Encoding and Reconstruction about Video Image via Compressed Sensing

Abstract

A new method for encoding and reconstruction high quality video image is given in this paper which uses the theory of compressed sensing. First the image frame of video is transformed into DCT domain. Then Image coding and decoding process using CS theory is given, frame I in image sequences is coded by frame coding mode after doing CS sampling to the DCT coefficients and the difference vector dv of the t-th fame for fame P. CS reconstruction and IDCT are done during decoding. Finally, the high quality reconstructed image is obtained. The experimental results shows that for images with sparseness, the image coding and decoding system integrated with CS theory and its methods can be used to obtain reconstructed images with high quality, and comparing with DCT and IDCT method, the method has some improvement in the term of PSNR for general images.