Compressed Sensing Image De-noising Method Based on Regularization of Wavelet Decomposition

摘要

A method called compressed sensing image de-noising method based on wavelet decomposition was presented. In view of lack of sparsity in the above method, A new method called compressed sensing image de-noising method based on regularization of wavelet decomposition was presented, First, the image signal was decomposed by multi-scale wavelet, then high-frequency coefficients of each level was divided into two positive and negative by regularization; each level high-frequency coefficients was sampled by compressed sensing, and measured coefficients can be acquired; At last, measured coefficients was reconstructed, and then de-noising image can be acquired according to inverse wavelet transform. The simulation show that regularization can effectively increase the capacity of compressed sensing de-noising.