中文摘要：

目前存在的CS恢复算法中大都采用固定的基函数，也就是在确定的域中对信号进行分解，比如：DCT域、小波域和梯度域，但这些域都忽略了自然信号的非平稳特性，缺乏白适应能力，从而不能够将图像分解得足够稀疏，也就使得CS恢复的效果很差，限制了CS在图像方面的应用。提出了一种基于分离Bregman迭代方法求解协同稀疏模型正则化的图像压缩感知恢复算法，能够在有效地刻画图像的局部平滑性和非局部自相似性的同时，获得更高质量的图像恢复效果。实验证明了本文提出算法的有效性，并且在峰值信噪比PSNR方面，比目前主流最好的算法高1dB。

英文摘要：

Most of conventional CS recovery methods utilized a set of fixed bases （ e. g. DCT, wavelet and gradient do main） for the entirety of a signal, which are irrespective of the non - stationarity of natural signals and cannot achieve high enough degree of sparsity, thus resulting in poor recovery performance. This paper proposes a new collaborative sparsity regularized framework for image compressive sensing recovery based on split Bregman iteration, which enforces image intrinsic local smoothness and nonlocal self- similarity, greatly improving image recovery performance. Experimental results demonstrate that the novel CS recovery strategy achieves significant performance improvements over the current state - of - the- art schemes by 1 dB in the peak signal- to- noise ratio （PSNR）.