| 基于Sentinel-1和DEM数据的南岭高植被覆盖区地形线性特征提取方法 |
| |
| 引用本文： | 鲁泽恩,田玉刚,柳庆威,王晶,朱宁,徐亚东.基于Sentinel-1和DEM数据的南岭高植被覆盖区地形线性特征提取方法[J].地球科学,2021,46(4):1349-1358. |
| |
| 作者姓名： | 鲁泽恩 田玉刚 柳庆威 王晶 朱宁 徐亚东 |
| |
| 作者单位： | 中国地质大学地理与信息工程学院,湖北武汉430074;中国地质大学地球科学学院,湖北武汉430074;流域关键带演化湖北省重点实验室,湖北武汉430074 |
| |
| 基金项目： | 中国地质调查局地质调查项目 |
| |
| 摘 要： | 在南岭高海拔山脉、植被茂密等地区开展地质调查难度大，人员和物资难以进入，因此需提前根据该区域的地形线性特征合理设计野外调查路线；此外地形线性特征可为区域构造运动信息的认知提供辅助知识.遥感技术以其宏观性、多尺度、多层次的特点成为地质研究和地质勘查的重要手段，然而在高植被覆盖区，常见的光学影像难以穿透植被，而微波雷达遥感则因为较好的植被穿透性，使得其应用成为可能.因此，提出了一种基于雷达卫星Sentinel-1和数字高程模型（digital elevation model，DEM）数据的地形线性特征提取方法.该方法首先利用似然比边缘检测算法提取Sentinel-1影像中的边缘特征，然后对DEM进行线性特征增强以生成山体阴影影像，再利用Canny边缘检测算子提取DEM数据中的山脊线与山谷线等主要线性特征，以DEM提取出的特征线为中心建立缓冲区并与雷达影像的提取结果做相交处理，将得到的线性特征利用道格拉斯-普克算法进行局部直线拟合，最后得到研究区的地形线性特征.结果表明，该方法综合考虑了雷达影像的微观细节信息和DEM数据的宏观趋势信息，在保留主要线性地形特征的同时剔除了伪边缘和噪声点，提取效果较好.
|
| 关 键 词： | 雷达卫星 数字高程模型 南岭 高植被覆盖区 线性特征 遥感 |
| 收稿时间： | 2020-10-06 |
Topographical Linear Feature Extraction Method Based on Sentinel-1 and DEM in Areas with High Vegetation Coverage of Nanling |
| |
| Lu Zeen,Tian Yugang,Liu Qingwei,Wang Jing,Zhu Ning,Xu Yadong.Topographical Linear Feature Extraction Method Based on Sentinel-1 and DEM in Areas with High Vegetation Coverage of Nanling[J].Earth Science-Journal of China University of Geosciences,2021,46(4):1349-1358. |
| |
| Authors: | Lu Zeen Tian Yugang Liu Qingwei Wang Jing Zhu Ning Xu Yadong |
| |
| Abstract: | Geological surveying is challenging in high-altitude mountains with dense vegetation in Nanling because it is inaccessible both for human and material resources. It is necessary to design reasonable field survey routes in advance according to the topographical linear characteristics of the area. Moreover, the topographical linear characteristics can provide the auxiliary information for local tectonic movement. Remotely sensed technology has become an important method for geological research and geological survey due to its macroscopic, multi-scale and multi-level characteristics. However, in areas with high vegetation coverage, optical remote sensing images are difficult to penetrate vegetation, while microwave radar images can do because of its good vegetation penetration. Therefore, the topographical linear feature extraction method based on Sentinel-1 (radar satellite) and DEM data in areas with high vegetation coverage is proposed. This method first uses the likelihood ratio edge detection algorithm to extract the edge features in Sentinel-1 images. Then, the linear features of DEM are enhanced to generate mountain shadow images, and then the main linear features such as ridge lines and valley lines are extracted by Canny edge detection operator. Finally, taking the feature line extracted from DEM as the center, a buffer zone is established and intersected with the result from radar images, and then the local straight line fitting is done by using the Douglas-Puck algorithm to obtain the final topographical linear features in the study area. The results show that the proposed method comprehensively considers the micro-detail information of radar images and the macro-trend information of DEM data, and can remove false edges and noise points while retaining the main topographical linear features, and the extraction effect is good. |
| |
| Keywords: | |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《地球科学》浏览原始摘要信息 |
|
点击此处可从《地球科学》下载全文 |
|
