从低分辨率压缩视频流恢复高分辨率图像

Restoring high-resolution ratio image from low-resolution ratio compressed video flow

在压缩视频高分辨率处理领域引入图像序列中的运动矢量、压缩噪声和其他冗余的方法,总结了获取和压缩系统模型以及原有的HR灰度和位移模型。将Bayesian框架内的问题公式化,定义了观测程序用到的获取系统,并进一步提出了Bayesian框架内的所有技术。该方法有助于本领域不同方法之间的比较。

Movement vector, compressed noise and other redundancy of image sequence were introduced to the field of high- resolution ratio process in compressed video. Acquiring ＆ compressing system model, the original HR gray and movement model were summarized. The problems in the frame of Bayesian were formulized, and the acquiring system used by observation program was defined. Furthermore, all the technologies in the frame of Bayesian were given to compare the different methods,