集成特征分量的高分二号影像阴影检测

Shadow Detection of Integrated Characteristic Components for GF-2 Image

针对高分辨率遥感影像中阴影检测精度易受水体、植被等因素干扰的问题,通过分析高分二号影像中典型地物的光谱特征,构建了一种集成特征分量与面向对象分类相结合的阴影检测方法。构建的特征分量包括:主成分第一分量PC1、亮度分量I、归一化差分植被指数NDVI及水体指数WI。将各特征分量进行归一化处理,建立包含波段均值、标准差等特征的规则集,对影像的I和PC1分量进行多尺度分割,结合面向对象的方法进行阴影检测。选取不同区域遥感影像进行实验,实验结果表明:与传统基于像素的阴影提取方法相比,该方法提取出的阴影斑块完整,且能有效地减弱水体和植被的影响。

The shadow detection accuracy in the high-resolution remote sensing images is easily disturbed by water,vegetation and so on.This study proposed a shadow detection method based on object-oriented method and established characteristic components by analyzing the spectral characteristics of typical features in GF-2 satellite images.The following components were constructed to detect shadow information:first principal com⁃ponent(PC1),brightness component I,Normalized Difference Vegetation Index(NDVI)and Water Index(WI).And then,we normalized each characteristic component to establish a rule set containing features such as band mean,standard deviation.Brightness I and PC1 were chosen as the main data source for multi-resolution segmentation,at last,performed object-oriented method on the segmented images to detect shadow.Selected different areas of GF-2 images for the proposed method,and experimental results show that the proposed meth⁃od could extract complete shadow patches and effectively reduce the influence of water bodies and vegetation compared with pixel-based method.