基于Google Earth的建筑物高度信息自动提取研究  

Research on Automatic Extraction of Building Height Information Based on Google Earth

作　　者：夏凤莲 刘超[1,2,3] 刘春阳[1,2,3] 徐胜华[4] XIA Feng-lian;LIU Chao;LIU Chun-yang;XU Sheng-hua(School of Spatial Information and Geomatics Engineering,Anhui University of Science and Technology,Huainan 232001,China;Anhui Key Laboratory of Space-Ground Collaborative Monitoring and Early Warning of Mine Mining Disasters,Anhui University of Science and Technology,Huainan 232001,China;Engineering Research Center of Coal Industry for Collaborative Monitoring of Environment and Disaster in Mining Areas,Anhui University of Science and Technology,Huainan 232001,China;Chinese Academy of Surveying and Mapping,Beijing 100036,China)

机构地区：[1]安徽理工大学空间信息与测绘工程学院,安徽淮南232001 [2]安徽理工大学矿山采动灾害空天地协同监测与预警安徽普通高校重点实验室,安徽淮南232001 [3]安徽理工大学矿区环境与灾害协同监测煤炭行业工程研究中心,安徽淮南232001 [4]中国测绘科学研究院,北京100036

出　　处：《宜春学院学报》2024年第9期4-8,共5页Journal of Yichun University

基　　金：国家自然科学基金面上项目(编号:42071384);安徽省教育厅科学研究重点项目(编号:2022AH050849)。

摘　　要：针对现有建筑物高度获取方法成本较高、费时费力且时效性较差的问题。基于Google Earth免费遥感影像数据,本文提出一种自动化程度较高的建筑物高度提取方法。首先对原始遥感影像进行自适应伽马校正,增强图像对比度;使用K-Means无监督分类方法,对建筑物阴影进行提取;然后使用三角形绿度指数,对误提取为建筑物的植被像元进行剔除;最后进行后处理操作即可得到建筑物阴影提取结果。基于建筑物阴影提取结果与其高度之间的几何关系,利用典型已知的建筑物高度数据反推出相关参数,并根据自动提取的建筑物阴影长度求取建筑物高度信息。实验结果表明,本方法提取的建筑物高度的均方根误差为0.94 m,平均精度为97.55%。说明该方法能较为完整的提取建筑物阴影,准确的估算建筑物高度,同时自动化程度也得到了一定提高。Based on Google Earth free remote sensing image data,this article proposes a highly automated building height extraction method,aiming to solve the problem that the existing methods for obtaining the height of buildings are high cost,time-consuming and labor-intensive,and have poor timeliness.First,adaptive gamma correction is performed on the original remote sensing image to enhance the image contrast.Then the K-Means unsupervised classification method is used to extract the shadows of buildings,and the triangle greenness index is used to identify and eliminate plant pixels that are mistakenly extracted as buildings.Finally,perform post-processing operations to obtain the building shadow extraction results.Based on the geometric relationship between the building height and its shadow,the relevant parameters are deduced using typical known building height data,and the building height information is obtained based on the automatically extracted building shadow length.Experimental results show that the root mean square error of the building height extracted by this method is 0.94m,and the average accuracy is 97.55%.This method can completely extract the shadow of the building,accurately estimate the height of the building,and partly improve automation level.

关 键 词：建筑物高度 阴影提取 无监督分类 谷歌地球 

分 类 号：P237[天文地球—摄影测量与遥感]