This paper presents two techniques, local feature

\r\nextraction using image spectrum and low frequency spectrum

\r\nmodelling using GMM to capture the underlying statistical

\r\ninformation to improve the performance of face recognition

\r\nsystem. Local spectrum features are extracted using overlap sub

\r\nblock window that are mapped on the face image. For each of this

\r\nblock, spatial domain is transformed to frequency domain using

\r\nDFT. A low frequency coefficient is preserved by discarding high

\r\nfrequency coefficients by applying rectangular mask on the

\r\nspectrum of the facial image. Low frequency information is non-

\r\nGaussian in the feature space and by using combination of several

\r\nGaussian functions that has different statistical properties, the best

\r\nfeature representation can be modelled using probability density

\r\nfunction. The recognition process is performed using maximum

\r\nlikelihood value computed using pre-calculated GMM components.

\r\nThe method is tested using FERET datasets and is able to achieved

\r\n92% recognition rates.<\/p>\r\n","references":"[1] G. Shakhnarovich, V. Moghaddam, Face recognition in subspaces,\r\nChapter 2, in: Handbook of Face Recognition, Springer-Verlag\r\nLondon Limited, 2011, pp. 19\u201349.\r\n[2] A. M. Patil, S. R. Kolhe, P. M. Patil, 2D face recognition techniques:\r\na survey, International Journal of Machine Intelligence 2 (2010) 74\u2013\r\n83.\r\n[3] M. Turk, A. Pentland, Face recognition using Eigenfaces, in: IEEE\r\nInternational Conference on Computer Vision and Pattern\r\nRecognition (CVPR), 1991, pp. 586\u2013591.\r\n[4] P. Belhumeur, J. Hespanha, D. Kriegman, Eigenfaces vs. Fisherfaces:\r\nrecognition using class specific linear projection, IEEE Transactions\r\non Pattern Analysis and Machine Intelligence 19 (7) (1997) 711\u2013720.\r\n[5] M. I. Ahmad, W. L. Woo, S. S. Dlay, \u201cMultimodal biometric fusion\r\nat feature level: Face and palmprint,\u201d Proc. of International\r\nSymposium on Communication Systems Networks and Digital Signal\r\nProcessing (CSNDSP), pp. 801-804, 2009.\r\n[6] Bhat, V. S., Pujari, J. D.,\"Face Recognition Using Holistic Based\r\nApproach\". International Journal of Emerging Technology and\r\nAdvanced Engineering, 4(7), pp. 134-141, 2014.\r\n[7] X. He, S. Yan, Y. Hu, N. Partha, H.-J.Zhang, Face recognition using\r\nLaplacianfaces, IEEE Transactions on Pattern Analysis and Machine\r\nIntelligence 27 (3) (2005) 328\u2013340.\r\n[8] M.-H. Yang, N. Ahuja, D. Kriegman. Face recognition using kernel\r\neigenfaces, in: IEEE International Conference on Image Processing\r\n(ICIP), vol. 1, 2000, pp. 37\u201340.\r\n[9] K.-I. Kim, K. Jung, H.-J.Kim, Face recognition using kernel principal\r\ncomponent analysis, IEEE Signal Processing Letters 9 (2) (2002) 40\u2013\r\n42.\r\n[10] M.-H. Yang, Kernel eigenfaces vs. kernel fisherfaces: face\r\nrecognition using kernel methods, in: IEEE International Conference\r\non Automatic Face and Gesture Recognition (AFGR), 2002, pp. 205\u2013\r\n211.\r\n[11] T. Ahonen, A. Hadid, M. Pietik\u00e4inen, Face description with local\r\nbinary patterns: application to face recognition, IEEE Transactions on\r\nPattern Analysis and Machine Intelligence 28 (12) (2006) 2037\u20132041.\r\n[12] Y. Rodriguez, S. Marcel, Face authentication using adapted local\r\nbinary pattern histograms, in: European Conference on Computer\r\nVision (ECCV), 2006, pp. 321\u2013332.\r\n[13] F. Cardinaux, C. Sanderson, S. Marcel, Comparison of MLP and\r\nGMM classifiers for face verification on XM2VTS, in: International\r\nConference on Audio- and Video-based Biometric Person\r\nAuthentication (AVBPA), Springer, Berlin, 2003\r\n[14] F. Samaria, S. Young, HMM-based architecture for face\r\nidentification, Image and Vision Computing 12 (8) (1994) 537\u2013543. [15] A. Nefian, M. Hayes, Hidden Markov models for face recognition, in:\r\nIEEE International Conference on Acoustics, Speech, and Signal\r\nProcessing (ICASSP), vol. 5, 1998, pp. 2721\u20132724.\r\n[16] A. Martinez, Face image retrieval using HMMs, in: IEEE Workshop\r\non Content- Based Access of Image and Video Libraries, 1999, pp.\r\n35\u201339.\r\n[17] Hongmei Li, Dongming Zhou, RencanNie, Analysis of Face\r\nRecognition Methods in Linear Subspace, Lecture Notes in Electrical\r\nEngineering Vol.269, 2013, pp 3045-3051\r\n[18] S. Lucey and T. Chen, \"A GMM parts based face representation for\r\nimproved verification through relevance adaptation,\" Proceedings of\r\nthe IEEE Computer Society Conference on Computer Vision and\r\nPattern Recognition 2, 2004, pp. II855-II861.","publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 97, 2015"}