Block effect reduction via model based compressive sensing

In this chapter, we propose a novel block effect reduction algorithm based on Model based compressive sensing (MCS). Block effect reduction can be considered as image recovery from a degraded image. It is exactly what compressive sensing does. According to MCS, our approach can catch the tree structured sparseness of natural images in wavelet domain and the discontinuity between adjacent blocks in JPEG images. Hence, our approach has a good performance in visual quality and PSNR as shown in our intensive experiments. © 2012 Springer Science+Business Media B.V.