【摘要】 为研究网络视频清晰度优化问题,提出了一种新型的针对网络视频的超分辨率方法。网络视频是人们获取信息的重要途径,然而大部分网络视频的分辨率无法满足人们对于高清晰度视频的要求。提出的方法利用多帧低分辨率视频序列之间的互补信息,重建出高分辨率视频序列。针对网络视频的特点,结合压缩感知理论,充分挖掘视频内部稀疏特征与自相关信息以实现网络视频的超分辨率。对网络视频进行建模仿真,结果表明,提出的方法可以很好的重建出高清晰度的网络视频。更多还原

【Abstract】 This paper proposed an innovative method for network video super resolution. Nowadays,network vid- eo is an important way for people to obtain information. However,most of the network videos are unable to meet people’s demand for high- resolution. The method proposed in this paper was designed to take the advantage of com- plementary information between multi- frame low- resolution images,to reconstruct the high- resolution sequence. The method takes into account the network video’s characteristic,fully taps the video’s internal sparse feature and self- relevant information through compressive sensing theory to achieve super- resolution. In this paper,we presented the simulation for network video,and the simulation results show that the proposed method has a good performance on the reconstruction for a high- resolution network video.更多还原