基于稀疏支撑集先验的压缩感知图像序列重建算法.pdfVIP

第36 卷第6 期 Vo l. 36 No.6 南京理工大学学报 2012 年 12 月 Joumal of Nanjing University of Science and Techno!ogy Dec.2012 基于稀疏支撑集先验的压缩感知图像序列重建算法 李星秀I ，韦志辉2 ，肖 亮2 (南京理工大学l.理学院;2. 计算机科学与工程学院，江苏南京 21∞94) 摘 要:针对现有压缩感知图像序列重建算法重建精度不高、模型参数设直较多的问题，提出了 一种结合稀疏支撑集先验和残差补偿的算法。在已知前一帧图像重建结果的基础上，通过求解 1 个最小化加权II 范数问题得到当前帧图像的初始估计。通过对估计残差进行压缩感知重建 并对初始估计加以补偿，得到当前帧图像的最终重建结果。与其他同类算法相比，该算法减少 了阐值参数的设直。实验结果表明，在相同的测量值数目下，该算法重建图像的相对误差、峰值 信噪比和结构相似度指标均优于同类比较算法。 关键词:稀疏支撑集;压缩感知;图像序列;残差补偿 中图分类号:TN911.72 文章编号: 1005-9830 ( 2012 ) 06 -0973 -06 Compressed Sensing Image Sequence Reconstruction Algorithm ßased on Sparse Support Prior LI Xing-xiu l ,WEI Zhi-hui2 ,XIAO Liang2 (1. School of Sciences ;2. School of Computer Science and Engineering ,NUST , Nanjing 210094 , China) Abstract: Aiming at the problems of low accuracy and more model parameters of traditional compressed sensing image sequence reconstruction algorithms , a novel algorithm combining sp町se support prior and residual compensation is proposed. The initial estimation of the current image is obtained by solving a weighted 1 norm minimization problem based on knowing the reconstruction of 1 the previous image. The final estimation of the current image is generated by the compressed sensing reconstruction of the estÎmation error and the compensation of the original estimation. Compared with other similar algorithms , the proposed algorithm reduces the number of threshold p缸ameters. Experimental results show that the proposed algorithm is superior to other similar algo