讨论了压缩感知理论用于图像稀疏重建的基本流程. 采用正交匹配追踪重建算法和正交归一化的随机高斯测量矩阵，对离散余弦变换和离散小波变换两种稀疏表示算法进行分析比较，通过调节实验图像的分块大小和采样率大小、采样率和稀疏表示算法对重构效果和效率的影响. 在图像的稀疏表示方面，离散余弦变换整体上比离散小波变换性能更好. 为了在重构效果与效率之间取得平衡，需要合理选择分块大小和采样率.
Application of compressed sensing to sparse reconstruction of image is discussed. An orthogonal matching pursuit algorithm for reconstruction and Gaussian random matrix for measurement are used. We analyze and compare DCT and DWT both theoretically and experimentally. By adjusting the sub-block size and sampling rate of the experimental images, we make a comprehensive comparison of sub-block size,sampling rate and influences of the two algorithms on effectiveness and efficiency of sparse reconstruction in terms of runtime, reconstruction error and visual effects. In sparse image representation, DCT exhibits better overall performance than DWT. In order to achieve an optimal balance between reconstruction effectiveness and efficiency, a reasonable choice of sub-block size and sampling rate is required.
[1] CANDES E J, ROMBERG J, TAO T. Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information [J]. IEEE Transactions on Information Theory, 2006, 52(2): 489-509.
[2] LIU Jing, MALLICK M, HAN Chongzhao, YAO Xianghua, LIAN Feng. Similar sensing matrix pursuit: an efficient reconstruction algorithm to cope with deterministic sensing matrix [J]. Signal Processing, 2014, 95: 101-110.
[3] LIU Xiaoman, ZHU Yonggui. A Fast method for TV-L1-MRI image reconstruction in compressive sensing [J]. Journal of Computational Information Systems, 2014, 10(2): 691-699.
[4] MICCHELLI C A, SHEN Li Xin, XU Yue Sheng, ZENG Xue Ying. Proximity algorithms for the L1/TV image denoising model [J]. Advances in Computational Mathematics, 2013, 38(2): 401-426.
[5] XU Zhiqiang. Deterministic sampling of sparse trigonometric polynomials [J]. Journal of Complexity, 2011, 27(2): 133-140.
[6] BOURGAIN J, DILWORTH S J, FORD K, KONYAGIN S, KUTZAROVA D. Explicit constructions of RIP matrices and related problems [J]. Duke Mathematical Journal, 2011, 159(1): 145-185.
[7] ZHANG Tong. Sparse recovery with orthogonal matching pursuit under RIP [J]. IEEE Transactions on Information Theory, 2011, 57(9): 6215-6221.
[8] YU Huimin, FANG Guangyou. The application of compressive sensing in the three dimensional imaging of ground penetrating radar [J]. Journal of Electronics and Information Technology, 2010, 32(1): 12-16.
[9] FU Qiang, LI Qiong. The research of constructing the measurement matrix in compressive sensing [J]. Computer and Telecommunication, 2011, 7(1): 39-41.
[10] WANG Yan, LIAN Qiusheng, LI Kai. MRI image reconstruction based on joint regularization and compressed sensing [J]. Optical Technology, 2010, 36(3): 350-355.
[11] QU Xiaobo, GUO Di, NING Bende. Undersampled MRI reconstruction with the patch-based directional wavelets [J]. Magnetic Resonance Imaging, 2012, 30(7): 964-977.
[12] LI Zhilin, CHEN Houjin, LI Jupeng, YAO Chang, YANG Na. An efficient algorithm for compressed sensing image reconstruction [J]. Acta Electronica Sinica, 2011, 39(12): 2796-2800.
[13] HOU Jinman, HE Ning, LV Ke. Image fast reconstruction method based on compressive sensing [J]. Computer Engineering, 2011, 37(19), 215-217.
[14] YIN Hongpeng, LIU Zhaodong, CHAI Yia, JIAO Xuguo. Survey of compressed sensing [J]. Control and Decision, 2013, 28(10): 1441-1445.
[15] YUAN Quan, ZHANG Cheng, CHEN Jianjun, YAO Junxia, LI Yingran, WANG Huan. Image compressed sensing based on wavelet transform [J]. Computer Engineering, 2012, 38(20): 209-218.
[16] CARMI A Y, MIHAYLOVA L S, GODSILL S J. Introduction to compressed sensing and sparse filtering [M]. Berlin Heidelberg: Springer, 2014: 1-23.
