An active contour model for image segmentation based on elastic interaction

The task of image segmentation is to partition an image into non-overlapping regions based on intensity or textural information. The active contour methods provide an effective way for segmentation, in which the boundaries of the objects are detected by evolving curves. In this paper, we propose a new edge-based active contour method, which uses a long-range and orientation-dependent interaction between image boundaries and the moving curves while maintaining the edge fidelity. As a result, this method has a large capture range, and is able to detect sharp features of the images. The velocity field for the moving curves generated by this elastic interaction is calculated using the fast Fourier transform (FFT) method. Level set representation is used for the moving curves so that the topological changes during the evolution are handled automatically. This new method is derived based on the elastic interaction between line defects in solids (dislocations). Although it is derived originally for two dimensional segmentation, we also extend it to three dimensions. The features of the new method are examined by experiments on both synthetic images and medical images of blood vessels. Comparisons are made with the existing active contour methods.