Image denoising algorithm based on fast discrete curvelet transform
Yao Wang · jOURNAL OF southwest University for Nationalities · 2007
Curvelet transform has received more and more attention in recent years because it can better represent anisotropy of objects with discontinuities along edges, but the original digital curvelet transform is redundant. The Fast Discrete Curvelet Transform which has the optimally sparse representation of objects with edges is applied in this paper. Experiments show that the image denoising algorithm based on FDCT not only keeps the edge of image but also obtain better visual quality and yields denoised images with higer PSNR.