Research On Super-Resolution Reconstruction Algorithm Of Compressed Video Based On MAP

Super-resolution reconstruction is to estimate a high-resolution image from a low-resolution image sequence. Because of the extensive application of the video compress techniques, the super-resolution reconstruction of compressed video has become a great focus.This paper firstly presents the meanings, applications and recent developments of super-resolution reconstruction, and analyzes the algorithms such as IBP, POCS and MAP. Secondly we discuss the compressing standards with the example of MPEG-2, and the key techniques, which include the subsample, the motion compensation prediction, the discrete cosine transform (DCT), and the quantization in detail. Then we study the algorithm of compressed video super-resolution reconstruction based on MAP and execute the simulation by building the compressed video model and surveying the models of quantization noise and motion. Finally, the possible direction of future researches on video super-resolution reconstruction is proposed.