The NAMlet transform

摘要: A novel image sparse representation method, named NAMlet transform, is proposed by fully considering the structure of nature images.The transform based on asymmetric homogeneous blocks, which could preserve detail information can remove restrictions size images. Image methods have been widely applied in many processing fields, such as computer vision, de-noising, super resolution, and visual tracking. An efficient method improve accuracy. However, few traditional consider from point anti-packing problem. Thus, these are not only restricted image, but also lose a great amount using symmetric blocking method. In this paper, we an called Transform. The NAMlets haar-type wavelets, non-symmetric blocks obtained non-symmetry model. all pixels same bit-plane. reduce lost size. experiment results show strong superiority for comparison with some state-of-the-art methods.