Unifying variational approach and region growing segmentation

International audienceRegion growing is one of the most popular image segmentationmethods. The algorithm for region growing is easilyunderstandable but criticized for its lack of theoretical background.In order to overcome this weakness, we propose todescribe region growing in a new framework using a variationalapproach that we called Variational Region Growing(VRG). Variational approach is commonly used in imagesegmentation methods such as active contours or levelsets, but is rather original in the context of region growing. Itrelies on an evolution equation derived from an energy minimization,that drives the evolving region towards the targetedsolution. Here, the energy minimization and the VRG robustnessto the initial seeds location are performed on gray-leveland color images.