Description

Getting information from different domains is explored to give a good face representation for recognition. Collecting information in image space, scale and orientation domains can provide rich important clues not seen in either individual of these domains. Thus we introduce gabor analysis and local binary pattern analysis in a combined way to propose a new face recognition system. GV-LBP (E-GV-LBP) descriptor models the neighboring changes around the central point in the joint domains simultaneously for face representation. Statistical uniform pattern mechanism(LBP) is adopted Local histogram features based upon the uniform patterns are extracted. Discriminant classification is finally performed based upon weighted histogram intersection