基于IR-MAD的高分辨率遥感影像交叉融合变化检测方法  

Cross Detection Method of High Resolution Remote Sensing Images Based on IR-MAD

作　　者：欧阳松南[1] OUYANG Songnan(Dongguan Geographic Information and Planning Research Center,Dongguan Guangdong 510805,China)

机构地区：[1]东莞市地理信息与规划编制研究中心

出　　处：《北京测绘》2019年第9期1029-1033,共5页Beijing Surveying and Mapping

摘　　要：高分辨率遥感影像数据在带来丰富地物信息的同时,也对变化检测提出了新的问题和挑战。本文从高分辨率遥感影像数据的空间结构特征和光谱分布信息入手,针对多光谱数据交叉融合的变化检测方法展开研究。采用基于GSA(Gram-Schmidt Adaptive Pansharpening Algorithm)法对全色和多光谱影像数据处理,生成四幅交叉融合影像;将迭代加权多元变化检测(Iteratively Regularized Multivariate Alteration Detection,IR-MAD)算法应用于高分辨率融合影像,提取变化信息。结果表明,本研究方法能够有效提取变化信息,并降低数据配准不一致所造成的误检测。High resolution remote sensing image data brings not only rich information of terrain,but also new problems and challenges to change detection.Starting with the spatial structure characteristics and spectral distribution information of high resolution remote sensing image data,this paper studies the change detection method based on cross-fusion of hyperspectral data.Four cross-fused images are generated by using block-based GSA(Gram-Schmidt Adaptive Panshar pening Algorithm)and dimensionality reduction method,and the Iterative Regularized Multivariate Alteration Detection(IR-MAD)algorithm is applied to high-resolution fusion images to extract change information.The results show that this method can effectively extract change information and reduce the false detection caused by inconsistent data registration.

关 键 词：高分辨率 影像交叉融合 迭代加权多元变化检测(IR-MAD) 变化检测 

分 类 号：P237[天文地球—摄影测量与遥感]