Image segmentation based on geodesic aided Chan-Vese model
Thi-Thao Tran, Van-Truong Pham, Yun-Jen Chiu, Kuo‐Kai Shyu · 2010
In this paper, a novel model for intensity inhomogeneous image segmentation is proposed. The proposed model uses the local information of the image to be segmented; concurrently, it incorporates the geodesic active contour (GAC) model into Chan-Vese (C-V) model in energy function. Thus, the proposed model is effective when dealing with intensity inhomogeneous images. Practical experiments prove that the proposed model can obtain exact segmented results, especially with the intensity inhomogeneous images even with hole, noise and complex background.