Speckle removal by distance-driven anisotropic diffusion of SAR temporal stacks

Radar images are often collected over the same region over time. To provide smoothing of such imagery without effacing temporal changes in the scene, we put forth an anisotropic diffusion technique. Traditionally, with synthetic aperture radar images, the mean of time series is utilized to produce a single despeckled image that discards temporal information. In contrast, we propose a statistical approach designed to reduce speckle noise in each image. The new filter incorporates temporal information essential to detection of potential change events for transportation infrastructure. Results demonstrate the efficacy of the approach, showing lower mean squared error than leading methods.