An Improved Snake Model Based on Image Gravitation and Optimized Greedy Algorithm

A improved active contour model is presented in this paper, function of energy and method of numerical value overlap are improved. The problem of the active contour model´s sensitivity to its initial position and the poor convergence to boundary concavities which exists in the traditional Snake for curve evolution are solved because Image gravitation term and area term are added in the new model; Snake points rapidly converge near image edge and increased or decreased self-adapted by optimal greedy algorithm. So, The active contour model´s sensitivity to its initial position, the poor convergence to boundary concavities which exists in the traditional Snake and sensitivity to noise are solved, and the computational complexity decrease. Experimental results show the new model is efficient and precise.