高分辨率遥感影像中的云阴影去除

Cloud Shadow Removal in High-resolution Remote Sensing Images

文章提出了一种人机交互的半自动云阴影去除算法。首先通过人机交互半自动方式将图像分割为云区域、阴影区域、清晰地物区域以及过渡区域四个部分,然后通过将云阴影区域的直方图映射到清晰区域来对阴影区域灰度进行补偿,最后对边界进行后处理得到一幅过渡平滑的去除了云阴影的图像。实验结果表明,该方法对高分辨率遥感影像中的云阴影去除是有效的。

In this paper,a semi-automated approach for cloud shadow removal in high-resolution satellite images is proposed.The image is flint segmented into four parts： clouds regions,shadow regions,non-shadowed regions and transition regions through semi-automated interaction.Then the shadow regions are matched to the non-shadowed regions through histogram transformation.At last,post-processing on the shadow regions＇ edges is proposed in order to get a smooth result image with shadow removed.The experiment results show that this method is valid.