基于改进SP算法的压缩感知图像重构.pdfVIP

第 30 卷第 7 期 计算机应用与软件 Vol.30 No.7 2013 年7 月 Computer Applications and Software Jul. 2013 基于改进 SP 算法的压缩感知图像重构 吴延海问迪 (西安科技大学通信与信息工程学院 陕西西安 71ω54) 摘 要 针对压缩感知子空间追踪 SP(subspace pursuit) 算法必须以信号稀疏度为先验知识，而现实中图像稀疏度未知这一问题， 提出改进 SP 算法 MSP( modified subspace pursuit) 。首先对信号的稀疏度进行自适应估计，其次在迭代过程中，通过给定的步长因子 对稀疏度进行更新，使之逐渐逼近正确子空间，当重构误差小于阂值时，停止迭代，实现稀疏信号的重构。重构图像表明 :MSP 算法 在运算时间和重构精度上均优于其他同类算法，实现了图像的快速精确重构。 关键词 压缩感知子空间追踪重构图像处理采样稀疏度 中圄分类号霄301 文献标识码 A DOI: 10. 3969/j. issn. 1侃)()-386x. 2013. 07.053 CO岛fPRESSIVE SENSING IMAGE RECONSTRUCTION BASED ON MSP Wu Yanhai Yan Di (School 01 Communication and li呐Irmatwn Ej咱neer吨， Xi an University o/Science and Technology ,Xi an 710054 ,Shaanxi , China) Abstract Compressive sensing subspace pursuit (SP) has to take 由e sparsity of signals as its priori knowledge , while the image sparsity actuality cannot be firstly got. Aiming at this problem , the modified-SP (MSP) algorithm is proposed in this paper. First , it adaptively estimates the sparsity of the si伊al; then in iteration process , it updates the sparsity by the given step factor to enable gradually approaching to the coπect subspace , and stops iterating when the reconstruction eπor is less than the threshold , thus achieves the reconstruction of sp缸W signal. Reconstructed images show that the MSP algorithm outperforms other similar algorithms in comput础。n time and reconstruction accuracy , and realises fast and accurate image reconstruction. Keywords Compressive sensing Subspace pursuit Reconstruction Image processing Sampling Sparsity [11 ]提出的子空间匹配追踪 SP 算法和文献[12 ]提出的压缩感 。引言