基于时空相关性的分布式压缩感知多假设预测重构算法OA北大核心CSCDCSTPCD
Multihypothesis prediction algorithm of DCVS based on spatio temporal correlativity
综合考虑到视频序列本身的不同特性以及时空相关性，将传统视频编码中的多假设预测运动估计思想引入到分布式压缩感知视频编码系统中，提出一种新的基于时空相关性的分布式压缩感知多假设预测重构算法。在编码端增加CS帧的块模式判别，在解码端CS帧根据模式判别进行相应的基于块的多假设预测估计和残差稀疏重构。仿真实验表明，与传统的分布式压缩感知多假设预测重构算法相比，所提出的算法能够较好地改善视频序列的重建质量。
For overall consideration of different features of video sequences and the temporal and spatial correlation， and applying the multihypothesis motion estimation prediction to the distributed compressed video sensing(DCVS) framework from the original video
MENG Yu;ZHU Jin-xiu;ZHANG Yao;ZHU Shun-wu;MENG Zhen-qi
College of Internet of Things Engineering，Hohai University，Changzhou Jiangsu 213022，ChinaChangzhou Key Laboratory of Sensor Networks & Environmental Perception，Changzhou Jiangsu 213022，ChinaCollege of Internet of Things Engineering，Hohai University，Changzhou Jiangsu 213022，ChinaChangzhou Key Laboratory of Sensor Networks & Environmental Perception，Changzhou Jiangsu 213022，ChinaCollege of Internet of Things Engineering，Hohai University，Changzhou Jiangsu 213022，China
信息技术与安全科学
分布式压缩感知多假设预测时空相关性稀疏重构
distributed compressed video sensingmultihypothesis predictiontemporal and spatial correlationsparse recovery
《计算机应用研究》 2014 (2)
637-640,4
评论