Image enhancement using diffusion equations

In this paper, a new algorithm is proposed for reducing noise from images using diffusion equations. The proposed algorithm is based on the fact that the intensity of adjacent pixels in a real image may vary gradually. Thus, in numerical solution of diffusion equations, for preserving edges, neighbors longer than one pixel must be considered. In the presented method, in addition to vertical and horizontal edges, slant edges are also considered to better preserve edges. This algorithm has been examined using a variety of standard images, and its performance has been compared with the existing algorithms. Experimental results show that, in comparison with other existing methods, the proposed algorithm has a better performance in denoising and preserving image edges.