Image Denoising Using Graph-Based Frequency Domain Low-pass Filtering
Meijuan Liu, Ying Wei · 2019
Image denoising is always a very challenging task in image processing, which requires appropriate signal priori to regulate the problem. In recent years, the development of spectrum theory and graph signal has given us a new direction. By taking the image as a graph signal, we can construct a suitable underlying image to make the image relatively smooth, and combine the signal priori to image denoising. Based on this idea, and combined with graph Laplacian matrix,a method of image low-pass filtering in frequency domain is proposed in this paper. Experiments show that the proposed method is effective in image denoising, and superior to the traditional Wiener filtering and Gaussian filtering.