Abstract:

To solve the problem of weak robustness in image segmentation of general active contour, an active contour model combined region-based model with gradient-based model is proposed using the entropy energy and gradient sum. Firstly, active contour model is constructed via fitting the gradient information and region information by use of entropy energy and gradient of image. Secondly, the level set is embed in the mentioned model to follow the curve change of topological structure. Finally, the narrow band method is used to increase the speed of evolution. The experiment results show that the proposed model has a better robustness with faster evolution speed.The study of the model is able to provide a new research way to develop the theory of image segmentation.

Key words: image segmentation, entropy energy, level set, active contour model