首都师范大学学报（自然科学版）2018，Vol.39Issue(3)：85-90,6.
一种多特征自适应组合的遥感影像变化检测方法
A Change Detection Method for Remote Sensing Images Based on Adaptive Multi-Feature Fusion
摘要
Abstract
Traditional change detection for remote sensing image mainly uses the spectral information of the image without considering the shape, texture and other characteristics of image information, so the accuracy of detection result is limited. Aiming at change detection of multi-temporal remote sensing images, this paper studies a change detection method for remote sensing images based on adaptive multi-feature fusion. By selecting the spectral and texture features of the image as the representative, the automatic adjustment analysis of the feature weight is carried out, and finally the accurate change detection of the remote sensing image is achieved. At last, experiments are carried out with WorldView-2 satellite images and the results show that this method can adjust the weights of the two features well and improve the accuracy and integrity of the detection results. The correct rate is 87.79％.关键词
变化检测/多特征组合/面向像斑/自适应调节Key words
change detection/features combination/object-oriented/adaptive control分类
信息技术与安全科学引用本文复制引用
马红,刘浩..一种多特征自适应组合的遥感影像变化检测方法[J].首都师范大学学报（自然科学版）,2018,39(3):85-90,6.基金项目
重庆市社会事业与民生保障科技创新专项 (cstc2017shmsA30027) （cstc2017shmsA30027）
住房城乡建设部2015 年科学技术项目(2015-K8-009). （2015-K8-009）
