Efficient multiple frame images recovery based on distributed compressed sensing

Distributed compressed sensing (DCS) shows great potential in reducing data acquisition in multiple frame. The images from multiple frame are highly related with each others. The efficiency of recovering these images is poor without taking the high correlation of multiple frame images into account. A novel reconstruction algorithm based on distributed compressed sensing is proposed. In order to take full use of the intra-signal correlation, a key frame image is recovered using compressive sampling match pursuit (CoSaMP). It is taken as the prior information. Then the compressive sampling match pursuit is modified by bringing in the prior information. Our presented method avoids reconstructing the common component of images repeatedly and reduces the computational complexity. Experimental results show that the proposed approach reduces a large number of required measurements at decoder and the calculation speed is faster than the existing state-of-the-art methods.