Improved intensive restraint topology adaptive snake model
Sun Zheng · Guangdian gongcheng · 2007
An intensive restraint topology adaptive snake model is designed for extracting contours and skeletons from gray images and tracking dynamic objects from image sequences. Firstly, the original image and snake curve are discretized by orthogonal grids. The snake nodes are constrained to move only from one grid vertex to another along grid lines. Then, topological transformation of the model is implemented through vertex splitting. Topological collision is detected and eliminated automatically. The effectiveness of the algorithm is demonstrated on simulated images and clinical medical images. Experimental results show that the proposed model is efficient in object capturing and topological transformation. The computation cost is also rather lower than that of original topology adaptive snake.