Human Face Recognition with Different Statistical Features

This paper examines application of various feature domains for recognition of human face images to introduce an efficient feature extraction method. The proposed feature extraction method comprised of two steps. In the first step, a human face localization technique with defining a new parameter to eliminate the effect of irrelevant data is applied to the facial images. In the next step three different feature domains are applied to localized faces to generate the feature vector. These include Pseudo Zernike Moments (PZM), Principle Component Analysis (PCA) and Discrete Cosine Transform (DCT). We have compared the effectiveness of each of the above feature domains through the proposed feature extraction for human face recognition. The Radial Basis Function (RBF) neural network has been utilized as classifier. Simulation results on the ORL database indicate the effectiveness of the proposed feature extraction with the PZM for human face recognition.

[1]  Weiyang Zhou,et al.  Verification of the nonparametric characteristics of backpropagation neural networks for image classification , 1999, IEEE Trans. Geosci. Remote. Sens..

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

[3]  M. A. Grudin,et al.  On internal representations in face recognition systems , 2000, Pattern Recognit..

[4]  Chin-Chuan Han,et al.  Why recognition in a statistics-based face recognition system should be based on the pure face portion: a probabilistic decision-based proof , 2001, Pattern Recognit..

[5]  Karim Faez,et al.  Human Face Recognition Using Shape Information and Pseudo Zernike Moments , 2000, International Symposium on Vision, Modeling, and Visualization.

[6]  Hilary Buxton,et al.  Face Recognition using Radial Basis Function Neural Networks , 1996, BMVC.

[7]  J. Haddadnia.  Human Face Recognition Using Radial Basis Function Neural Network , 2000 .

[8]  Narendra Ahuja,et al.  Detecting Faces in Images: A Survey , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[9]  Karim Faez,et al.  Neural network based face recognition with moment invariants , 2001, Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205).

[10]  Erik Hjelmås,et al.  Face Detection: A Survey , 2001, Comput. Vis. Image Underst..

[11]  Roland T. Chin,et al.  On Image Analysis by the Methods of Moments , 1988, IEEE Trans. Pattern Anal. Mach. Intell..

[12]  Paul M. Embree,et al.  C Language Algorithms for Digital Signal Processing , 1991 .

[13]  Carlos E. Thomaz,et al.  Design of radial basis function network as classifier in face recognition using eigenfaces , 1998, Proceedings 5th Brazilian Symposium on Neural Networks (Cat. No.98EX209).

[14]  Karim Faez,et al.  A hybrid learning RBF neural network for human face recognition with pseudo Zernike moment invariant , 2002, Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290).