Image Processing in FEMLAB using Diffusion Filters

Zama, F.
Department of Mathematics, Bologna University, Bologna, Italy

PDE-based methods apply in a large variety of image processing and computer vision areas such as denoising, smoothing, segmentation and restoration. Aim of this work is to apply the primitives provided by FEMLAB to some well known PDE-based models used in image processing. This application can be utilized by students, approaching PDE-models in image processing, as well as by researchers who can find an easily extensible framework for more sophisticated evolutive models.