Compressive Sensing Algorithm And Its Application In Imaging

Compressive sensing is a new theory of signal processing. It breaks through the limit of Nyquist sampling rate and can recover a sparse signal successfully below Niquist sampling rate. Compressive sensing theory is a revolution in the field of signal sampling and it proposes new ideas and methods for signal processing. Due to the sparsity of natural signal, the application of compressive sensing can improve the performance of signal processing system effectively. In this paper, we will give the modified optimization of measurement matrix and modified reconstruction algorithm. We will also do some researches on compressive imaging system. The main work of this paper can be summarized as follows:Firstly, we introduce an optimization algorithm of measurement matrix based on iteration and modify it. The reconstructing ability of OMP depends on the cross-coherence of sensing matrix. In this paper, we introduce an adaptive strategy of threshold selection and keep the sparse matrix unchanging. The modified optimization algorithm based on iteration reduces the cross-coherence between atoms and the reconstructing performance of OMP improved.Secondly, a modified reconstruction algorithm based on LAOMP is proposed. Due to the slowness of iterative convergence and the lack of error correction, we change the selection method and the number of atoms adaptively which waiting for prediction. The reconstruction accuracy is improved and the number of iteration is reduced with the parallel multi-atom chosen and the backtracking mechanism.Finally, we use compressive sensing algorithm in imaging system. Due to the huge number of sampling and the bad quality of imaging, we use compressive sensing in traditional imaging system. With the modified recovery algorithm, the number of sampling is reduced and the quality of imaging is improved.