Improved Edge Enhancing Diffusion Filter for Speckle-Corrupted Images

In the context of automatic image processing, noise is often the hardest problem to cope with. This is, in particular, true for images corrupted by multiplicative speckle noise, such as synthetic aperture radar images. To tackle such an issue, this letter proposes a novel anisotropic diffusion filter that manages to simultaneously fulfill competing requirements: reduce noise on homogeneous regions, preserve weak edges, and keeping hard targets intact. The capabilities of our filter were proved by comparing it with other state-of-the-art filters on a 1-look Cosmo-SkyMed image.