Fast and robust fuzzy active contours

摘要

This paper presents a new approach for the detection of image objects, based on active contours by exploiting techniques of curve evolution. The proposed model of active contours is based on the minimization of a fuzzy energy, which can be seen as a particular case of a minimal partition problem. Unlike to existing methods, the adopted fuzzy energy does not depend on the gradient of the image, but is related to the image color and spatial segments. The proposed fuzzy energy is used as the model motivation power evolving the active contour, which will stop on the desired object boundary. The fuzziness of the energy and the exploitation of both local gray-scale and spatial information provide a balanced technique with a strong ability to reject "weak", as well as, "strong" local minima. The proposed approach is faster comparing to existing active contour models, due to the fact that the fuzzy energy calculations are computed directly instead of solving the Euler-Lagrange equations. The theoretical properties and the obtained results show that the proposed fuzzy energy-based active contour provides good object detection robustness when compared to other state of the art snake methods based on the gradient or other kind of energies.