一种优化的可能性C-均值聚类图像分割算法

An optimized C-means clustering image segmentation algorithm

针对截集式可能性C-均值聚类算法没有考虑邻域差异性信息,导致该算法对于有些噪声图像分割效果不理想的问题,通过在典型性值的计算中加入邻域信息,对算法中重要的变量典型性值进行修正,提出了一种优化的可能性C-均值聚类图像分割算法。通过仿真实验验证了算法的有效性,结果表明,该算法在有噪的情况下可以将目标和背景清晰的分割分离,达到较好的分割效果。

For the cut-off-likelihood C-means clustering algorithm,it does not consider the difference information of the neighborhood,which leads to the problem that the algorithm is not ideal for some noisy image segmentation.The typical values of important variables are modified,and an optimized possibility C-means clustering image segmentation algorithm is proposed.The effectiveness of the algorithm is verified by simulation experiments.The results show that the algorithm can separate the target and the background clearly under noisy conditions,and achieve better segmentation results.