基于边缘提取算法的岩层三角面产状半自动提取  

Semi-Automatic Extraction of Triangular Facet Attitude Based on Edge Extraction Algorithm

作　　者：林楠 徐遥辰 高博文 翁旭华 陈宁华[1] Lin Nan;Xu Yaochen;Gao Bowen;Weng Xuhua;Chen Ninghua(School of Earth Sciences,Zhejiang University,Hangzhou 310027,China)

机构地区：[1]浙江大学地球科学学院,浙江杭州310027

出　　处：《地球科学》2021年第10期3753-3763,共11页Earth Science

基　　金：国家自然科学基金项目(No.41372205)。

摘　　要：岩层三角面的准确识别和快速检测是利用遥感技术获取地表地层产状信息的有效途径.提出了利用高分辨率卫星遥感立体像对,通过自动迭代调整高斯分布模型方差的Canny边缘检测算法实现岩层三角面半自动检测和地层产状快速提取方法.以南天山吐格尔明背斜为实验区,使用Worldview-2立体像对构建数字高程模型和0.5 m分辨率的数字正射影像,利用岩层三角面检测算法提取实验区的岩层三角面和产状.结果表明,提出的方法客观高效,提取准确率达到90.2%,能够准确反映构造趋势的变化,为复杂构造带的地表建模和构造分析提供依据.Accurate identification and rapid detection of triangular facet is an effective way to use remote sensing technology to obtain stratum attitude information. In this paper, it proposes a method for semi-automatic detection of triangular facet and rapid extraction of bedding attitude. Using high-resolution satellite remote sensing stero image pairs, the method is based on Canny edge detection algorithm with an automatic iterative adjustment of Gaussian distribution model variance. Taking Tugerming anticline in the South Tianshan Mountains as the experimental area, a digital elevation model and digital orthophoto maps with the resolution of 0.5 m were constructed using Worldview-2 stereo pairs, and 81 triangular facets in the experimental area were extracted. The result shows that the method proposed in this paper is objective and efficient, and the extraction accuracy rate reaches 90.2%. The extracted attitude accurately reflect the changes in structure, providing a basis for surface modeling and structural analysis of complex structural belts.

关 键 词：产状提取 高分辨率遥感影像 立体像对 边缘提取算法 遥感 

分 类 号：P627[天文地球—地质矿产勘探]