Open AccessDWT-based Illumination Normalization and Feature Extraction for Enhanced Face Recognition
Author(s) -
K. Manikantan,
Milan S. Shet,
Minal Patel,
S. Ramachandran
Publication year - 2012
Publication title -
international journal of engineering and technology
Language(s) - English
Resource type - Journals
ISSN - 2227-524X
DOI - 10.14419/ijet.v1i4.353
Subject(s) - artificial intelligence , pattern recognition (psychology) , normalization (sociology) , computer science , facial recognition system , feature extraction , discrete wavelet transform , dimensionality reduction , face (sociological concept) , particle swarm optimization , invariant (physics) , computer vision , wavelet , mathematics , wavelet transform , algorithm , social science , sociology , anthropology , mathematical physics
Face recognition (FR) under varying lighting conditions is challenging, and exacting illumination invariant features is an effective approach to solve this problem. In this paper, we propose to utilize Discrete Wavelet Transform (DWT) for normalizing the illumination variance in images as well as for feature extraction. Individual stages of the FR system are examined and an attempt is made to improve each stage. A Binary Particle Swarm Optimization (BPSO) based feature selection algorithm is used to search the feature space for the optimal feature subset. Experimental results, obtained by applying the proposed algorithm on YaleB and Color FERET face databases, show that the proposed system outperforms other FR systems. A significant increase in the recognition rate and a substantial reduction in the number of features is observed. Dimensionality reduction obtained is more than 99% for both YaleB and Color FERET databases.