Improved Approach for Snake Model Based on GHough and Dynamic Programming

Abstract

Snake models are used more and more widely in applications of image analysis and computer vision. There are two challenging issues for the application of snake models: how to select the initial contour of the object automatically and how to obtain good convergence result. This paper introduces an improved approach for snake model that integrates with GHough transformation and dynamic programming algorithm. In the paper GHough transformation is used to help selecting the initial contour without the interference manually, and the dynamic programming is used to improve the convergence precision. The experimental results shows that such approach can not only solve the problem of selecting initial contour automatically, but also get good convergence results.