A fast image denoising algorithm based on local weighting
Yong Chen, Zhou Wenzhang · 2017
For noise removal, a fast image denoising algorithm based on local weighting is proposed in this paper. Firstly, the bilateral filter is used to estimate the image noise. Secondly, the noise of the original image is reduced by the improved noise model. Besides, the fundamental energy of the original image is obtained by using kernel anisotropic diffusion. Then, by setting up local window, the difference between the energy image and the reduced noise image in local window is calculated, and the energy difference between the parts of the energy image is obtained. The differences are used to form the corresponding weights. Thus, the noise removal is achieved by local weighting, with images enhanced and details recovered. The experimental results demonstrate that the proposed method not only has a good denoising effect, but also has a high processing speed.