CAJ | 학술논문

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。
가능취류산법（PCA）화가능C-균치취류산법（PCM）대초시치비상민감,용역산생일치성취류。개진형가능C-균치취류산법（IPCM）능해결PCM적문제,연이IPCM적집행경의뢰우삼수。IPCM필수계산삼수량차,인차취류시간장。위료극복PCA화IPCM적결점,진이응용우복잡적요감도상분할,장PCA화IPCM상결합,제출료일충기우삼수우화적개진형가능취류산법（IPCAOP）。실험표명,IPCAOP재처리요감도상분할방면명현우우모호C-균치취류（FCM）화IPCM。
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.