Edge detection and image segmentation based on nonlinear anisotropic diffusion

Partial differential equations (PDEs) have led to an entire new field in image processing and computer vision. Typical PDE techniques for image smoothing and edge extraction regard the original image as initial state of a parabolic (diffusion-like) process, and extract filtered versions from its temporal evolution. The diffusion coefficient varies spatially in such a way that intra-region smoothing is preferred to inter-region smoothing, thus preventing edge blurring and edge dislocation. We review some useful applications of this particular diffusion process on noise reduction and edge detection. We also propose a pixel based clustering scheme for color image segmentation, which incorporates Perona-Malik diffusion filtering. Preliminary experimental results show a significant improvement in the segmentation results.