摘要
压缩感知理论是在已知信号具有可压缩性或通过变换具有稀疏性的条件下,对其信号进行采集,稀疏和重构的新理论。其中稀疏信号重构算法是其中关键的一部分,对信号恢复的精确性及时效性验证有着重要的意义。该文在总结目前已有的重构算法的基础上,提出了一种新的基于压缩感知的双连续超松弛迭代重构算法。该算法通过参数估计自适应的寻找合适的稀疏度K的值来平衡重构精度和重构速度之间的矛盾。实验结果表明,这种算法能够有效地提高了重构图像的主观视觉效果和峰值信噪比,加快了压缩传感图像重构算法的收敛速度,提高了重构精度。因此是一种综合性能较好的压缩感知重构算法。
Compressed sensing（CS） theory is a novel data collection,sparse and reconstruction theory in the condition that signal is sparse or compressible.The sparse signal reconstruction algorithm is the key part in compressed sensing,and it is of great significance to verify the reconstruction accuracy and efficiency.This paper summarizes the current reconstruction algorithms,and then proposes a new double successive over-relaxation（DSOR）.In order to strike a balance between the efficiency and accuracy,DSOR algorithm adapts parameter estimation to find the appropriate value of sparse level.The results of experiments show that proposed algorithm can effectively improve the subjective visual quality and peak signal-to-noise ratio,and accelerate the convergence of reconstruction algorithm.Thus,it is a better reconstruction algorithm in general.
出处
《杭州电子科技大学学报（自然科学版）》
2011年第6期79-82,共4页
Journal of Hangzhou Dianzi University：Natural Sciences
关键词
压缩感知
重构算法
稀疏信号重构
超松弛
compressed sensing
reconstruction algorithm
sparse signal reconstruction
over-relaxation
