Image Denoising Based on Hybrid Fourier and Neighborhood Wavelet Coefficients

Image denoising is a classical topic and a difficult problem in the field of image processing. Fourier transform can effectively sparsely represent the smoothing texture part of the image, but can't effectively represent mutations in the image. The wavelet transform can sparsely represent sharply changing parts in the image, but can't effectively represent the texture and the slowly changing parts of the image. A new method based on hybrid Fourier and neighborhood wavelet coefficient is presented, and experiments show that the method is valid and can improve the visual effect, and that the objective indicator PSNR is better than NeightShrink method.