Research On Block-sparse Signal Reconstruction And Image Block Sampling Algorithm Based On Compressive Sensing

Compressed Sensing(CS)is a kind of signal processing technology which takes the compression and sampling as the core idea at the same time.It breaks through the sampling limits of the traditional Nyquist sampling theorem,and can restore the signal with high fidelity with small sampling samples,so that it is highly concerned in the field of signal processing,image processing and wireless communication and so on.On the basis of CS theory,block-sparse signal and image block compressed sensing are two popular research directions in recent years.In the foundation of the deeper research on the reconstruction problem of block sparse signal and the sampling problem of image block compressed sensing,and mainly research on the following aspects in this papers.Firstly,the advantages and disadvantages of common measurement matrix and classical greedy tracking algorithm in CS theory were analyzed and summarized.On the basis of the theory of compressed sensing,in the case of the block sparsity requires a priori and.the initial parameter setting of stage-long has a great influence on the algorithm performance,this paper proposed an adaptive reconstruction algorithm based on estimating of block sparsity in compressed sensing.The algorithm obtained an estimated value of block sparsity by estimates calculation of the block sparsity,and then it initialized the residual and stage-long with the estimated value.At the same time,it reconstructed the block-sparse signal by combining with subspace tracking,the principle of correlation maximization and the principle of regularization.Adaptive reconstruction algorithm based on estimating of block sparsity in compressed sensing was realized on Matlab2015a,and a series of experiments have been performed to verify the effectiveness of the algorithm.According to experiment results,this algorithm not only can obtain better recovery probability,but also can effectively shorten the time required for signal reconstruction.Next,in the case of in image block compressed sensing,using the same sampling rate to sample each block is easy to make some blocking artifacts appearing in the reconstructed image,this paper proposed an adaptive block compressed sensing method based on the difference of image information to solve this problem.The algorithm distinguished the amount of information contained in the image by the definition of the difference coefficient of image information,and then it obtained the number of image block samples by the proposed adaptive sampling strategy,next it reconstructed each image block according to the sampling results and combined each image block into a complete image.Adaptive block compressed sensing method based on the difference of image information was realized on Matlab2015a,and a series of experiments have been performed to verify the effectiveness of the algorithm.According to experiment results,this algorithm not only can optimize the allocation of sampling resources,and can effectively improve the reconstructed image quality.In summary,the results that research on block sparse signal reconstruction algorithm and image block compressed sensing sampling algorithm have positive practical significance for the application of compressed sensing in block-sparse signal processing and image processing.