[1]张丽坤,孙建德,李静.视觉关注转移的事件检测算法[J].智能系统学报,2012,7(4):333-338.
ZHANG Likun,SUN Jiande,LI Jing.Event detection based on visual attention shift[J].CAAI Transactions on Intelligent Systems,2012,7(4):333-338.
点击复制
ZHANG Likun,SUN Jiande,LI Jing.Event detection based on visual attention shift[J].CAAI Transactions on Intelligent Systems,2012,7(4):333-338.
视觉关注转移的事件检测算法
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
7
期数:
2012年第4期
页码:
333-338
栏目:
学术论文—机器学习
出版日期:
2012-08-25
- Title:
- Event detection based on visual attention shift
- 文章编号:
- 1673-4785(2012)04-0333-06
- Keywords:
- intelligent surveillance; event detection; occurrence of event; visual attention model; visual attention rhythm; key frame extraction
- 分类号:
- TP18；TN911.73
- 文献标志码:
- A
- 摘要:
- 智能监控系统已广泛应用于银行、超市、公交车等公共场合，监控视频的事件检测已经成为智能监控中的关键技术．提出了一种基于视觉关注转移的事件检测方法，该方法首先分别通过对视频帧进行动态和静态受关注模型的提取得到视觉关注显著图，然后根据视觉关注显著图的时域特性形成视觉关注节奏曲线，根据视觉关注节奏的变化强度选取关键帧，以关键帧形式表示受关注事件的发生．实验结果表明，算法提取的关键帧可以准确地标示监控视频中特征事件的发生，并且可以做到实时地检测事件．
- Abstract:
- Intelligent surveillance systems have been widely used in banks, supermarkets, buses, and other public places. Event detection in a surveillance video is a key technology for this field. In this paper, a visual attention shiftbased event detection algorithm was proposed for intelligent surveillance in which the dynamic and static visual attention regions were detected to obtain the visual saliency map. After that, the visual attention rhythm was derived from the visual saliency map temporally. According to the visual attention rhythm, the key frames were selected to label the occurrence of the events. Experimental results demonstrate that the proposed algorithm can label the occurrence of the events with the extracted key frames correctly, and that the event detection is performed in realtime.
- 参考文献/References:
-
［1］JIANG F, WU Y. A dynamic hierarchical clustering method for trajectorybased unusual video event detection［J］. IEEE Transactions on Image Processing, 2009, 18(4): 907913.
［2］LIU C, WANG G J, NING W X, et al. Anomaly detection in surveillance video using motion direction statistics［C］// IEEE International Conference on Image Processing. Hong Kong, China, 2010: 717720.
［3］JIANG P, QIN X L. Keyframebased video summary using visual attention clues［J］.IEEE Transactions on Multimedia, 2010, 17(2): 6473.
［4］LAI J L, YI Y. Key frame extraction based on visual attention model［J］. Journal of Visual Communication and Image Representation, 2012, 23(1): 114125.
［5］AMIRI A, FATHY M, NASERI A. Keyframe extraction and video summarization using QRdecomposition［C］//IEEE International Conference on Multimedia Technology and Applications. Wuhan, China, 2010: 134139.
［6］ZHANG J, SUN J D, YAN H, et al. Visual attention model with crosslayer saliency optimization［C］//IEEE International Conference on Intelligent Information Hiding and Multimedia Signal Processing. Dalian, China, 2011: 240243.
［7］ITTI L, KOCH C, NIEBUR E. A model of saliency based visual attention for rapid scene analysis［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence, 1998, 20(11): 12541259.
［8］RUBINSTEIN M, SHAMIR A, AVIDAN S. Multioperator media retargeting［J］.ACM Transactions on Graphics, 2009, 23(3): 18.
- 相似文献/References:
-
[1]毋? 非,封化民,申晓晔.容错粗糙模型的事件检测研究[J].智能系统学报,2009,4(2):112.
WU Fei,FENG Hua-min,SHEN Xiao-ye.Research on event detection based on the tolerance rough set model[J].CAAI Transactions on Intelligent Systems,2009,4():112.
备注/Memo
收稿日期：2012-03-02.
网络出版日期：2012-07-12.
基金项目：国家自然科学基金资助项目（61001180）.
通信作者：孙建德.
E-mail:jd_sun@sdu.edu.cn.
作者简介：
张丽坤，女，1986年生，硕士研究生，主要研究方向为多媒体信号处理．
孙建德，男，1978年生，副教授．主要研究方向为基于内容的多媒体分析、基于视觉关注模型的图像/视频分析、图像/视频复制检测、多媒体信息内容安全和数字水印等．承担及完成科研项目10余项，授权发明专利8项，发表学术论文40余篇．
李静，女，1979年生，主要研究方向为多媒体通信．
更新日期/Last Update:
2012-09-27