《计算机应用》唯一官方网站 ›› 2023, Vol. 43 ›› Issue (8): 2330-2337.DOI: 10.11772/j.issn.1001-9081.2022101566

• 第十九届CCF中国信息系统及应用大会 • 上一篇    

面向视频数据的时空伴随模式挖掘算法

张潇誉1, 于自强1,2(), 刘承栋1, 李博涵3, 靖常峰4   

  1. 1.烟台大学 计算机与控制工程学院, 山东 烟台 264005
    2.自然资源部城市国土资源监测与仿真重点实验室, 广东 深圳 518034
    3.南京航空航天大学 计算机科学与技术学院, 南京 211106
    4.中国地质大学(北京) 信息工程学院, 北京 100083
  • 收稿日期:2022-09-26 修回日期:2022-11-08 接受日期:2022-11-11 发布日期:2023-02-14 出版日期:2023-08-10
  • 通讯作者: 于自强
  • 作者简介:张潇誉(1999—),女,山东枣庄人,硕士研究生,主要研究方向:视频数据处理
    刘承栋(2000—),男,山东淄博人,硕士研究生,主要研究方向:视频数据结构化查询
    李博涵(1979—),男,副教授,博士,CCF高级会员,主要研究方向:时空数据库、知识图谱、自然语言处理
    靖常峰(1979—),男,山东济南人,教授,博士,主要研究方向:城市运行管理物联网技术、城市时空大数据建模、时空分析与挖掘。
  • 基金资助:
    国家自然科学基金资助项目(62172351)

Spatial-temporal co-occurrence pattern mining algorithm for video data

Xiaoyu ZHANG1, Ziqiang YU1,2(), Chengdong LIU1, Bohan LI3, Changfeng JING4   

  1. 1.School of Computer and Control Engineering,Yantai University,Yantai Shandong 264005,China
    2.Key Laboratory of Urban Land Resources Monitoring and Simulation,Ministry of Natural Resources,Shenzhen Guangdong 518034,China
    3.College of Computer Science and Technology,Nanjing University of Aeronautics and Astronautics,Nanjing Jiangsu 211106,China
    4.School of Information Engineering,China University of Geosciences,Beijing,Beijing 100083,China
  • Received:2022-09-26 Revised:2022-11-08 Accepted:2022-11-11 Online:2023-02-14 Published:2023-08-10
  • Contact: Ziqiang YU
  • About author:ZHANG Xiaoyu,born in 1999, M. S. candidate. Her research interests include video data processing.
    LIU Chengdong, born in 2000, M. S. candidate. His research interests include structured querying of video data.
    LI Bohan, born in 1979, Ph. D., associate professor. His research interests include spatial-temporal database, knowledge graph, natural language processing.
    JING Changfeng, born in 1979, Ph. D., professor. His research interests include internet of things technology for urban operation and management, urban spatial-temporal big data modeling, spatial-temporal analysis and mining.
  • Supported by:
    National Natural Science Foundation of China(62172351)

摘要:

时空伴随模式是具有时空伴随关系的视频对象组合。为了从海量视频数据中快速发现符合查询条件的时空伴随模式,提出一种基于三重剪枝匹配策略的时空伴随模式发现算法——MPA。首先,利用已有的视频对象识别和跟踪模型对视频对象进行结构化提取;然后,对提取的连续帧中大量重复出现的视频对象进行压缩存储并构建索引;最后,设计基于前缀树的时空伴随模式发现算法,以快速发现符合查询条件的时空伴随模式。在真实数据集和合成数据集上的实验结果表明,与暴力搜索算法(BFA)相比,所提算法的效率提高了30%左右,且数据量越大,效率提高越明显。因此,所提算法能够快速发现海量视频数据中满足查询条件的时空伴随模式。

关键词: 视频对象, 结构化, 时空伴随模式, 索引结构, 剪枝策略

Abstract:

Spatial-temporal co-occurrence patterns refer to the video object combinations with spatial-temporal correlations. In order to mine the spatial-temporal co-occurrence patterns meeting the query conditions from a huge volume of video data quickly, a spatial-temporal co-occurrence pattern mining algorithm with a triple-pruning matching strategy — Multi-Pruning Algorithm (MPA) was proposed. Firstly, the video objects were extracted in a structured way by the existing video object detection and tracking models. Secondly, the repeated occurred video objects extracted from a sequence of frames were stored and compressed, and an index of the objects was created. Finally, a spatial-temporal co-occurrence pattern mining algorithm based on the prefix tree was proposed to discover the spatial-temporal co-occurrence patterns that meet query conditions. Experimental results on real and synthetic datasets show that the proposed algorithm improves the efficiency by about 30% compared with Brute Force Algorithm (BFA), and the greater the data volume, the more obvious the efficiency improvement. Therefore, the proposed algorithm can discover the spatial-temporal co-occurrence patterns satisfying the query conditions from a large volume of video data quickly.

Key words: video object, structuration, spatial-temporal co-occurrence pattern, index structure, pruning strategy

中图分类号: