An improved nonlinear diffusion algorithm for image denoising

摘要

The original nonlinear diffusion algorithm for image denoising is proposed by Perona and Malik in 1990. This method diffuse in different directions along the image gradient, but it can not make greatly well on the edge. So with the features of wavelet transform which can save the detail information of an image, an improved nonlinear diffusion algorithm is proposed. In this method, we construct the diffusion coefficient with wavelet coefficients which can reduce image noise without blurring edge. The computer simulations show that the proposed method has better edge protection and excellent denoising performance.