Algorithm for Face Recognition Using HMM and SVD Coefficients

Face Recognition stands high as a significant research area since it has plenty of application domains in pattern recognition, image processing, biometrics etc. Researchers contributed lot of algorithms and techniques to uncover the mask of face recognition arena. In this paper, a Left-Right Hidden Markov Models (HMM) based face recognition algorithm along with Singular Value Decomposition (SVD) Coefficients is discussed. Human face is divided into seven facial regions and a small number of quantized SVD Coefficients were trained to choose the facial features. Order Statistic Filtering is used as a preprocessing operation for efficient computation. Using SVD Coefficients, a face is considered as a numerical sequence representing block of images which can be easily modeled by discrete HMM. The system is tested on Olivetti Research Laboratory (ORL) face database consist of 400 images of 40 persons in .pgm format. For training, five face images of a person are considered and our proposed system achieves a recognition rate of 96.5% with a computational speed of 0.22 seconds per image. The experimental results reveal that our proposed system outperforms many of the traditional face recognition methods tested on ORL database.