欢迎访问《金属矿山》杂志官方网站,今天是 分享到:
×

扫码分享

金属矿山 ›› 2023, Vol. 52 ›› Issue (09): 173-179.

• 机电与自动化 • 上一篇    下一篇

基于三维视频融合的矿山视频监控技术研究

苗作华1,2 汤 阳1 任 磊1 王梦婷1 谢 媛1 战 川1 李宏亮3
  

  1. 1. 武汉科技大学资源与环境工程学院,湖北 武汉 430081;2. 冶金矿产资源高效利用与造块湖北省重点实验室,湖北 武汉 430081;3. 北京道亨软件股份有限公司,北京 100012
  • 出版日期:2023-09-15 发布日期:2023-11-03
  • 基金资助:
    国家自然科学基金项目(编号:41071242,41701624,41271449);湖北省大学生创新训练项目(编号:S202010488021);教育部产学合作协同育人项目(编号:202102136008)。

Study on Mine Video Monitoring Technology Based on 3D Video Fusion

MIAO Zuohua1,2 TANG Yang1 REN Lei1 WANG Mengting1 XIE Yuan1 ZHAN Chuan1 LI Hongliang3 #br#   

  1. 1. School of Resources and Environmental Engineering,Wuhan University of Science and Technology,Wuhan 430081,China;2. Hubei Key Laboratory for Efficient Utilization and Agglomeration of Metallurgic Mineral Resources,Wuhan 430081,China;3. Beijing Daoheng Software Co. ,Ltd. ,Beijing 100012,China
  • Online:2023-09-15 Published:2023-11-03

摘要: 传统矿山视频监控存在视频数据量大、视频图像之间关联性差、缺乏与三维空间位置信息的直接联动 等问题,监管人员在面对多个碎片化的监控视频图像时难以有效复现场景及位置,这使得突发事件发生时不能及时 解决相关问题,严重影响了应急救援工作及时有效开展。 针对以上问题,提出了一种基于三维视频融合的矿山视频 监控方法,利用三维视频融合技术,将实时监控视频作为纹理构建几何体,通过特征点选择、匹配与矿山三维模型进 行融合,在此基础上为了使三维视频纹理融合效果更好,采用拉普拉斯网格形变算法对视频几何体纹理的畸变进行 校正。 试验结果表明:三维视频融合具有较强的视觉观感效果,视频与视频之间相互关联,整合了碎片化的视频;视频 与三维模型融合,解决了三维空间数据与视频不联动的问题,弥补了传统视频监管技术的不足,丰富了矿山监测的技 术手段。

关键词: 三维可视化, 三维视频融合, 矿山视频监控, 网格形变, 纹理校正

Abstract: The traditional mine video monitoring has the problems of large amount of video data,poor correlation between video images,and lack of direct linkage with the three-dimensional spatial location information. It is difficult for regulators to effectively reproduce the scene and location in the face of multiple fragmented monitoring video images,which makes it impossible to solve relevant problems in time when emergencies occur,and seriously affects the timely and effective development of mine emergency rescue work. In view of the above problems,a mine video monitoring method based on three-dimensional video fusion is proposed. Using three-dimensional video fusion technology,the real-time monitoring video is used as a texture to construct a geometry,and the feature points are selected and matched to fuse with the mine three-dimensional model. On this basis,in order to make the three-dimensional video texture fusion effect better,the Laplacian grid deformation algorithm is adopted to correct the distortion of the video geometry texture. The experimental results show that the three-dimensional video fusion has a strong visual perception effect. The video and video are interrelated,and the fragmented video is integrated. The fusion of video and three-dimensional model solves the problem of non-linkage between three-dimensional space data and video,makes up for the defects of traditional video monitoring technology,and enriches the means of mine monitoring.

Key words: 3D visualization,3D video fusion,mine video surveillance,mesh deformation,texture correction