提出一种压缩感知成像框架结构.该结构采样端用新建的采样矩阵实现数字微镜阵列分区控制,可增强信息获取的准确性,测量得到与新数字微镜阵列对应的压缩采样值；重构端由采样值优化重构出低分辨率图像后,根据分区控制过程建立压缩感知理论框架下的超分辨重建模型,利用梯度稀疏约束优化算法进行求解,恢复出原高分辨率图像.实验结果表明： 数字微镜阵列分区控制与超分辨重建相结合的方法可以明显降低压缩感知成像系统的计算量,缩短成像时间,并且具有较高的图像重构质量.

A compressive sensing based imaging framework was proposed. At the sampling end, in order to enhance the acquisition accuracy of information, a new measurement matrix was designed to carry out zone control on digital micromirror device and the new digital micromirror device with zone control was used to make compressive measurements. At the recovery end, a low resolution image was first recovered by solving an optimization problem from the received measurements. Then, the modeling of a super-resolution problem in a compressive sensing framework was constructed according to the zone control process, and a total variation algorithm was exploited to solve such a compressive sensing based super-resolution problem for the high resolution image. Experimental results show that the proposed method based on zone control of digital micromirror device and super-resolution reconstruction can greatly shorten the imaging time and has very low computational complexity and excellent recovery performance.