Abstract Selecting a low dimensional feature subspace from t housands of features is a key phenomenon for optima l classification. Linear Discriminant Analysis (LDA) is a basic well recognized supervised classifier that is effectively employed for classification. However, two problems arise in intra class during discriminant analysis. Firstly, in training phase the number of samples in intra class is smaller than th e dimensionality of the sample which makes LDA unst able. The other is high computational cost due to redundant and irrelevant data points in intra class. An Adaptive Margin Fisher’s Criterion Linear Discriminant Analysis (AMFC/LDA) is proposed that a ddresses these issues and overcomes the limitations of intra class problems. Small Sample Size (SSS) problem is resolv ed through modified Maximum Margin Criterion (MMC), which is a form of customized LDA and convex hull. Inter class is d efined using LDA while intra class is formulated using quick hull respectively. Similarly, computational cost is redu ced by reformulating within class scatter matrix th rough minimum Redundancy Maximum Relevance (mRMR) algorithm while preserving discriminant information. The proposed algorithm reveals encouraging performance. Finally, a compari son is made with existing approaches.

References

[1] Auffarth B., Lopez M., and Cerquides J., Hopfield Networks in Relevance and Redundancy Feature Selection Applied to Classification of Biomedical High2Resolution Micro2CT Images, in Proceedings of Spanish MEC Project , Spain, pp. 16231, 2008.

[2] Belhumeur P., Hespanha J., and Kriegman D., Eigenfaces vs Fisher Faces: Recognition using Class Specific Linear Projection, IEEE Transactions Pattern Analysis and Machine Intelligence, vol. 19, no. 7, pp. 7112720, 1997.

[3] Bland M., Some Problems with Sample Size, Presentation at the Joint Meeting of the Dutch Pathological Society and the Pathological Society of Great Britain & Ireland , Leeds, pp. 12 6, 2008.

[4] Chen L2F., Mark H2Y., Ko M2T., Lin J2C., and Yu G2J., A New LDA2Based Face Recognition System which Can Solve the Small Sample Size Problem, Pattern Recognition , vol. 33, no. 10, pp. 171321726, 2000.

[5] Chong L., Wanquan L., and Senjian A., Face Recognition with Only One Training Sample, in Proceedings of the 25 th Chinese Control Conference , Harbin, pp. 221522219, 2006.

[6] Dai D2Q. and Yuen P2C., Face Recognition by Regularized Discriminant Analysis, IEEE Transactions on Systems, Man , and Cybernetics , Part B: Cybernetics , vol. 37, no. 4, pp. 10802 1085, 2007.

[7] Gu Q. and Zhou J., Local Relevance Weighted Maximum Margin Criterion for Text Classification, in Proceedings of SIAM 157 Face Recognition Using Adaptive Margin Fisher s Cri terion and Linear Discriminant International Conference on Data Mining , vol. 21, pp. 113521146, 2009.

[8] Hariharan B., Manor L., Vishwanathan N., and Varma M., Large Scale Max2Margin Multi2 Label Classification with Priors, in Proceedings of the 27th International Conference on Machine Learning , Haifa, Palestine, pp. 4232430, 2010.

[9] Hu H., Zhang P., and Torre D., Face Recognition using Enhanced Linear Discriminant Analysis, IET Computer Vision , vol. 4, no. 4, pp. 1952208, 2010.

[10] Hua Y. and Yang J., A Direct LDA Algorithm for High2Dimensional Data with Application to Face Recognition, Pattern Recognition , vol. 34, no. 2001, pp. 206722070, 2001.

[11] Jiang Y. and Guo P., Comparative Studies of Feature Extraction Methods with Application to Face Recognition, in Proceedings of IEEE International Conference on Systems , Man and Cybernetics , Montreal, Canada, pp. 362723632, 2007.

[12] Juwei L., Plataniotis K., and Venetsanopoulos A., Regularization Studies of Linear Discriminant Analysis in Small Sample Size Scenarios with Application to Face Recognition, Elsevier Science Direct Pattern Recognition Letters , vol. 26, no. 2, pp. 1812191, 2005.

[13] Kabir M., Jabid T., and Chae O., Local Directional Pattern Variance (LDPv): A Robust Feature Descriptor for Facial Expression Recognition, the International Arab Journal of Information Technology , vol. 9, no. 4, pp. 3822 391, 2012.

[14] Kalavdekar P., Face Recognition using Extended Fisher Face with 3D Morph Able Model, the International Journal of Computer Applications , vol. 1, no. 16, pp. 3442523, 2010.

[15] Lajevardi S. and Hussain Z., Novel Higher2 Order Local Autocorrelation2Like Feature Extraction Methodology for Facial Expression Recognition, IET Image Process , vol. 4, no. 4 pp. 1142119, 2010.

[16] Li H., Jiang T., and Zhang K., Efficient and Robust Feature Extraction by Maximum Margin Criterion, IEEE Transaction on Neural Network , vol. 17, no. 1, pp. 1572165, 2006.

[17] Li W., Ruan Q., and Wan J., Two2Dimensional Uncorrelated Linear Discriminant Analysis for Facial Expression Recognition, in Proceedings of IEEE 10th International Conference on Signal Processing , Beijing, China, vol. 34, pp. 13622 1365, 2010.

[18] Liu C., Learning the Uncorrelated, Independent, and Discriminating Color Spaces for Face Recognition, IEEE Transactions on Information Forensics and Security , vol. 3, no. 2, pp. 2132 222, 2008.

[19] Lu H., Plataniotis K., Anastasios N., and Venetsanopoulos N., Uncorrelated Multilinear Discriminant Analysis with Regularization and Aggregation for Tensor Object Recognition, IEEE Transaction on Neural Networks , vol. 20, no. 1, pp. 1032123, 2009.

[20] Pannagadatta K. and Jebara T., Maximum Relative Margin and Data2Dependent Regularization, Journal of Machine Learning Research , vol. 11, pp. 7472788, 2010.

[21] Qiu X. and Wu L., Nonparametric Maximum Margin Criterion for Face Recognition, IEEE CiteSeer , vol. 25, no. 7, pp. 177121782, 2005.

[22] Razzak M., Khan M., Alghathbar K., and Yousaf R., Face Recognition using Layered Linear Discriminant Analysis and Small Subspace, in Proceedings of the 10 th IEEE International Conference on Computer and Information Technology , Bradford, UK, pp. 140721412, 2010.

[23] Robinson P. and Clarke W., Comparison of Principal Component Analysis and Linear Discriminant Analysis for Face Recognition, in Proceedings of AFRICON, Windhoek, Namibia, pp. 126, 2007.

[24] Sharif M., Mohsin S., Javed M., and Ali M., Single Image Face Recognition using Laplacian of Gaussian and Discrete Cosine Transforms, the International Arab Journal of Information Technology , vol. 9, no. 6, pp. 5622570, 2012.

[25] Smith L., A Tutorial on Principal Components Analysis, Journal of Cognitive Neuroscience , vol. 3, no. 1, pp. 228, 2002.

[26] Tan X., Chen S., Zhou Z2H., and Liu J., Face Recognition under Occlusions and Variant Expressions with Partial Similarity, IEEE Transaction on Information Forensics and Security , vol. 4, no. 2, pp. 4212426, 2009.

[27] Tian Q., Barbero M., Gu Z., and Lee S., Image Classification by the Foley2Sammon Transform, Optical Engineering , vol. 25, no. 7, pp. 8342840, 1986.

[28] Yan J., Zhang B., Yan S., Yang Q., Li H., Chen Z., Xi W., Fan W., Ma W., and Cheng Q., IMMC: Incremental Maximum Margin Criterion, in Proceedings of the 10 th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining , New York, USA, vol. 1, pp. 7252730, 2004.

[29] Yang W., Wang J., Ren M., Yang J., Zhang L., and Liu G., Feature Extraction Based on Laplacian Bidirectional Maximum Margin Criterion, Pattern Recognition , vol. 42, no. 11, pp. 232722334, 2009.

[30] Yang X., Tek B., Beddoe G., and Slabaugh G., Feature Selection for Computer2Aided Polyp Detection using MRMR, in Proceedings of Medical Imaging, Computer/Aided Diagnosis , San Diego, USA, vol. 7624, pp. 128, 2010. 158 The International Arab Journal of Information Techn ology, Vol. 11, No. 2, March 2014

[31] Zhang X. and Tang Z., A Fast Convex Hull Algorithm for Binary Image, Informatica Journal , Guangxi , vol. 34, pp. 3692376, 2010.

[32] Zhao H. and Yuen P2C., Incremental Linear Discriminant Analysis for Face Recognition, IEEE Transactions on Systems , Man , and Cybernetics , Part B: Cybernetics , vol. 38, no. 1, pp. 2102221, 2008.

[33] Zhao W., Karishnaswamy A., Chellappa R., Swets D., and Weng J., Discriminant Analysis of Principal Components for Face Recognition, in Proceedings of the 3 th IEEE International Conference Automatic Face and Gesture Recognition , Nara, Japan, vol. 83, pp. 3362341, 1998.

[34] Zhao W. and Chellappa R., Image Based Face Recognition Issues and Methods, in Proceedings of Image Recognition and Classification , USA, pp. 3752402, 2002.

[35] Zheng W., Heteroscedastic Feature Extraction for Texture Classification, IEEE Signal Processing Letters, vol. 16, no. 9, pp. 7662769, 2009. Marryam Murtaza is a lecturer at University of Wah, Pakistan. She has more than four years teaching and research experience. Her research interests include digital image processing and software engineering. She had completed her MS degree in computer science from CIIT Wah in 2011. Muhammad Sharif is working as an assistant professor at the Department of Computer Science, COMSATS Institute of Information Technology, Pakistan. He is a PhD scholar at COMSATS Institute of Information Technology, Islamabad Campus. He has more than 16 years of experience of teaching undergraduate and graduate classes. Mudassar Raza is a lecturer at COMSATS Institute of Information Technology, Pakistan. He has more than four years of experience of teaching undergraduate classes at CIIT Wah. He has also been supervising final year projects to undergraduate students. His areas of interest are d igital image processing, and parallel and distributed computing. Jamal Hussain Shah is a research associate in Computer Science Department at COMSATS Institute of Information Technology, Pakistan. His research areas are digital image processing and networking. He has more than 3 years experience in IT2related projects, he develop ed and designed ERP systems for different organization s of Pakistan.