An Implicit Contour Morphing Framework Applied to Computer-Aided Severe Weather Forecasting
Dominique Brunet, David M. L. Sills · IEEE Signal Processing Letters · 2015
We propose a contour morphing framework that allows large deformations and topological changes while guaranteeing non self-intersecting contours. Instead of explicitly matching contour points, the proposed algorithm combines an implicit contour representation with state-of-the-art deformable image registration. Synthetic and real-life examples from a meteorological application are presented to demonstrate the efficacy of the framework.