Adaptive image denoising method based on absolute median difference and BM3D

The invention relates to a self-adaptive image denoising method based on absolute median difference and BM3D, and the method comprises the steps: dividing an input image into an image block set with a specific size, carrying out the iteration screening of uniform image blocks through local variance, carrying out the correction of a main dimension feature value through the absolute median difference, so as to obtain the noise standard deviation of a current image, and carrying out the denoising of the current image. According to the method, noise intensity estimation is obtained, partial parameter setting of a BM3D filtering algorithm is completed according to the obtained noise intensity estimation, in basic estimation of the BM3D algorithm, noise is preliminarily removed in a mode of setting a hard threshold value, then final estimation serves as an improvement means, matching of similar blocks is carried out by using a result of basic estimation, and the matching accuracy of the similar blocks is improved.