Nonsubsampled Pyramid Contourlet Transform and Its Application

The contourlet transform [CT] is a new two-dimensional representation for images that can capture the intrinsic geometrical structure of pictorial information. The construction proposed in this paper is based on a nonsubsampled pyramid structure and critically sampled directional filter banks. The proposed transform impartially takes into account the redundancy and the data amounts. Our experiments in image denoising and enhancement applications show that the proposed method provides better performance than wavelets and existing contourlets.