热红外与可见光数据的云边缘高度匹配

发布时间：2018-04-24 22:17

本文选题：云边缘高度 + 欧氏距离变换　；参考：《遥感学报》2016年06期

【摘要】：高分辨率数据中云高度的差异性突显,特别是边缘处高度在云阴影识别和地表辐射估算等方面成为需要考虑的重要因素。热红外数据获取云高度分辨率较低、缺乏细部差异性特征,为解决这一问题,首先将对应的热红外和可见光数据进行特征点配准,再将基于热红外数据计算的云高度重采样至高分辨率,然后以基于欧式距离变换的围线搜索方法及距离加权将热红外云边缘高度匹配至对应的可见光图像,最后根据云阴影的相似度匹配方法确定真实云高度。结果表明,算法在遵循热红外云高信息分布变化规律的同时,可以得到较准确的高分辨率云边缘高度,一定程度上解决了热红外技术获取云高在分辨率上的局限,扩展了其在云高反演方面的作用。
[Abstract]:The difference of cloud height in high resolution data is prominent, especially the height at the edge becomes an important factor to be considered in cloud shadow recognition and surface radiation estimation. Thermal infrared data acquisition cloud height resolution is low, lack of detail difference feature, in order to solve this problem, the corresponding thermal infrared and visible light data feature points registration, Then the cloud height calculated by thermal infrared data is resampled to high resolution, and then the edge height of the thermal infrared cloud is matched to the corresponding visible light image by the circumferential search method based on Euclidean distance transform and the distance weighting method. Finally, the real cloud height is determined according to the similarity matching method of cloud shadow. The results show that the algorithm not only obeys the rule of thermal infrared cloud height distribution, but also can get accurate cloud edge height with high resolution. To a certain extent, the limitation of thermal infrared technology to obtain cloud height in resolution can be solved. Its role in cloud height inversion is extended.
【作者单位】：南京信息工程大学应用气象学院;中国科学院遥感与数字地球研究所遥感科学国家重点实验室;
【基金】：中国科学院重点部署项目(编号:KZZD-EW-TZ-18) 国家自然科学基金(编号:41371360)~~
【分类号】：TP391.41

【相似文献】