摘要
随着调查监测技术的进步,广东省自然资源常态化监测工作汇聚了多种感知数据类型,包括遥感影像、监控视频、无人机照片、移动终端照片等。为充分利用这些数据,本文依托多视图立体匹配方法,深入探讨利用多源感知数据自动构建三维地理场景的方法,并从调查监测的角度出发,对成果的可读性和实用性进行实证分析。通过对比不同数据源单独建模与本文方法建模成果,充分证实了本文所提方法的有效性,实现了数据成果的再利用,并探索出一种新的三维地理场景构建方法,推动常态化监测工作成果从二维向三维转变。
With the advancement of investigation and monitoring technology,the normalized monitoring of natural resources in Guangdong Province has gathered various types of perception data,including remote sensing images,surveillance videos,drone photos,and mobile terminal photos.To fully utilize these data,this article relied on the multi-view stereoscopic matching method to deeply explore the method of automatically constructing three-dimensional(3D)geographic scenes by using multi-source perception data.From the perspective of investigation and monitoring,the readability and practicality of the results were empirically analyzed.By comparing the modeling results of different data sources separately with the method proposed in this paper,the effectiveness of the proposed method was fully confirmed.The reuse of data results was achieved,and a new method for constructing 3D geographic scenes was explored,promoting the transformation of normalized monitoring work from two-dimensional(2D)scenes to 3D scenes.
作者
吴晓雯
易雅琴
林东铨
WU Xiaowen;YI Yaqin;LIN Dongquan(Surveying and Mapping Institute Lands and Resource Department of Guangdong Province,Guangzhou,Guangdong 510663,China;Key Laboratory of Natural Resources Monitoring in Tropical and Subtropical Area of South China,Ministry of Natural Resources,Guangzhou,Guangdong 510663,China;Guangdong Science and Technology Collaborative Innovation Center for Natural Resources,Guangzhou,Guangdong 510663,China;Guangdong Geographical Science Data Center,Guangzhou,Guangdong 510663,China)
出处
《北京测绘》
2024年第9期1300-1305,共6页
Beijing Surveying and Mapping
基金
广东省科技计划(2021B1212100003)
广东省自然资源厅科技项目(GDZRZYKJ2023001)。
关键词
调查监测
多视图几何
三维地理场景
物联感知数据
多源数据
investigation and monitoring
multi-view geometry
three-dimensional(3D)geographic scene
perception data of internet of things
multi-source data
