基于参数优化改进型可能聚类的遥感图像分割  

Remote Sensing Image Segmentation Using Improved Possibilistic Clustering Algorithm with Optimized Parameters

作　　者：武斌[1,2] 黄庆丰[1,2] 魏元春[1,2] 

机构地区：[1]安徽农业大学信息与计算机学院,安徽合肥230036 [2]安徽农业大学林学与园林学院,安徽合肥230036

出　　处：《农业网络信息》2011年第12期31-34,共4页Agriculture Network Information

摘　　要：可能聚类算法(PCA)和可能C-均值聚类算法(PCM)对初始值非常敏感,容易产生一致性聚类。改进型可能C-均值聚类算法(IPCM)能解决PCM的问题,然而IPCM的执行更依赖于参数。IPCM必须计算参数两次,因此聚类时间长。为了克服PCA和IPCM的缺点,进而应用于复杂的遥感图像分割,将PCA和IPCM相结合,提出了一种基于参数优化的改进型可能聚类算法(IPCAOP)。实验表明,IPCAOP在处理遥感图像分割方面明显优于模糊C-均值聚类(FCM)和IPCM。Possibilistic clustering algorithm（PCA） and possibilistic c-means clustering（PCM） were very sensitive to initialization value which had undesirable tendency to produce coincident clusters.An improved possibilistic c-means（IPCM） algorithm was proposed to solve the problems of PCM.However,the performance of IPCM depended heavily on the parameters.IPCM must compute the parameters twice,so it was time-consuming.In order to solve the problems of PCA and IPCM,an improved possibilistic clustering algorithm with optimized parameters（IPCAOP） was proposed to combine PCA and IPCM.Our experimental results showed that compared with fuzzy c-means clustering（FCM） and IPCM,the proposed algorithm was better in remote sensing image segmentation.

关 键 词：模糊C-均值聚类 可能性模糊C-均值聚类 改进的可能性模糊C-均值聚类 遥感图像 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]