Study On Compressed Sensing Of Color Images Based On The Dual Tree Complex Wavelet

Compressed sensing (CS) is capable of recovering the original signal from the projections which is less than the Nyquist sampling ratio using the sparse priors of it. Nowdays, the CS has been widely applied in many areas, such as compressive gray images, and so on; however there are few papers for CS of color images. This paper performs the researches on reconstruction of color images based on CS after learning CS theory and the existent reconstruction algorithms, mainly including following three aspects.First of all, aiming to overcome the shortcomings of lacking of diretion selectivity and translation variance of the orthogonal wavelet, color images are sparsity represented by the dual tree complex wavelet, then transform coefficients of each channel are processed seperatly. An algorithm on compressed sensing of color image reconstruction based on the dual tree complex wavelet and iterative shrinkage is proposed. The experiment results show the validity of the algorithm.Secondly, to increase the quality of the reconstructed color image, inter-scale dependency of transform coefficients of color images'three channels is described by multivariate model. After solving the equation of CS of color image, the approximate closed roots are obtained, and then the dual tree complex wavelet is combined with it to reconstruct the color image. An algorithm on compressed sensing of color images based on multivariate model in the dual tree complex wavelet domain is proposed. The results of experiments show that, the reconstructed color image has better vision quality.At last, intra-scale dependency of transform coefficients of color images'three channels is described by local Gaussian model. From the maximum a posteriori estimation, the threshold operators of color image are received after deducing and solving the new equation of CS of color image. A new CS of color images reconstruction algorithm is proposed by exploiting the advantage of the dual tree complex wavelet on sparsity representation of images. An algorithm on compressed sensing of color images based on local Gaussian model in the dual tree complex wavelet domain is proposed. The results of experiments show that the algorithm can improve the Peak Signal-to-Noise Ratio and the quality of the reconstructed color image.