Image Denoising by 2-D Anisotropic Wavelet Diffusion
There are two popular methods for image denoising : nonlin ear diffusion and wavelet shrinkage. Recently they were proven equivalent and combined to generate a new way of image denoising: wavelet diffu sion. But this equivalence is restricted to some specific case and not for a 2-D anisotropic case. In this paper, we prove that a discrete computa tion step of 2-D anisotropic nonlinear diffusion is equivalent to a sequence of the DWT decomposition, high-frequency subbands regularization, and reconstruction. So we can apply different nonlinear diffusivity function to different wavelet coefficients to guide the processes to retain useful data and suppress noises. This makes wavelet diffusion more widely usable. Experimental results show our method is effective in image denoising.
image denoising discrete wavelet transform nonlinear diffusion wavelet diffusion
Chenglin Mao Hong Shen
Department of Computer Science,University of Science and Technology of China,China Department of Computer Science,University of Science and Technology of China,China School of Compute
国际会议
南宁
英文
83-95
2009-12-04（万方平台首次上网日期，不代表论文的发表时间）