基于粒子群优化算法和改进的Snake模型的图像分割算法

Image segmentation algorithm based on the PSO and improved Snake model

基于活动轮廓(Snake)模型的目标轮廓提取是图像分割中一种重要的方法.为了克服传统Snake模型在图像分割中不能向凹处收敛和收敛不准确的缺点,提出了一种粒子群优化算法与改进的Snake模型相结合的图像分割算法.改进的Snake模型,即在传统的Snake模型的基础上增加了一个向心能量,增加此能量可以使初始化曲线向目标的凹处收敛.又由于粒子群优化算法具有获得全局最优的能力,可以使曲线能更准确地收敛到目标的边界.通过实验证明此方法可以取得很好的分割效果.

Getting the contour of an object according to the Snake model is an important method in the image segmentation. Because traditional Snake model cannot reach the concave of the object and the result of convergence is not accurate, an image segmentation algorithm based on the PSO and improved Snake model is proposed. The improved Snake model is generated by adding centripetal energy to traditional Snake model. The curve can reach the concave of the object because of the centripetal energy. Because the PSO has the ability of getting the global optimization, the curve can exactly reach the edge of the object. It is proved by experiment that preferable image segmentation result is gotten based on the algorithm.