Tracking of Dynamic Image Sequence Based on Intensive Restraint Topology Adaptive Snake

An approach for detecting and tracking dynamic objects with complex topology from image sequences based on intensive restraint topology adaptive snake mode is presented. Firstly, the image and snake are decomposed by orthogonal grid lines. The snake nodes are constrained to move only from one grid vertex to another along grid lines. Then, topological transformation is implemented through splitting snaxels (snake nodes). For tracking along the sequence, the automation is improved by reducing manual interaction to only select a point randomly inside the region of interest from the first frame. The algorithm was demonstrated with simulated image and clinical coronary angiographic sequences. Results validated the accuracy and efficiency.