Abstract: In this paper, we address the problem of principal component analysis technique which is the modeling method of active shape model.Since the principal component analysis usually ignores nonlinear relativity among a large number of pixels, kernel method is combined to research object localization.A nonlinear active shape model based on kernel principal component analysis is presented in this paper.The proposed method reconstructs the pre-image in input space through the idea in multidimensional scaling.The validity of this method is demonstrated by the results of preliminary experiments.