计算机与现代化

• 图像处理 • 上一篇    下一篇

智能视频监控中的遗留物检测技术

  

  1. (黄河科技学院信息工程学院,河南 郑州 450063)
  • 收稿日期:2014-09-11 出版日期:2014-12-22 发布日期:2014-12-22
  • 作者简介:张具琴(1980-),女,河南信阳人,黄河科技学院信息工程学院讲师,硕士,研究方向:电磁场与无线通信技术; 通信作者:王海洋(1989-),男,河南驻马店人,本科生,研究方向:无线通信技术。
  • 基金资助:
    郑州市无线与移动通信网络应用技术科技创新团队项目(X2012GC1137)

Abandoned Objects Detection in Intelligent Video Surveillance

  1. (School of Information Engineering, Huanghe Science and Technology College, Zhengzhou 450063, China)
  • Received:2014-09-11 Online:2014-12-22 Published:2014-12-22

摘要: 对智能视频监控中的遗留物检测技术进行研究,提出一种能够自动检测视频监控环境中的遗留物并发出警报的检测方案,给出详细的检测算法分析。该方案基于动态阈值的背景差分算法和背景更新算法,提高系统对于复杂场景的适应性,能够准确地检测出复杂背景中的箱、包等遗留物,并能够提供关键帧,便于找到遗留物的失主。实验结果表明了该方案的有效性。

关键词: 遗留物检测, 动态阈值, 背景差分法, 自适应, 最大类间方差法

Abstract: A detection technology in intelligent video surveillance was researched. A method of automatically detecting abandoned objects and alerting in video surveillance environment was given, and also the detection algorithm was analysed in detail. By dynamic threshold background subtraction algorithms and background updating algorithm, the adaptability of the system in complex scenes was improved. This method can accurately detect the abandoned objects in the complex background, such as box, bag and so on, and provide key frames. The effectiveness of this method was confirmed by experimental results.

Key words: abandoned object detection, dynamic threshold, background subtraction method, adaptive, method of maximum classes square error

中图分类号: