

applications having high security zone. But with increase in age the facial features

also keep on changes and the database needs to be updated regularly which is one

very tedious task. Hence we need to address this issue of facial aging and come up

with a solution that identifies a person without any age limits. In this thesis, effective

age group estimation using face features like texture and shape from human face

image is proposed. For getting efficient results, the geometric features of the facial

image like wrinkle geography, face angle, left to right eye distance, eye to nose

distance, eye to chin distance and eye to lip distance are calculated [90]. Based on the

texture and shape information, age classification is done by making use of

classification algorithms

2. EXISTING SYSTEM

As we already discussed in literature survey age group classification is one of the

research topics from last few years. Many research already done research on the age

group classification with different algorithms (Surf algorithm, PCA and LDA etc.)

and different classification techniques were used. In the age group classification the

most difficult part is to identify the different pattern of the faces[78]. Many authors

have tried and failed. As per our research many researchers failed to observe the exact

pattern of different age group. Pattern recognition, having two critical task one

features extraction and another is classification. Researcher tried to work hard on the

machine learning algorithm and many of them is ignore features extraction

improvisation. As the result they cannot obtain good output, but all the literature

author work excellent on classification algorithm which include (Neural network,

KNN, SVM and Fuzzy logic)[65].

5 www.ssijmar.in

Block diagram of proposed work

3.3 FLOW CHART OF AGE CLASSIFICATION

The brief description of each block is described below:

IMAGE DATA

PREPROCESS USING

GEOMETRICAL AND WRINKLE

FEATURE

CLASSIFICATION USING

KNN

CLASSIFICATION PERFOM

ANCE

Image aquisition

• Dataset from THE OPEN UNIVERSITY, ISRAEL

Data preprocess

• Face detect viola jone apply

Find Parts• Find face parts

Data prepration

• Find the face features of the images(Haar feature, PCA and HOG)

Classification

• Apply classification(KNN, SVM and Neural Network.

Performance

• Analysis machine learning perfomance .

6 www.ssijmar.in

FLOW CHART OF PROPOSED SYSTEM

`

Start

Preprocess and feature extraction

Insert 3class AGE

classification

dataset

Detect Face

Resize image

Performance

Apply

Classification

Exit

Face

Detectio

n

Data Divide

Yes

No NOT A

VALID

DATA

7 www.ssijmar.in

2.1 FEATURE EXTRACTION

One of the main key issue of any characterization frameworks is to locate an arrangement

of reliable features as the basis for classification. In general these features cam be

categorized into two categories. These are wrinkle features and geometric features. Let us

discuss each one of them in detail[59].

2.2 WRINKLE FEATURES

One of the most important property of wrinkle features is that it determines the age of a

person. Estimation of feature F5 can be done as follows :

F5= (sum of pixels in forehead region / number of pixels in forehead region) + (sum of

pixels in left eyelid region / number of pixels in left eyelid region) + (sum of pixels in

right eyelid region / number of pixels in right eyelid region) + (sum of pixels in left eye

corner region / number of pixels in left eye corner region) + (sum of pixels in right eye

corner region / number of pixels in right eye corner region).

F5 can be estimated by making use of the grid features of face image that is completely

dependent on the wrinkle geography in face image.

For the estimation of F5 features, a few steps have to be followed as discussed below:

As the age keeps on increasing, wrinkles on face turn out to be clearer. Aged individuals

regularly have clear wrinkles on the face in the following areas as mentioned below :

a) The forehead has horizontal furrows.

b) The eye corners have crow’s feet.

c) The cheeks have clear cheekbones, sickle molded pouches, and profound lines

between the cheeks and the upper lips.

Since there are evident changes in wrinkle intensities and even some form clear lines,

thus in this Project we make use of Sobel edge magnitudes, approximating gradient

magnitudes in order to judge the level of wrinkles[15]. The Sobel edge magnitude is

larger, if the pixel belongs to wrinkles. The reason behind the larger magnitude is that the

difference of gray levels is self-evident. From this perspective, a pixel is named as a

wrinkle pixel if its sobel edge size is bigger than some limit. Figure 7 (a) and (c)

demonstrate a youthful grown up and an old grown up[69].

8 www.ssijmar.in

2.3 GEOMETRICAL FEATURES

As indicated by the investigations of facial representation and emotional cosmetics, there

occurs a lot of change in the facial features as the age keeps on increasing. In this phase,

global features in combination with the grid features are extracted from the face images.

The global features include the distance between two eye balls, chin to eye, nose tip to

eye and eye to lip[76].

By making use of four distance values, there occurs calculation of four features namely

F1, F2, F3 and F4 as mentioned below:

F1 = (distance from left to right eye ball) / (distance from eye to nose).

F2 = (distance from left to right eye ball) / (distance from eye to lip).

F3 = (distance from eye to nose) / (distance from eye to chin).

F4 = (distance from eye to nose) / (distance from eye to lip).

It is clear that new born babies have a number of wrinkles on their faces. The head bone

structure in new born ones is not fully grown. Moreover the ration of primary features is

highly different from those in other life spans. Hence we can conclude that it is more

reliable to use geometric features as compared to wrinkle features when it is to be judged

that whether an image is a baby or not[82].

In case of infants, the head is near a circle. The distance between two eyes is almost equal

to the distance from eyes to mouth. As the head bone grows, the head becomes oval

shaped and accordingly there occurs a sudden increase in the distance from the eyes to the

mouth. Above and beyond the ratio between baby’s eyes and noses is equal to the

distance between noses and mouths which in turn are almost equal to one while as in case

of adults it is larger than 1[88].

9 www.ssijmar.in

3. CLASSIFICATION (KNN ALGORITHMS)

3.3.1 KNN Classification: The k-nearest neighbor algorithm is a classification algorithm

which classifies an object on the basis of where the majority of the neighbor belongs to

[76]. To choose the number of neighbors is optional and it depends on the users. If k is

equal to 1 then it is classified [10] as a class of neighbor is nearest. Normally the object is

classified on the basis of labels of its k nearest neighbors by finding out the majority vote.

If k is 1, the object is classified as the class of the object which is nearest to it. When there

are two classes, it is considered that k must be an odd integer. However, there can still be

times when k is an odd integer while performing multiclass classification. After

converting each image to a vector image of fixed-length having real numbers, we will

then use the most common distance function for KNN that is Euclidean distance [85].

Fig 3.8 :KNN classification. At the query point of the circle depending on the k value of 1,

5, or 10, the query point can be a rectangle at (a), a diamond at (b), and a triangle at (c).

The KNN is classifies an object where the majority of the neighbor belongs to. The

choice of the number of neighbors is discretionary and up to the choice of the users. If k

is 1 then it is classified [10] whichever class of neighbor is nearest[13].

result = knnclassify(Sample, Training, Group, k)

3.2 Histogram of Oriented Gradients(HoG):

The next step is to extract the features of the hand gesture. This system uses the HoG

descriptor (Histogram oriented gradient) to present the hand shape. HoG descriptor counts

the number of times a gradient orientation occurs in a localized are of the image[22]. It

uses a histogram of intensity gradient to depict the shape of the object. This technique is

resilient under change of shadow and illumination. Due to this, it's a popular method for

hand gesture detection[43].

10 www.ssijmar.in

The implementation method of the HoG algorithm descriptor is given as follows. Firstly,

the cells are divided into smallest possible regions of an image. These regions are called

cells. For each of these cells, a histogram of of gradient orientations or edge orientations

is computed. Each cell is separated and discreted into corresponding angular bins in

accordance with its gradient orientation. The weighted gradient of each cell is contributed

to its respective angular bin. The adjacent cell with same gradient orientation are grouped

together and these spatial regions are known as blocks[57]. These groupings into blocks

is the basis for histograms' normalization. The normalized group represents the block

histogram which in turn represent the descriptor [21].

3.3) Principal Component Analysis(PCA)

PCA is one of the best available statistical methods available that is used for image

compression and gesture recognition. The basic ideology behind the PCA algorithm is the

reduction of the dimensionality of an image and also maintaining maximum variance. The

features which remain then are the ones relevant for recognition[59].

Whenever there is 2-dimensional data, then due to the presence of more than 2 variables,

the visualization of of the relationships becomes complex. PCA reduces the

dimensionality of the data such that the two actual variables are reduced to less number of

new one dimensional variables which are called Principal Components. This is done by

using a single variable for a group of variables. The principal components are a linear

representation of the actual variables[64].

These principal components can also be represented in the form of vectors called Eigen

Vectors. The Eigen Vectors collectively create a feature space known as Eigen space

which is calculated by the eigen vectors of a co-variance matrix derived from a hand

gesture set. Each input gesture image corresponds to eigen vectors which represents the

feature vector of the image[79].

3.4 SVM (Support Vector Machine)

SVM Classification: A support vector machine (SVM) is a non probabilistic linear binary

classifier, which can analyze input data and predict which of the two classes it belong to. It

works by building a hyper plane separating the two classes which is of higher dimension. A

11 www.ssijmar.in

good separation is obtained by a hyper plane that is very far from any data point of each class

[11], since further the separation of the data, better the performance[64].

Fig 3.6.4: Shows the formation of hyperplane and also the how the image is classified

between red and blue using SVM.

Fig. 3.6.4(B) Overall flow diagram of project

12 www.ssijmar.in

CONCLUSION

This thesis thoroughly explains a novel method for the age group classification. Proposed

technique based on wrinkle and geometrical features provides a robust method that identifies

the age group of individuals from a set of different images capturing various aged faces.

From these images features are then extracted such as distances between various face

elements, analysis of wrinkle geography and then calculation are performed for finding out

face angles. The results are then compared at the end to find the best way to calculate age

ranges for the face images present in the database. Based on the observed results, images are

further classified into 3 groups on the basis of SVM and KNN algorithm. It is normally

observed that wrinkle geography feature i.e., F5 provides better results to predict human age

range in comparison to other features. Hence we can conclude that wrinkle geography

analysis is one good approach to estimate human age range for an individual. For better eye

and eyeball detection, images should be captured without spectacles. Viola Jone algorithm

focuses on the front face that is why the image needs to be a straight frontal face. As we are

working on the individual face age group identification so for that purpose image should

contain single human face only. This thesis has shown results with 76% accuracy for two age

group, 64% accuracy for three age group. As the numbers of group are increased for

classification the accuracy of classification is decreased. There is a strong possibility for

further extension of the work which includes extracting more feature points that can improve

accuracy of age group classification. By introducing more features the age range can also be

further enhanced.

FUTURE SCOPE

The future work is to add more category in the field of age group recognition classes to the

given system. Also since the proposed system is limited to classify only for front images, so

modeling 3-D face using various cameras to increase the efficiency of the proposed facial age

recognition system can be used for future work. We can also implement face age detection by

using fuzzy logic and genetic algorithm.

13 www.ssijmar.in

REFERENCE

BIBILOGRAHY

Books:

1. Jang, J.S.R., Sun, C.T. and Mizutani, E. (1997) Neuro Fuzzy and Soft Computing: A

Computational Approach to Learning and Machine Intelligence. United States of

America: Prentice-Hall.

2. Mitra, S.K. (2006) Digital Signal Processing: A Computer-Based Approach. New

York:McGraw-Hill.

3. proaktis, J G. and Manolakis D.G. (2007) Digital Signal Processing:

PrinciplesAlgorithms and Application, New Jersey: Prentice-Hall.Tournals and

Conferences

REFERENCE

[1] A.Lanitis and C.J.Taylor, “Towards Automatic Face Identification Robust to Ageing

Variation”, IEEE Trans. on Pattern Analysis and Machine Intelligence, vol.24, no.24, p.442-

455, 2002.

[2] A.Lanitis, C.Draganova, and C.Christodoulou, “Comparing different classifiers for

automatic age estimation”, IEEE Trans.Syst.Man, Cybern.B, Cybern, vol34, no.1, pp.621-

628, Feb.2004.

[3] V. Blanz and T. Vetter, “Face recognition based on fitting a 3D morphable model”, IEEE

Transactions on Pattern Analysis and Machine Intelligence, 25(9):1063 –1074, September

2003.

[4] R. Kimmel A. M. Bronstein, M. M. Bronstein, “Three-dimensional face recognition”, Intl.

Journal of Computer Vision, 64(1):5–30, August 2005.

14 www.ssijmar.in

[5] N.Ramanathan and R. Chellappa, “Face verification across age progression”, in Proc.

IEEE Conf. Computer Vision and Pattern Recognition, San Diego, CA, 2005, vol.2, pp.462-

469.

[6] N.Ramanathan and R. Chellappa, “Modeling Age Progression in young faces”, in Proc.

IEEE Conf. Computer Vision and Pattern Recognition (CVPR), vol.1, pp.387-394, 2006.

[7] G. Guo, A. Jain, W. Ma, and H. Zhang, “Learning similarity measure for natural image

retrieval with relevance feedback,” IEEE Trans.Neural Netw., vol. 13, no. 4, pp. 811–820,

Apr. 2002.

[8] G. Guo and S. Li, “Content-based audio classification and retrieval by support vector

machines,” IEEE Trans. Neural Netw., vol. 14, no. 1,pp. 209–215, Jan. 2003.

[9] G. Guo, S. Li, and K. Chan, “Support vector machines for face recognition,” Image Vis.

Comput., vol. 19, no. 9-10, pp. 631–638, 2001.

[10] G. Guo, S. Li, and K. Chan, “Face recognition by support vector machines,” in Proc.

IEEE Conf. FG, 2000, pp. 196–201.

[11] [X. F. He and P. Niyogi, “Locality preserving projections,” presented at the NIPS, 2003.

[12] T. Joachims, “Making large-scale SVM learning practical,” in Advances in Kernel

Methods—Support Vector Learning, B. Schölkopf,C. Burges, and A. Smola, Eds. Cambridge,

MA: MIT Press, 1999.

[13] Y. Kwon and N. Lobo, “Age classification from facial images,” Comput. Vis. Image

Understand., vol. 74, no. 1, pp. 1–21, 1999.

[14] A. Lanitis, C. Taylor, and T. Cootes, “Toward automatic simulation of aging effects on

face images,” IEEE Trans. Pattern Anal. Mach. Intell.,vol. 24, no. 4, pp. 442–455, Apr. 2002.

[15] Abed, M. A., Ismail, A. N., &Hazi, Z. M. (2010). Pattern recognition using genetic

algorithm. International Journal of Computer and Electrical Engineering,2(3), 1793-8163.

15 www.ssijmar.in

[16] [Gundogan, K. K., Alatas, B., &Karci, A. (2004). Mining classification rules by using

genetic algorithms with non-random initial population and uniform operator. Turk J Elec

Engin, 12(1)

.

[17] Maulik, U., &Bandyopadhyay, S. (2000). Genetic algorithm-based clustering

technique. Pattern recognition, 33(9), 1455-1465.

[18] Kwan, H. K., &Cai, Y. (1994). A fuzzy neural network and its application to pattern

recognition. Fuzzy Systems, IEEE Transactions on, 2(3), 185-193.

[19] Karnik, N. N., & Mendel, J. M. (1998, May). Introduction to type-2 fuzzy logic

systems.In Fuzzy Systems Proceedings, 1998. IEEE World Congress on Computational

Intelligence., The 1998 IEEE International Conference on (Vol. 2, pp. 915-920). IEEE.

[20] Castillo, O., &Melin, P. (2008). Type-2 Fuzzy Logic: Theory and Applications.

Heidelberg, Germany: Springer-Verlag.

[21] Melin, P., & Castillo, O. (2005). Hybrid intelligent systems for pattern recognition using

soft computing: an evolutionary approach for neural networks and fuzzy systems (Vol. 172).

Springer Science & Business Media.

[22] Tizhoosh, H. R. (2005). Image thresholding using type II fuzzy sets. Pattern

recognition, 38(12), 2363-2372.

Abiyev, R. H., &Kaynak, O. (2010).Type 2 fuzzy neural structure for identification and

control of time-varying plants. Industrial Electronics, IEEE Transactions on, 57(12), 4147-

4159.

[23] Mitchell, H. B. (2005). Pattern recognition using type-II fuzzy sets. Information

Sciences, 170(2), 409-418.

Wang, C. H., Cheng, C. S., & Lee, T. T. (2004).Dynamical optimal training for interval type-

2 fuzzy neural network (T2FNN). Systems, Man, and Cybernetics, Part B: Cybernetics, IEEE

Transactions on, 34(3), 1462-1477.

[24] Ozkan, I. (2004, July). Entropy assessment for type-2 fuzziness.In Fuzzy Systems,

2004.Proceedings.2004 IEEE International Conference on (Vol. 2, pp. 1111-1115).IEEE.

[25] Zeng, J., & Liu, Z. Q. (2004, August). Type-2 fuzzy hidden Markov models to phoneme

recognition.In Pattern Recognition, 2004.ICPR 2004.Proceedings of the 17th International

Conference on (Vol. 1, pp. 192-195).IEEE.

[26] Zeng, J., & Liu, Z. Q. (2004, August). Type-2 fuzzy hidden Markov models to phoneme

recognition.In Pattern Recognition, 2004.ICPR 2004.Proceedings of the 17th International

Conference on (Vol. 1, pp. 192-195).IEEE.

[27] Rhee, F. C. H., & Choi, B. I. (2007, July). Interval type-2 fuzzy membership function

design and its application to radial basis function neural networks. InFuzzy Systems

Conference, 2007.FUZZ-IEEE 2007. IEEE International (pp. 1-6). IEEE.

16 www.ssijmar.in

[28] Park, K. J., Oh, S. K., &Pedrycz, W. (2009, August). Design of interval type-2 fuzzy

neural networks and their optimization using real-coded genetic algorithms.In Fuzzy Systems,

2009.FUZZ-IEEE 2009. IEEE International Conference on (pp. 2013-2018). IEEE.

[29] Own, C. M. On the switch between Type-2 fuzzy sets and intuitionistic fuzzy sets. In 12

th conference on Artificial intelligence & applications (TAAI 2007).

[30] Mendoza, O., Melín, P., & Castillo, O. (2009). Interval type-2 fuzzy logic and modular

neural networks for face recognition applications. Applied Soft Computing, 9(4), 1377-1387.

[31] Own, C. M. (2009). Switching between type-2 fuzzy sets and intuitionistic fuzzy sets: an

application in medical diagnosis. Applied Intelligence, 31(3), 283-291.

[32] Bandyopadhyay, S., Murthy, C. A., & Pal, S. K. (1998). Pattern classification using

genetic algorithms: Determination of H. Pattern Recognition Letters,19(13), 1171-1181.

[33] Alfehaid, W. M., Khan, A. I., & Amin, A. H. M. (2012, December). A combined pattern

recognition scheme with genetic algorithms for robot guidance using Wireless Sensor

Networks. In Control Automation Robotics & Vision (ICARCV), 2012 12th International

Conference on (pp. 759-764). IEEE.

[34] Melin, P. (2010, August). Interval type-2 fuzzy logic applications in image processing

and pattern recognition.In 2010 IEEE International Conference on Granular Computing (pp.

728-731).IEEE.

[35] Zhenjiang, M., &Baozong, Y. (1993). A new analytical method for AM neural networks

and its application to pattern recognition. In Neural Networks, 1993., IEEE International

Conference on (pp. 1570-1575). IEEE.

Lin, C. J., Wang, J. G., & Lee, C. Y. (2009). Pattern recognition using neural-fuzzy networks

based on improved particle swam optimization. Expert Systems with Applications, 36(3),

5402-5410.

[36] Zhao, W., Chellappa, R., Phillips, P. J., & Rosenfeld, A. (2003). Face recognition: A

literature survey. ACM computing surveys (CSUR), 35(4), 399-458.

[37] Darwish, S. M., & Mohammed, A. H. (2014). Interval Type-2 Fuzzy Logic to the

Treatment of Uncertainty in 2D Face Recognition Systems. International Journal of Machine

Learning and Computing, 4(1), 24-30.

Begum, S. A., & Devi, O. M. (2011).Fuzzy algorithms for pattern recognition in medical

diagnosis. Assam University Journal of Science and Technology, 7(2), 1-12.

[38] Al-Allaf, O. N. (2014). Review of face detection systems based artificial neural

networks algorithms. arXiv preprint arXiv:1404.1292.

[39] Schmalzel, J. L., & Johnson, J. F. (1993, May). Pattern recognition based on fuzzy logic.

In Instrumentation and Measurement Technology Conference, 1993.IMTC/93. Conference

Record., IEEE (pp. 681-684). IEEE.

[40] Kwan, H. K., &Cai, Y. (1994). A fuzzy neural network and its application to pattern

recognition. Fuzzy Systems, IEEE Transactions on, 2(3), 185-193.

17 www.ssijmar.in

[41] Mahdy, Y. B., & El-Melegy, M. T. (1996, October). Encoding patterns for efficient

classification by nearest neighbor classifiers and neural networks with application to

handwritten Hindi numeral recognition. In Signal Processing, 1996., 3rd International

Conference on (Vol. 2, pp. 1362-1365). IEEE.

[42] Niño-de-Rivera, L., Nakano-Miyatake, M., Sánchez, J. C., Perez-Meana, H., & Sanchez-

Sinencio, E. (1995, August). VLSI implementation of an extended Hamming neural network

for non-binary pattern recognition. In Circuits and Systems, 1995., Proceedings., Proceedings

of the 38th Midwest Symposium on (Vol. 2, pp. 973-977). IEEE.

Kandel, A., Zhang, Y. Q., &Bunke, H. (1997, July).A genetic fuzzy neural network for

pattern recognition. In Fuzzy Systems, 1997., Proceedings of the Sixth IEEE International

Conference on (Vol. 1, pp. 75-78). IEEE.

[43] Melin, P. (2010, August). Interval type-2 fuzzy logic applications in image processing

and pattern recognition.In 2010 IEEE International Conference on Granular Computing (pp.

728-731).IEEE.

[44] Budiwati, S. D., Haryatno, J., & Dharma, E. M. (2011, July). Japanese character (Kana)

pattern recognition application using neural network. InElectrical Engineering and

Informatics (ICEEI), 2011 International Conference on (pp. 1-6). IEEE.

[45] Alfehaid, W. M., Khan, A. I., & Amin, A. H. M. (2012, December). A combined pattern

recognition scheme with genetic algorithms for robot guidance using Wireless Sensor

Networks. In Control Automation Robotics & Vision (ICARCV), 2012 12th International

Conference on (pp. 759-764). IEEE.

[46] Jun, T. (2012, October). Genetic optimization of BP neural network in the Application of

Suspicious Financial Transactions Pattern Recognition.InManagement of e-Commerce and e-

Government (ICMeCG), 2012 International Conference on (pp. 280-284). IEEE.

[47] Silva, W. L. S., & de Oliveira Serra, G. L. (2012, November). Proposal of an Intelligent

Speech Recognition System.In Intelligent Systems (GCIS), 2012 Third Global Congress

on (pp. 356-359).IEEE.

[48] Alhusain, S., Coupland, S., John, R., & Kavanagh, M. (2013, November). Design pattern

recognition by using adaptive neuro fuzzy inference system. InTools with Artificial

Intelligence (ICTAI), 2013 IEEE 25th International Conference on (pp. 581-587). IEEE.

[49] Silva, W. L. S., & de Oliveira Serra, G. L. (2014, July). A novel intelligent system for

speech recognition.In Neural Networks (IJCNN), 2014 International Joint Conference on (pp.

3599-3604).IEEE.

[50] Silva, W. L. S., & de Oliveira Serra, G. L. (2014, July). A novel intelligent system for

speech recognition.In Neural Networks (IJCNN), 2014 International Joint Conference on (pp.

3599-3604).IEEE.

[51] Shelke, S., &Apte, S. (2015, January). A fuzzy based classification scheme for

unconstrained handwritten Devanagari character recognition.InCommunication, Information

& Computing Technology (ICCICT), 2015 International Conference on (pp. 1-6). IEEE.

18 www.ssijmar.in

[52] Chengjian, S., Zhu, S., & Shi, Z. (2015, May). Image annotation via deep neural

network.In Machine Vision Applications (MVA), 2015 14th IAPR International Conference

on (pp. 518-521).IEEE.

[53] Ren, J., Jiang, X., & Yuan, J. QUANTIZED FUZZY LBP FOR FACE RECOGNITION.

[54] Weng, C., Yu, D., Seltzer, M. L., &Droppo, J. (2015). Deep neural networks for single-

channel multi-talker speech recognition. IEEE/ACM Transactions on Audio, Speech and

Language Processing (TASLP), 23(10), 1670-1679.

[55] Taylor, C. E., & Meldrum, D. R. (2000). Algorithm design, user interface, and

optimization procedure for a fuzzy logic ramp metering algorithm: A training manual for

freeway operations engineers (No. WA-RD 481.1,).Washington State Department of

Transportation.

[56] Kóczy, L. T., Vámos, T., &Biró, G. (1999). Fuzzy signatures. Proceedings of

EUROFUSESIC, 99, 210-217

Gedeon, T. D., Kóczy, L. T., Wong, K. W., & Liu, P. (2001). Effective fuzzy systems for

complex structured data. In Proceedings of IASTED International Conference Control and

Applications (pp. 184-187).

[57] Vámos, T., Kóczy, L. T., &Biró, G. (2001, July). Fuzzy signatures in data

mining.In IFSA World Congress and 20th NAFIPS International Conference, 2001. Joint

9th (pp. 2842-2846). IEEE.

[58] Zhu, D., Mendis, B. S. U., Gedeon, T., Asthana, A., &Goecke, R. (2009). A hybrid

fuzzy approach for human eye gaze pattern recognition. In Advances in Neuro-Information

Processing (pp. 655-662).Springer Berlin Heidelberg.

[59] Mendis, B. S. U., &Gedeon, T. D. (2008, June). A comparison: Fuzzy signatures and

choquet integral. In Fuzzy Systems, 2008. FUZZ-IEEE 2008.(IEEE World Congress on

Computational Intelligence). IEEE International Conference on (pp. 1464-1471). IEEE.

[60] Zhu, D., &Gedeon, T. (2008). Fuzzy logic for cooperative robot communication.

In Agent and Multi-Agent Systems: Technologies and Applications (pp. 401-410). Springer

Berlin Heidelberg.

[61] Kulkarni, G. H., &Waingankar, P. G. (2007, August). Fuzzy logic based traffic light

controller. In Industrial and Information Systems, 2007.ICIIS 2007. International Conference

on (pp. 107-110). IEEE.

[62] Ambre, M. S. (2004). A Design Methodology for the Implementation of Fuzzy Logic

Traffic Controller using Field Programmable Gate Arrays.

[63] Khalid, M., Liang, S. C., &Yusof, R. (2004, July). Control of a complex traffic junction

using fuzzy inference.In Control Conference, 2004.5th Asian (Vol. 3, pp. 1544-1551).IEEE.

[64] Niittymäki, J. (2001). Installation and experiences of field testing a fuzzy signal

controller. European journal of operational research, 131(2), 273-281.

19 www.ssijmar.in

[65] Khalid, M. B. Module 1: Basics of Fuzzy Sets and Fuzzy Relations.Unpublished Course

Notes, UniversitiTeknologi Malaysia.

[66] Islam, M. S., Bhuyan, M. S., Azim, M. A., Teng, L. K., & Othman, M. (2006,

September). Hardware Implementation of Traffic Controller using Fuzzy Expert System.

In Evolving Fuzzy Systems, 2006 International Symposium on (pp. 325-330).IEEE.

[67] Azim, M. A., & Huda, M. N. Fuzzy Traffic Control System.

[68] Walker, M. (2001). Introduction to genetic programming. Tech. Np: University of

Montana.

[69] Koza, J. R. (1992). Genetic programming: on the programming of computers by means

of natural selection (Vol. 1). MIT press.

[70] Duda, R. O., & Hart, P. E. (1973). Pattern recognition and scene analysis.

[71] Tou, J. T., & Gonzales, R. C. (1974). Pattern Recognition Principles.Addison-

Weley. Reading, MA.

[72] Fukunaga, K. (2013). Introduction to statistical pattern recognition.Academic press.

[73] Konar, A. (2006). Computational intelligence: principles, techniques and applications.

Springer Science & Business Media.

[74] Haykin, S., & Lippmann, R. (1994). Neural Networks, A Comprehensive

Foundation. International Journal of Neural Systems, 5(4), 363-364.

[75] Klir, G., & Yuan, B. (1995). Fuzzy sets and fuzzy logic (Vol. 4). New Jersey: Prentice

Hall.

[76] Golberg, D. E. (1989). Genetic algorithms in search, optimization, and machine

learning. Addionwesley, 1989.

[77] Burke, E. K., & Kendall, G. (2005). Search methodologies. Springer Science+ Business

Media, Incorporated.

Jain, A. K., Mao, J., &Mohiuddin, K. M. (1996). Artificial neural networks: A

tutorial. Computer, (3), 31-44.

Kohonen, T. (1995).Self-Organing Maps-Springer Series in Information Sciences, vol.

30. Berlim: Springer Verlag.

[78] Shah, S. A. A., ulAsar, A., &Shaukat, S. F. (2009). Neural network solution for secure

interactive voice response. World Applied Sciences Journal, 6(9), 1264-1269.

[79] Guo, X., Liang, X., & Li, X. (2007, August). A stock pattern recognition algorithm

based on neural networks. In Natural Computation, 2007.ICNC 2007.Third International

Conference on (Vol. 2, pp. 518-522).IEEE.

20 www.ssijmar.in

[80] Dai, W., & Wang, P. (2007, August). Application of pattern recognition and artificial

neural network to load forecasting in electric power system.In Natural Computation,

2007.ICNC 2007.Third International Conference on (Vol. 1, pp. 381-385).IEEE.

[81] He, L., Hou, W., Zhen, X., & Peng, C. (2006, October). Recognition of ECG patterns

using artificial neural network. In Intelligent Systems Design and Applications,

2006.ISDA'06.Sixth International Conference on (Vol. 2, pp. 477-481).IEEE.

[82] Mani, N., & Srinivasan, B. (1997, October). Application of artificial neural network

model for optical character recognition.In Systems, Man, and Cybernetics, 1997.

Computational Cybernetics and Simulation., 1997 IEEE International Conference on (Vol. 3,

pp. 2517-2520). IEEE.

[83] Han, Y. S., Min, S. S., Choi, W. H., & Cho, K. B. (1992, November). A learning pattern

recognition system using neural network for diagnosis and monitoring of aging of electrical

motor.In Industrial Electronics, Control, Instrumentation, and Automation, 1992. Power

Electronics and Motion Control., Proceedings of the 1992 International Conference on (pp.

1074-1077). IEEE.

[84] Ozdemir, E., &Gunduz-Demir, C. (2013). A hybrid classification model for digital

pathology using structural and statistical pattern recognition. Medical Imaging, IEEE

Transactions on, 32(2), 474-483.

[85] Ma, H., Chan, J. C., Saha, T. K., &Ekanayake, C. (2013). Pattern recognition techniques

and their applications for automatic classification of artificial partial discharge

sources. Dielectrics and Electrical Insulation, IEEE Transactions on,20(2), 468-478.

[86] Phadik, S., &Sil, J. (2008, December). Rice disease identification using pattern

recognition techniques.In Computer and Information Technology, 2008.ICCIT 2008. 11th

International Conference on (pp. 420-423). IEEE.

[87] Phadik, S., &Sil, J. (2008, December). Rice disease identification using pattern

recognition techniques.In Computer and Information Technology, 2008.ICCIT 2008. 11th

International Conference on (pp. 420-423). IEEE.

[88] Chandra, M., &Langari, R. (2006, July). Gearbox degradation identification using

pattern recognition techniques.In Fuzzy Systems, 2006 IEEE International Conference on (pp.

1520-1526).IEEE.

[89] Cao, L. (2002, November). Circuit power estimation using pattern recognition

techniques. In Proceedings of the 2002 IEEE/ACM international conference on Computer-

aided design (pp. 412-417). ACM.

[90] Liu, J., & Zhou, P. (2013). A novel myoelectric pattern recognition strategy for hand

function restoration after incomplete cervical spinal cord injury. Neural Systems and

Rehabilitation Engineering, IEEE Transactions on, 21(1), 96-103.

[91] Srihari, S. N., Cha, S. H., & Lee, S. (2001). Establishing handwriting individuality using

pattern recognition techniques.In Document Analysis and Recognition, 2001.Proceedings.

Sixth International Conference on (pp. 1195-1204). IEEE.

21 www.ssijmar.in

[92] Tou, J. T., & Gonzalez, R. C. (1974). Pattern recognition principles. Pattern Recognition

in Physics, 1.