期刊文献+

视频特征下的电视广告单元分割技术研究

Research on TV Advertising Unit Segmentation Based on Video Features
在线阅读 下载PDF
导出
摘要 本文基于视频特征,对电视广告单元分割技术进行了研究。对广告视频的检测算法框架流程展开深入分析。综合切变镜头检测算法以及消隐镜头,结合区域特征重要性对其进行划分,形成不同的镜头合集,在分类特征的基础上提取镜头。并分析了视频广告在时间、内容上的连续性,此外完善了广告镜头后期处理机制,全面消除错误镜头。此外还专门制定了镜头整合算法。此外还专门制定了镜头整合算法,合并广告镜头,以产出广告视频片段。 Based on video characteristics,this paper studies the segmentation teciinology of TV advertising unit.The frame flow of detection algorithm for advertising video is analyzed.The vicieo segment is segmented into a set of sliots,which is based on the importance of regional feature,and extracts the classification features based on the lens.The continuity of video advertisement in time and content is analyzed,and the post processing mechanismof accumulative advertising lens based on sliding window is put forward to eliminate some error shots.And the lens integration algorithm is proposed to merge the advertising lens to get advertising video clips.
作者 邓海生 DENG Haisheng(College of Information Engineering,Xijing University,Xi'an 110123.China)
出处 《电视技术》 2018年第12期75-78,共4页 Video Engineering
基金 西京学院 基于手机的老人行为识别系统设计与实现(XJ150123) 陕西省教育厅 基于数据驱动辨识的多智能体有限时间协调优化研究(16JK2243)
关键词 消隐镜头 切变镜头 视频特征 分割技术 电视广告 blanking lens cut shot video feature segmentation technolog TV advertisement
作者简介 邓海生(1980—),硕士,副教授,研究方向:Web中间件研究、数字水印技术等。
  • 相关文献

参考文献3

二级参考文献39

  • 1Wixson L, et al. Detecting salient motion by accumulating directionally-consistent flow. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22(8) : 774-780.
  • 2Karm ann K P, Moving object recognition using an adaptive background memory. The Netherlands: Elsevier Science Publishers, 1990: 289-307.
  • 3Friedman N, Russell S J. Image Segmentation in Video Sequences Approach: A Probabilistic approach//Proceedings of the Uncertainty in Artificial Intelligence UAI. Colorado, USA, 1997:175-181.
  • 4Fragkiadaki K, Shi Jian-Bo. Detection free tracking: Exploi- ting motion and topology for segmenting and tracking under cntanglement//Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition. Colorado. USA, 2011:2073-2080.
  • 5Li Zong-Min, Zhong Liang-Liang, Liu Yu-Jie. Efficient fore- ground layer exlraetion in video//Proeeedings of the IEEE Pacific Rim Conference on Muhimedia Conference. Shanghai, China, 2010, 6297:319-329.
  • 6Collins R, Lipton A J, Kanade T, et al. A system for video surveillance and monitoring. Technical Report= CMURI-TR- 00112, Carnegie Melon University, Pittsburgh, Peen, America, 2000:1-68.
  • 7Elias Herrero-Jaraba, et al. Detected motion classification with a double-background and a Neighborhood based differ ence. Patern Recognition Letters, 2003, 24(12) : 2079 -2092.
  • 8Joaquin S, Hugo J-H. A double layer background model to detect unusual events//Proceedings of the Advanced Con- cepls for Intelligent Vision Systems-AClVS. Las Vegas, USA, 2007:406-416.
  • 9Arnell F, Petersson L. Fast object segmentation from a moving camera//Proceedings of the Intelligent Vehicles Symposium. Las Vegas, USA, 2005:136-141.
  • 10Yang Chang Jiang, Duraiswami R, DcMcnthon D. Mean- shift analysis using quasi-newton methods//Proceedings of the IEEE International Conference on Image Processing. Barcelona, Spain, 2003:447-450.

共引文献8

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部