一种城市高大地物阴影检测方法

Shadow detection method of city large objects based on WorldView-2 image

针对水体、偏蓝色地物会影响阴影检测精度这一问题,该文提出了一种基于WorldView-2影像的城市高大地物阴影检测方法。首先,在统计分析阴影及典型地物光谱值的基础上,对C1C2C3色彩不变空间中的C3分量利用直方图阈值法分割阴影区域和非阴影区域,分离后的阴影区域中包含了绿色植被信息;然后,利用多波段运算提取植被信息;其次,对最初的阴影植被混合区域和植被区域做逻辑差运算,得到目标阴影区域;最后,对阴影提取结果进行小区域去除和形态学滤波等后处理。实验结果表明,该方法针对WorldView-2影像具有普适性,能够有效避免水体、偏蓝色地物的影响,能够精确、快速地检测阴影区域。

In view of the problem that water and bluish surface features can effect the accuracy of shad- ow detection, a shadow detection method based on WorldView-2 multispectral remote sensing images was proposed. Firstly, shaded areas and non-shaded areas were divided by histogram threshold method for C3 component in C1C2C3 invariant color space, and the shaded areas contained green vegetation information; secondly, the vegetation information was extracted by using multi-band operation based on counting and analysing shadow and spectral values of typical features; thirdly, the target shaded area was obtained based on difference computation between the zones containing shadow and vegetation area; finally, making post-processing for shadow extraction results, such as removal of small areas and morphological filte- ring. The results showed that this method was suitable for WorldView-2 images, which could accurately and quickly detect the shaded area without the influence of water and bluish surface features.