Matching Pursuit Algorithm For Image Reconstruction Based On Compressive Sensing

Compressive sensing (CS) is a novel signal sampling theory under the condition that the signal is sparse or compressible. It has the ability of compressing a signal during the process of sampling. Reconstruction algorithm is one of the key parts in compressive sensing, and it is of great significance to accurately reconstruct a signal and verify the sampling accuracy. In this paper, properties of the existing reconstruction algorithms are firstly analyzed. Based on that, the main contributions of this paper are summarized as follows.An improvement scheme for OMP algorithm is given. To increase the convergence speed of OMP algorithm, the image to be processed is divided into some blocks. The new scheme could significantly improve the computation efficiency although it may reduce the reconstruction accuracy, which is hardly noticeable.A novel enhancement scheme for image reconstruction is presented based on an idea of balance. In many existed image processing or reconstruction algorithms, images are always processed in a column-wise manner, which ignores the correlation between columns. In this paper, the images are processed both in column-wise and row-wise manners and the final results are obtained by using a balance scheme. The experimental results show that for different images and sampling rate, it could get better performance.And then a new adaptive matching pursuit (VssAMP) algorithm is presented by introducing an idea of variable step size. The proposed algorithm could control the accuracy of reconstruction by both variable step size and double thresholds although the sparsity of a signal is unknown. The experimental results show that the proposed algorithm can get better reconstruction performances and is superior to other algorithms both visually and objectively.