基于自适应卡尔曼的时域增强算法研究  

Research on time domain enhancement algorithm based on adaptive Kalman filter

作　　者：张金丽[1] 张淑芳[1] 

机构地区：[1]天津大学电子信息工程学院,天津300072

出　　处：《信息技术》2016年第8期9-13,共5页Information Technology

基　　金：国家自然科学青年基金(61101226)

摘　　要：在图像序列进行压缩感知重构的过程中,基于运动补偿的分块压缩感知重建算法利用了帧间残差图像的稀疏特性,有效提高了重构视频的质量。但该算法仅在空域对图像进行了维纳滤波,帧间存在抖动现象,视频主观质量较差。文章将自适应卡尔曼滤波算法应用到分块压缩感知重建算法的重建过程中,可以有效地去除视频帧间的噪声,使得图像的主观质量得到了改善。In the process of compressed sensing reconstruction for image sequence,block compressed sensing reconstruction algorithm based on the motion compensation effectively improved the quality of the reconstructed video frames by utilizing the characteristics of inter sparse residual image. But this algorithm just utilized Wiener filtering in the spatial domain, and for the inter jitter existed, the reconstructed video frames had a little bit poor subjective quality. In this paper,an adaptive Kalman filter algorithm is applied to block compressed sensing reconstruction algorithm in the reconstruction process,which effectively removes the noise between image frames and makes better subjective quality.

关 键 词：压缩重构 分块压缩感知重建 视频主观质量 时域增强 自适应卡尔曼滤波 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]