Abstract (EN)

The proposed method is related to parametric and geodesic active contours as well as minimal paths, in the context of image segmentation. Our geodesically linked active contour model consists in a set of vertices connected by paths of minimal cost. This makes up a closed piecewise defined curve, over which an edge or region energy functional is formulated. The greedy algorithm is used to move vertices towards a configuration minimizing the energy functional. This evolution technique ensures lower sensitivity to erroneous local minima than usual gradient descent of the energy. Our method intends to take advantage of explicit active contours, minimal paths and greedy evolution techniques.