测绘学报 ›› 2020, Vol. 49 ›› Issue (5): 632-643.doi: 10.11947/j.AGCS.2020.20190282

• 地图学与地理信息 • 上一篇    下一篇

三维GIS场景与多路视频融合的对象快速确定法

刘振东, 戴昭鑫, 李成名, 刘晓丽   

  1. 中国测绘科学研究院, 北京 100830
  • 收稿日期:2019-07-04 修回日期:2019-12-12 发布日期:2020-05-23
  • 通讯作者: 李成名 E-mail:cmli@casm.ac.cn
  • 作者简介:刘振东(1989-),男,硕士,助理研究员,研究方向为三维GIS、智能感知设备与三维场景融合、三维实景重建。E-mail:lzdgis@casm.ac.cn
  • 基金资助:
    国家自然科学基金(41871375);中国测绘科学研究院基本科研业务费(AR 1935;AR 1909;AR 1916;AR 1917)

A fast fusion object determination method for multi-path video and three-dimensional GIS scene

LIU Zhendong, DAI Zhaoxin, LI Chengming, LIU Xiaoli   

  1. Chinese Academy of Surveying and Mapping, Beijing 100830, China
  • Received:2019-07-04 Revised:2019-12-12 Published:2020-05-23
  • Supported by:
    The National Natural Science Foundation of China (No. 41871375);The Basal Research Fund of Chinese Academy of Surveying and Mapping (Nos. AR 1935;AR 1909;AR 1916;AR 1917)

摘要: 虚实融合技术是视频增强三维虚拟地理场景的关键环节之一,对提高真实沉浸式视觉体验具有重要作用。现有基于视频投影的虚实融合方法,在三维场景动态更新时,通常需要对用户视角下新场景内所有渲染对象进行遍历,来筛选出摄像头范围内需融合的对象,导致融合过程耗时长、效率不高。本文提出一种顾及拓扑关系的多路视频与三维GIS场景融合的对象快速确定方法。首先对三维场景视锥体内视频对象集渲染对象动态建立拓扑网络模型;然后基于拓扑网络模型建立视频与渲染对象的拓扑关系,完成融合对象的快速确定及融合渲染;最后利用滕州市三维场景进行了验证。试验结果表明,本文提出的方法在视频与三维GIS场景融合时,整体耗时少、效率高,极大地降低了融合的计算代价,具有良好的可行性及稳健性。

关键词: 监控视频, 拓扑信息, 融合对象, 三维GIS场景, 动态更新

Abstract: Virtual and real fusion technology is one of the key links in video enhancement of 3D geographical scene, which plays an important role in improving the real immersive visual experience. The existing video projection fusion method usually needs to traverse all the rendered objects in the new scene when the 3D scene is updated dynamically, in order to determine the objects that need to be fused in the camera range, which resulting in a long fusion process and low efficiency. A fast fusion method of multi-channel video and 3D GIS scene considering topological relationship is proposed. Firstly, a dynamic topology network model was established for video objects and rendering objects to accelerate the judgment of spatial relationship. Secondly, the topology relationship between video and rendering objects within 3D scene was established based on the topology network model, and the recorded topological relationship is used to determine and fuse the object quickly. Finally, the three-dimensional scene of Tengzhou City is verified. The experimental results show that the proposed method has the advantages of less time-consuming, high efficiency, greatly reducing the computational cost of fusion, and has good feasibility and robustness in the fusion of video and 3D GIS scene.

Key words: video, topology information, fusion object, 3D GIS scene, dynamic update

中图分类号: