Edge detection innovator based on wavelet coefficients for images corrupted by the white-Gaussian noise

Denoising of images is one of the vital topics in image manipulating. Approaches for denoising a chain of images aims to attenuate additive noise to the lowest possible rates by using both spatial and temporal areas. Conversely, extracting the edges of images that affected by the White-Gaussian noise was the major dilemma faced by many researchers. Many of the denoising image methods based on wavelet have been proposed to extract the edges from both the vertical and horizontal image gradients. In this paper, denoising of images obtained after thresholding of wavelet coefficients. At the same time, an adaptive average filtering for each pixel in the neighborhood of the processed pixel is used. The method could denoise each of the smooth piecewise as well as images of the natural textured as they were carried enough redundancy. Furthermore, the weights in this averaging were determined after finding similar patches in the neighborhood around pixels matched to describe their contents. Accordingly, the best extraction method for the vertical and horizontal image gradients is achieved after changing the magnitude of the threshold. These were extracted from the histogram of these gradients. Experiment results demonstrate that the proposed method simultaneously provided significant improvements in terms of the blockiness artifacts as well as enhancing the quality of images in terms of visual perception