PDF based face recognition system in different colour channels using discrete wavelet decomposition

Abstract

In this paper, a new high performance face recognition system based on the probability distribution functions of pixels obtained from intensity images in HSI and YCbCr colour channels and their decomposed images obtained by discrete wavelet decomposition is proposed. The PDFs of the equalized face images in spatial and subband domains in different colour channel are used as statistical feature vectors for the recognition of faces by minimizing the Kullback- Leibler Distance between the PDF of a given face and the PDFs of faces in the database. Majority voting and feature vector fusion methods have been employed to combine feature vectors obtained from different colour channels to improve the recognition performance.