视频数据的地理信息提取研究

Research on geographic information extracting based on video data

针对视频数据虽然能高效、逼真、详细地展现地理信息,但是因不具备地理框架而不能直接与GIS结合,参与空间分析的问题,该文通过多次真实环境条件下的尝试性实验,探索出一种从视频中提取地理信息的方案:采用运动目标检测技术自动识别出视频帧中实体的轮廓;通过半自动的处理清除提取数据的干扰轮廓;借助数据变换技术为视频帧及其提取的数据赋予准确的地理框架。基于真实环境的实验结果表明:该方案不仅可以有效地提取视频中的地理信息参与空间分析,还能够实现其空间分析结果的多视角可视化。本研究针对视频与GIS结合问题做出初步探索,为视频GIS的深层次研究提供帮助。

Despite displaying the real geographic information in detail and effectively,video data can not provide real geographic frame so that it could not be directly involved in spatial analysis.In order to solve the problem,this paper presented a method of extracting geographic information from video data through a series of experiment in real situation.The objects＇ contour on video frame could be identified automatically based on motion target detection technology,interference contours were eliminated through the semi-automatic processing,video frame and the extracted data were defined the real spatial coordinates through data transform functions.The results of real experiment indicated that the method could not only extract the geographic information from video data effectively for spatial analysis,but also the results of analysis could be visualized in multi-view.The paper makes a preliminary exploration for the combination of video and GIS,and provides assist to further study of VideoGIS.