Kernel principal component analysis for texture classification

Kernel principal component analysis (PCA) has recently been proposed as a n onlinear extension of PCA. The basic idea Is to first map the input space i nto a feature space via a nonlinear map and then compute the principal comp onents in that feature space. This letter illustrates the potential of kern el PCA for texture classification. Accordingly, supervised texture classifi cation mas performed using kernel PCA for texture feature extraction. By ad opting a polynomial kernel, the principal components were computed within t he product space of the input pixels making up the texture patterns, thereb y producing a good performance.