Journal of Geo-information Science >
A Geospatial Representation Framework for Geo-tagged Video Objects
Received date: 2014-12-18
Request revised date: 2015-02-23
Online published: 2015-09-07
Copyright
Geo-tagged video contains location information, and it is critical for true geographic representation. The geospatial representation of geo-tagged video is the key feature for the integration of video and GIS. Regarding to the disadvantage of geo-tagged representation methods for video objects with monotone spatial semantic information, a geographic representation framework for geo-tagged video objects is proposed. On the basis of extending OGC specifications for geographic information, this paper defined the respective objects in 7 types from 3 categories to describe the spatial information on two levels, including the video frame and video clip. The 3 categories include: (1) the video positions (point) to represent the location and attitude as the camera taking shoots; (2) the video trajectories (line) to portray the track of the video clip; and (3) the video field of view in plain view (polygon) or 3D (solid) space to describe the spatial extent of the video scene. The framework consists of the main spatial objects including the point, line, polygon and solid. It is more competent for demonstrating video spatial information. Meanwhile, the framework supports different levels of video data, such as the video frame and video clip. It achieves the loosely-coupled and perfectly-integrated integration of video and GIS, which does not need to alter the data structures. This paper discussed the data acquisition methods for the spatial information of video frames or clips in detail, which take use of the GPS receiver and 3D digital compass. We also developed 9 tables and defined their relations for the logical model to realize the geographic representation of geo-tagged video objects, and we analyzed the data visualization and retrieval methods by taking them as the application cases. The results show that the geographic representation framework for geo-tagged video extends the current spatial database standard. It is easy to implement and applicable in geographic visualization, video retrieval and spatial analysis or data mining.
Key words: geo-tagged video; geographic representation; GIS; data model; spatial database
HAN Zhigang , KONG Yunfeng , QIN Yaochen , QIN Fen . A Geospatial Representation Framework for Geo-tagged Video Objects[J]. Journal of Geo-information Science, 2015 , 17(9) : 1014 -1021 . DOI: 10.3724/SP.J.1047.2015.01014
Fig. 1 Geographic representation framework for geo-tagged video objects图1 可定位视频对象的地理空间表达框架 |
Fig. 2 Field of view of video frame in 2D图2 视频帧平面视域示意图 |
Fig. 3 Field of view of video frame in 3D图3 视频帧立体视域示意图 |
Fig. 4 Examples of the geographic representation of geo-tagged video object图4 可定位视频对象地理表达实例 |
Fig. 5 Comparison for actual and theoretical FOV of video frame图5 视频帧理论视域与实际视域对比示意图 |
Fig. 6 Logical model of the geographic representation of geo-tagged video object图6 可定位视频对象地理表达逻辑模型实例 |
Fig. 7 Application of the geographic representation of geo-tagged video object图7 可定位视频对象地理表达应用实例 |
The authors have declared that no competing interests exist.
[1] |
|
[2] |
|
[3] |
|
[4] |
|
[5] |
|
[6] |
|
[7] |
|
[8] |
|
[9] |
|
[10] |
|
[11] |
|
[12] |
|
[13] |
|
[14] |
|
[15] |
|
[16] |
|
[17] |
|
[18] |
|
[19] |
|
[20] |
|
[21] |
|
/
〈 | 〉 |