Palmprint Recognition by Applying Wavelet-Based Kernel PCA

This paper presents a wavelet-based kernel Principal Component Analysis(PCA) method by integrating the Daubechies waveletrepresentation of palm images and the kernel PCA method for palmprintrecognition. Kernel PCA is a technique for nonlinear dimensionreduction of data with an underlying nonlinear spatial structure.The intensity values of the palmprint image are first normalized byusing mean and standard deviation. The palmprint is then transformedinto the wavelet domain todecompose palm images and the lowest resolution subband coefficientsare chosen for palm representation. The kernel PCA method is thenapplied to extract non-linear features from the subband coefficients.Finally, similarity measurement is accomplished by using weightedEuclidean linear distance-based nearest neighbor classifier.Experimental results on PolyU Palmprint Databases demonstrate that theproposed approach achieves highly competitive performance with respectto the published palmprint recognition approaches.