Image selective smoothing and denoising using modified nonlinear diffusion (MAD) approach

Abstract

The research work presented in this paper is concerned with the identification of an approach suitable for image smoothing and denoising. The images often contain information at a number of different scales of resolution. We settle down this constraint, and specify the task of preserving important edges into the front-end processing, while keeping the properties of a scale-space. This is possible by the nonlinear scale-spaces. The objective of nonlinear diffusion filtering is to reduce smoothing in the presence of edges. An attempt is made to retain all the properties of the original model and to enhance the performance at different scale of resolutions. It shows improved performance in the presence of noise. In the proposed method the edges are better preserved because diffusion is controlled by the gray level differences of diagonal neighbors in addition to four nearest neighbors using coupled PDE formulation. The experiment is performed on Ultrasound images, the proposed Modified Anisotropic Diffusion (MAD) algorithm found very useful for denoising and enhancement the ultrasound biomedical images. (5 pages)