基于单层小波变换的加权压缩感知图像处理

Weighted Compressed Sensing for Image Processing Based on the Single Layer Wavelet Transform

针对简单正交基不能足够稀疏表示信号问题,提出了一种基于单层小波变换改进的加权压缩感知算法。根据图像小波变换的特点,对图像进行单层小波分解,保留低频系数,对高频系数进行测量;并提出设置加权系数矩阵,作用于信号小波正交变换后的高频稀疏系数,增强其系数的稀疏性,增强图像的重构质量;重构算法采用贪婪算法中的OMP算法。实验结果表明该算法对重构精度有进一步提高。

Since the orthogonal basis can not sparsify the signal efficiently,a new weighted compressed sensing algorithm is presented based on the single layer wavelet transform. According to the properties of the wavelet transform, we use a single-layer wavelet transform to decompose images preserve the low-pass coefficients and measured the high-pass wavelet coefficients. This algorithm employs the weighted coefficient matrix to enhance the high-pass coefficient sparsity and the quality of the recovered image. The orthogonal matching pursuit belongs to a greedy algorithm and is used for reconstruction. Simulation results show that the proposed algorithm can improve the quality of the recovered image.