Fast and Robust Active Contours Model for Image Segmentation

Abstract

Image segmentation using local region-based active contour models can segment images with intensity inhomogeneity effectively, but their segmentation results are sensitive to the initialization and easy to get incorrect results when dealing with texture images. This paper presents a novel active contour model (ACM) for image segmentation. The proposed method adopts local kernel mapping to enhance the discriminative ability to delineate nonlinear decision boundaries between classes. In addition, we introduce a modified convex model and propose a fast evolving scheme accordingly to deal with the minimization of the model energy function. The proposed approach is validated by a comparative study over a large number of experiments on synthetic and real images. The experiments demonstrate that our method is more efficient and robust for segmenting different kinds of images compared with the state-of-the-art image segmentation methods.