测绘通报 ›› 2022, Vol. 0 ›› Issue (8): 128-132.doi: 10.13474/j.cnki.11-2246.2022.0244

• 技术交流 • 上一篇    下一篇

亚米级遥感影像的建筑物高度反演方法及其验证

刘晓忠1,2,3, 冯存均1,2,3, 邓小渊1,2,3, 詹远增1,2,3, 于辉1,2,3, 王兴坤1,2,3   

  1. 1. 浙江省测绘科学技术研究院, 浙江 杭州 330010;
    2. 自然资源浙江省卫星应用技术中心, 浙江 杭州 330010;
    3. 自然资源部地理国情监测重点实验室, 浙江 杭州 330010
  • 收稿日期:2021-10-26 发布日期:2022-09-01
  • 通讯作者: 邓小渊。E-mail:dxy541@126.com
  • 作者简介:刘晓忠(1974-),男,高级工程师,主要从事卫星遥感应用技术研究工作。E-mail:gisleo@163.com
  • 基金资助:
    浙江省自然资源厅科技项目(2021-56);浙江省测绘科学技术研究院重大科技攻关项目(DFP2021A0701)

Building height extraction method based on sub-meter remote sensing image and its verification

LIU Xiaozhong1,2,3, FENG Cunjun1,2,3, DENG Xiaoyuan1,2,3, ZHAN Yuanzeng1,2,3, YU Hui1,2,3, WANG Xingkun1,2,3   

  1. 1. Zhejiang Academy of Surveying and Mapping Science and Technology, Hangzhou 330010, China;
    2. Zhejiang Application Center of Nature Resources Satellite Technology, Hangzhou 330010, China;
    3. Key Laboratory of National Geographic Census and Monitoring, MNR, Hangzhou 330010, China
  • Received:2021-10-26 Published:2022-09-01

摘要: 建筑物高度是现代化都市监测、规划、管理及各城市经济活动中的基础性数据,为实现建筑物高度信息的提取,本文提出了一种基于玻尔兹曼曲线的建筑物高度反演方法。首先,利用建筑物影像的光谱特性,采用多尺度分割和遥感指数分类的办法获取建筑物阴影感兴趣区域,根据玻尔兹曼曲线函数拟合获取阴影的亚像素位置,线性拟合得到阴影边界;然后,根据太阳、卫星、建筑物和其阴影的几何关系,构建高度反演模型,估算建筑物高度;最后,选择宁海为研究区,选取在轨的主流亚米级高分二号、高景一号、北京二号、WorldView-2卫星遥感数据进行精度验证。试验结果表明,计算的建筑物高度中误差优于2.5m,可用于一般的城市卫星遥感监测。

关键词: 亚米级遥感影像, 玻尔兹曼曲线, 亚像素阴影提取, 建筑物高度反演与验证

Abstract: Building height is the basic reference data in modern urban monitoring, planning, management and economic activities. In order to realize the extraction of building height information, this paper presents a method for inversion of building height based on Boltzmann curve. First of all, through the spectral characteristics of the building image, the method of multi-scale segmentation and remote sensing index classification are used to obtain the interesting area of the building shadow, then we obtain the sub-pixel position of the shadow according to the Boltzmann curve function in order to get the shadow boundary by linear fitting. After that, a height inversion model based on the geometric relationship between the sun, satellites, buildings and their shadows is built. Finally, Ninghai county is selected as the experimental area, and the main satellite remote sensing data of GF-2, GJ-1, Beijing-2 and WorldView-2 which spatial resolution are better than 1m on orbit for accuracy verification. The experimental results show that the medium error in the calculated building height is better than 2.5meters, which can be used for urban satellite remote sensing monitoring.

Key words: sub-meter remote sensing image, Boltzmann curve, sub-pixel shadow extraction, building height extraction and verification

中图分类号: