摘要
针对阴影提取过程中水体等地物会影响阴影检测精度的问题,根据高分辨率遥感影像在YC_(b)C_(r)空间阴影区域的性质,提出了一种顾及水体的结合YC_(b)C_(r)空间与主成分分析(PCA)的遥感阴影检测方法。首先需判断是否有水体,若没有,则利用PCA与经过伽马校正的YC_(b)C_(r)空间变换相结合的阴影检测指数(SII)进行阴影检测;若有,则在此基础上利用水体指数NDWI对融合影像进行水体检测以及小斑块去除后处理,去除水体提取阴影。实验结果表明,该方法检测的正确率高于90%,比NDUI、C1C2C3、SI阴影检测方法更准确,同时能去除水体等对阴影的影响,具有更好的普适性。
In the process of shadow extraction,water and other objects will affect the shadow detection accuracy.According to the nature of shad-ow area of high-resolution remote sensing images in YC_(b)C_(r) space,we proposed a remote sensing shadow detection method combining YC_(b)C_(r) space and principal component analysis(PCA)considering water.The first thing to determine was whether there was a body of water.If not,we used a shadow detection index SII formed by PCA and gamma-corrected YC_(b)C_(r) space conversion to detect shadow.If there was,on the basis of this,we used the water body index NDWI and the small patch removal to perform logical difference processing,to remove the water body extrac-tion shadow.The experimental results show that the correct extraction number of this method is higher than 90%.Compared with the NDUI,C1C2C3,SI shadow detection methods,it has higher accuracy,can remove the influence of water and other shadows,and has better performance.
作者
谢文韬
王书涵
雷宇斌
龙城仕
刘华光
刘正才
XIE Wentao;WANG Shuhan;LEI Yubin;LONG Chengshi;LIU Huaguang;LIU Zhengcai(School of Civil Engineering,Xiangtan University,Xiangtan 411100,China;National and Local Joint Engineering Laboratory of Geospatial Information Technology,Hunan University of Science and Technology,Xiangtan 411201,China;Hunan First Institute of Surveying and Mapping,Changsha 410004,China)
出处
《地理空间信息》
2024年第3期20-23,共4页
Geospatial Information
基金
湖南省自然科学基金资助项目(2020JJ5164)。
