Compressed sensing image reconstruction algorithm based on regional segmentation

The existing compressed sensing image reconstruction algorithms cannot combine reconstruction effect with reconstruction speed at the same time. A new image reconstruction algorithm is proposed for compressed sensing. The new algorithm retains the advantages of block-sampling compressed sensing by the image blocks segmentation. The subblocks of the image edges are extracted and then the edges structure information are added on the basis of matching pursuit algorithm (MP). The accuracy and speed of the MP algorithm has been improved and the blocking effect generated by block-sampling reconstruction is overcomed. Experimental results show that the new algorithm is better than other similar algorithms on the computation time and the accuracy of the reconstruction, and achieves the fast and accurate image reconstruction.