Image Denoising Method Based On The Low Rank Texture

In the process of image acquisition and transmission, the quality of the image is often reduced by various noises, which seriously affects the subsequent image processing. Therefore, image denoising is an important part in the field of image processing. In recent years, with the rise of compressed sensing theory, the theory of low rank matrix recovery has been paid more and more attention by researchers. Therefore, the image denoising based on the low rank structure has become a hot topic in this field.In this paper, based on the research of low rank matrix recovery theory, a low rank sparse decomposition denoising model based on matrix and tensor is established. The main work of this paper is as follows:By using the affine transformation of the image sub block to obtain the low rank texture, a low rank sparse decomposition model is established. The optimization problem of the model is solved by using the augmented Lagrange multiplier algorithm and the fast singular value truncation algorithm based on matrix decomposition, and then the low rank texture of the matrix has been restored. Finally, a more robust image denoising method is realized.Adaptive center weighted filtering method is used to detect the position of random impulse noise and filter. The filtered image is decomposed into a plurality of image sub blocks, which are then stacked into a tensor form. And then the corresponding tensor low rank sparse decomposition model is established. The alternating direction method is used to solve the optimization problem in each model. Therefore, the lower rank tensor is obtained and expanded into a matrix form to obtain the denoising result of the image sub block. The final denoising result is obtained by calculating the mean value of the overlapped area of the image sub block.