摘要
在目前经典的变化检测算法中,后验概率空间变化向量分析(CVAPS)方法广泛用于遥感影像的变化检测。然而,基于支持向量机(SVM)的CVAPS法无法有效处理高分一号影像中等分辨率遥感影像中的混合像元问题,且难以有效保证变化检测的精度。因此,本文通过引入空间信息,使用空间模糊C均值聚类(Spatial Fuzzy C Means, SFCM)有效地实现高分一号影像混合像元的分解,并结合简单贝叶斯网络(SBN),提出一种新的后验概率空间变化向量分析法SFCM-SBN-CVAPS。实验结果表明,本文算法的总体精度和Kappa系数均高于基于普通模糊C均值聚类(Fuzzy C Means, FCM)的CVAPS算法,且耗时更短,本文所提出的算法有助于提高遥感影像变化检测的精度和效率。
Among the current classical change detection algorithms,the change vector analysis in posterior probability space(CVAPS)method is widely used in the change detection of remote sensing images.However,the CVAPS method based on support vector machine(SVM)can not effectively deal with the mixed pixel problem in GF-1 medium resolution remote sensing image,and it is difficult to guarantee the accuracy of change detection effectively.Therefore,by introducing spatial information and using spatial fuzzy c-means clustering(SFCM),this paper effectively decomposes the mixed pixels of GF-1 imagery and proposes a new SFCM-SBN-CVAPS method combined with a simple Bayesian network(SBN).The experimental results show that this algorithm's overall accuracy and Kappa coefficient are higher than the CVAPS algorithm based on regular fuzzy c-means clustering(FCM)with a shorter time.Experiments show that the proposed algorithm is helpful to improve the accuracy and efficiency of remote sensing image change detection.
作者
杨洋
李轶鲲
杨树文
宋嘉鑫
YANG Yang;LI Yikun;YANG Shuwen;SONG Jiaxin(Faculty of Geomatics,Lanzhou Jiaotong University,Lanzhou 730070,China;National-Local Joint Engineering Research Center of Technologies and Applications for National Geographic State Monitoring,Lanzhou 730070,China;Gansu Provincial Engineering Laboratory for National Geographic State Monitoring,Lanzhou 730070,China)
出处
《测绘与空间地理信息》
2023年第4期34-37,42,共5页
Geomatics & Spatial Information Technology
基金
国家自然科学基金项目——基于高分辨率卫星影像的彩钢板建筑与城市空间结构演变关系研究(41761082)
国家重点研发计划(地球观测与导航)项目——星空地遥感立体监测技术(2017YFB0504201)
兰州交通大学(201806)优秀平台资助。
关键词
遥感影像变化检测
空间模糊C均值聚类
模糊C均值聚类
简单贝叶斯网络
后验概率空间变化向量分析
remote sensing image change detection
spatial fuzzy c-means clustering
fuzzy c-means clustering
simple Bayesian network
change vector analysis in posterior probability space
