Robust Contour Tracking Model Using a Variational Level-Set Algorithm
Tudor Barbu · Numerical Functional Analysis and Optimization · 2013
In this work, we provide a novel variational level-set based object contour tracking approach. Thus, a mathematically rigorous variant of the Chan-Vese algorithm for image segmentation via geometric active contour model is proposed here. With respect to the original contour detection algorithm, the level set function ϕ(t) defining the evolving contour S t = {x; ϕ(t, x) = 0} is iteratively computed from a nonlinear parabolic boundary value problem that is well posed in the space of functions with bounded variations. We provide a robust mathematical justification of the proposed level-set model.