Active Contour Models

Active contour models have been widely applied to image segmentation and analysis. It has been successfully used in contour detection for object recognition, computer vision, computer graphics, and biomedical image processing such as X-ray, MRI and Ultrasound images. The energy-minimizing active contour models or snakes were developed by Kass, Witkin and Terzopoulos in 1987. Snakes are curves defined in the image domain that can move under the influence of internal forces within the curve itself and external forces derived from the image data. Snakes perform well on certain types of images (such as well-defined, convex shapes). There have been several improvements proposed to the original snake or active contour model. These improvements include balloon snakes, adaptive snakes, and GVF snakes. In this project, I reviewed and implemented their algorithms as well as the original snake model. GCBAC (Graph Cut Based Active Contour) is one of alternative solutions to the object extraction problem. Although the GCBAC belongs to family of active contour models, it differs fundamentally from original active contours. In this project, I also review and implement the GCBAC algorithm as well.