集成特征分量的高分辨率遥感影像建筑物阴影检测

Building Shadow Detection with Integrated Characteristic Components for High Resolution Remote Sensing Images

针对高分辨率遥感影像,提出了一种集成影像特征分量与建筑物阴影形态特征的面向对象建筑物阴影提取方法。通过分析遥感影像的光谱特性,构建了主成分第一分量PC1、绿波段分量G、过绿指数EXG及色调特征H,并采用伽马变换对H波段进行增强处理,对各特征分量DN值归一化处理并进行了综合分析,根据建筑物阴影形态特征设立面积及长宽比规则,构建了规则集,实现了面向对象的建筑物阴影信息提取。选取不同时段不同地区的遥感影像进行建筑物阴影提取试验,试验表明,本文方法不仅能够有效地削弱水体、植被、深色地物的影响,而且能够去除非建筑物阴影的干扰,所获得的建筑物阴影斑块完整,无破碎图斑。

An object-oriented building shadow extraction method is proposed for high-resolution remote sensing images,which integrates the images characteristic components and the shadow＇s morphological characteristics of buildings. By analyzing the spectral characteristics of remote sensing images,the first component（ PC1）,the green band component（ G）,the excess green（ EXG） and the hue characteristic（ H） were constructed,then we used the gamma transform to enhance the H band,and normalized the DN value of each characteristic component and made a comprehensive analysis.According to the morphological characteristics of building shadow,we established the area and aspect ratio rule and built a rule set.Thereby,the object-oriented building shadow information extraction was achieved.Finally,we selected remote sensing images from different regions and time intervals to extract building shadow. Experimental result showed that the proposed method can not only effectively reduce the influence of water bodies,vegetation and dark objects,but also remove the shadow of non-building structures,and then obtain integral building shadow patches without fragmentation.