摘要
针对传统模糊C均值(Fuzzy C-Means,FCM)算法受初始聚类中心影响而陷入局部最优的问题,提出了一种基于遗传算法(Genetic Algorithm,GA)的改进模糊C均值聚类算法。主要通过差值法获得图像的差异图,利用主成分分析(Principal Component Analysis,PCA)法提取变化影像的主要特征信息,利用遗传算法(GA)群体搜索的优点对传统的模糊C均值算法进行改进。对遥感图像的变化检测实验表明,改进的聚类算法克服了传统算法的缺点,在保留图像细节特征的前提下能有效提高检测精度,相比其他几种常见的聚类算法更有优势。
Aiming at the problem that the traditional Fuzzy C⁃Means(FCM)algorithm falls into local optimization due to the influence of the initial clustering center,an improved Fuzzy C⁃Means clustering algorithm based on Genetic Algorithm(GA)is proposed.The difference map of the image is obtained by the difference method,the main feature information of the changed image is extracted by the Principal ComponentAnalysis(PCA),andthetraditionalFuzzyC⁃Meansalgorithm isimprovedbyusingthe advantages of Genetic Algorithm(GA)population search.Through the change detection experiment of remote sensing image,it shows that the improved clustering algorithm overcomes the shortcomings of the traditional algorithm,can effectively improve the detection accuracy on the premise of retaining the detailed features of the image,and has more advantages than several other common clustering algorithms.
作者
赵东波
李辉
ZHAO Dongbo;LI Hui(School of Electronic Information,Xi’an Aeronautical University,Xi’an 710077,China;School of Electronic Information,Northwestern Polytechnical University,Xi’an 710129,China)
出处
《电子设计工程》
2023年第9期156-160,共5页
Electronic Design Engineering
关键词
变化检测
模糊C均值
主成分分析
遗传算法
遥感图像
change detection
Fuzzy C⁃Means(FCM)
Principal Component Analysis(PCA)
Genetic Alg⁃orithm(GA)
remote sensing image