Improved image denoising method based on Curvelet transform

An improved image denoising method based on Curvelet transform is proposed in this article. Considering characteristic of Wavelet transform and traditional Curvelet transform, this method divides image up via window neighbourhood processing. It fuses two denoising images based Wavelet and Curvelet transform according to segmentation information and obtains higher quality and better effect image. Experiments show that with noise variance, the method in this paper can not only obtain higher PSNR but also keep more edge details. And it reduces “nick” and “ring” of restoration images.