Contourlet feature based kernel relevance weighted discriminant analysis for face recognition

This paper presents a novel face recognition method based on the contourlet for facial features representation and using an new kernel based algorithm, for discriminating purposes, namely kernel relevance weighted discriminant analysis (KRWDA). This nonlinear reduction dimension algorithm has several interesting characteristics. First, using kernel theory, it handles nonlinearity efficiently. Second, by incorporating a weighting function into discriminant criterion, it overcomes overemphasis on well-separated classes and hence can work under more realistic situations. Finally, it can effectively deal with the small sample size problem by using a QR decomposition on the scatter matrices. We have performed multiple face recognition experiments to compare the proposed method with other dimensionality reduction methods showing its good performance.