Face recognition based on skin color information and Support Vector Machine

A new face recognition method based on Mahalanobis distance and Support Vector Machine (SVM) using skin color information is proposed. According to face skin color distribution in YCbCr color space,The Mahalanobis distance map of the image is obtained and use Independent Component Analysis (ICA) to extract features and establish Eigenfaces space. A novel SVM method is proposed based on loop-symmetrical division for solving multi-classification problem. This algorithm is loop-arranged and symmetrical divided classes of multi-classification problem, and constructed error-correcting codes matrix. The class of an unknown sample is obtained by the decode function based on error-correcting codes matrix. Experiment shows the new face recognition method is effective.

[1]  Jason Weston,et al.  Multi-Class Support Vector Machines , 1998 .

[2]  Elli Angelopoulou,et al.  Understanding the color of human skin , 2001, IS&T/SPIE Electronic Imaging.

[3]  Vladimir Vezhnevets,et al.  A Survey on Pixel-Based Skin Color Detection Techniques , 2003 .

[4]  David J. Kriegman,et al.  Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection , 1996, ECCV.

[5]  Azriel Rosenfeld,et al.  Face recognition: A literature survey , 2003, CSUR.

[6]  David Zhang,et al.  Palmprint recognition using eigenpalms features , 2003, Pattern Recognit. Lett..

[7]  M. Turk,et al.  Eigenfaces for Recognition , 1991, Journal of Cognitive Neuroscience.

[8]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[9]  Li Li,et al.  3D face recognition by constructing deformation invariant image , 2008, Pattern Recognit. Lett..

[10]  Ming-Hsuan Yang,et al.  Kernel Eigenfaces vs. Kernel Fisherfaces: Face recognition using kernel methods , 2002, Proceedings of Fifth IEEE International Conference on Automatic Face Gesture Recognition.