The Weight-Block Compressed Sensing and its Application to Image Reconstruction

Compressed sensing (CS) is a novel theory for simultaneous data sampling and compression. The block compressed sensing can reduce the computation complexity and storage space for compressed sensing. In this paper, the weight-block compressed sensing technique coupled with the edge information is presented for improving the reconstructed image quality. Firstly, we segment the original image into block by block. Based on the edge characteristic of every sub-block, we will select the different measurements that needed for each block. This algorithm can preserve the edge and reduce the aliasing in comparison to the traditional block-compressed sensing. Experimental results show that the proposed algorithm can improve the PSNR comparing with the usual method.