城市航空影像中基于颜色恒常性的阴影消除

基于颜色恒常性理论,对真彩色和彩红外城市航空影像中高大建筑物形成的阴影进行消除。首先采用光谱比技术和最大类间方差法(Otsu)阈值分割技术进行城市航空影像中建筑物阴影的检测,进而就颜色恒常计算的Shades of Gray算法中明可夫斯基范式(Minkowski norm)的p取不同值情况下的阴影去除效果进行实验,利用亮度、对比度及平均梯度值比较阴影去除效果的好坏。实验表明:在基于航空影像阴影区域及非阴影区域划分的基础上,本文方法比一般的阴影区反差拉伸方法效果好;且与一般场景影像的阴影去除不同,对两类航空影像,p取2时阴影去除效果最佳,说明这两类影像不能简单看成是一个灰色世界影像。

Removing shadows from urban aerial images based on color constancy

A method is used to remove high building shadows in true color and color infrared urban aerial images based on the theory of color constancy.The spectrum ratio and Otsu threshold segmentation methods are used to detect building shadows in true color and color infrared urban aerial images.Then, based on the shadow detection result,one of the color constancy algorithms shades of gray(SoG) is used to remove the shadows in aerial images with different p values of Minkowski norm.Finally, the shadow removal resul...