An adaptive active contour model driven by weighted local and global image fitting constraints for image segmentation

This paper presents an adaptive active contour model (ACM) to handle both homogeneous and inhomogeneous images. The object function of the presented model is roughly composed of the local energy constraint and the global energy constraint. First, we choose the weighted local image fitting constraint as the local energy constraint term. Second, the weighted global image fitting constraint is defined as the global energy constraint term inspired by the weighted local image fitting constraint. Moreover, a monotone time-varying function within the range of zero to one is defined as the energy weight to regulate the proportions of the local and global energy constraints, which makes the curve evolution be divided into three stages. Some experiments are performed on synthetic images and real-world images, and the results indicate that the proposed model is superior to the popular ACMs in segmentation accuracy and efficiency and is more robust to the initial contours.