一种基于主动轮廓模型的蚁群图像分割算法

Ant colony optimization with active contour models for image segmentation

中文关键词  主动轮廓模型  蚁群优化算法  图像分割

英文关键词  active contour models  ant colony optimization  image segmentation

中文摘要

      通过对主动轮廓模型进行图像分割的过程研究发现,其多阶段决策问题与蚁群算法的决策过程非常相似．文中根据主动轮廓模型的特点构建了一类新的蚁群求解算法,把图像分割问题转化成最优路径的搜索问题,为获取精确的图像轮廓提供了新方法．证明了该方法以概率1收敛到最优解,即可以在能量函数的约束下找到最好的边界．本方法还可以推广到其他主动轮廓模型的图像分割问题中．仿真结果表明,本文提出的分割方法比文献中的遗传算法更为有效．

英文摘要

      It is found that the multistage decision algorithm is similar to ant colony optimization (ACO) for image segmentation with active contour models (ACM). A new algorithm based on ACM is proposed in the paper, which converts the problem of image segment to a problem of searching for the best path in a constrained region and thus provides a new approach to obtain precise contour, The algorithm is then proved to be convergent with probability one, and will reach the best feasible boundary with minimum energy function value. Moreover, this algorithm can also be used to solve other mutational ACM problems. The simulation results show that the proposed approach is more effective than the genetic algorithm in literature.