基于目标检测及跟踪的视频摘要技术研究

doi:10.11896/j.issn.1002-137X.2016.11.057

计算机科学 ›› 2016, Vol. 43 ›› Issue (11): 297-299.doi: 10.11896/j.issn.1002-137X.2016.11.057

基于目标检测及跟踪的视频摘要技术研究

田合雷,丁胜,于长伟,周立

合肥工业大学仪器科学与光电工程学院合肥230009,合肥工业大学计算机与信息学院合肥230009,合肥工业大学仪器科学与光电工程学院合肥230009,合肥工业大学计算机与信息学院合肥230009

出版日期:2018-12-01 发布日期:2018-12-01
基金资助:
本文受国家自然科学基金 (61371155),安徽省科技攻关基金项目(1301b042023)资助

Research of Video Abstraction Based on Object Detection and Tracking

TIAN Helei, DING Sheng, YU Changwei and ZHOU Li

Online:2018-12-01 Published:2018-12-01

摘要/Abstract

摘要： 为了能够在不丢失有用信息的前提下对海量监控视频进行摘要,提出一种基于目标检测及跟踪的视频摘要技术。首先根据混合高斯模型对视频进行背景建模,获得监控视频的背景；然后通过背景减法对运动目标进行检测,得到前景运动目标；其次对检测到的运动目标采用分层关联的思想进行目标跟踪,得到运动目标的完整信息；最后将这些运动目标及视频背景重新组装成摘要视频。实验结果表明,该方法能够有效地对监控视频进行浓缩,形成的摘要视频能够完整地保存原有视频信息,减少了存储空间,降低了成本,方便了相关人员及时获取有用信息,提高了工作效率。

关键词: 视频摘要,视频浓缩,目标检测,目标跟踪,背景建模,分层关联

Abstract: ion Based on Object Detection and Tracking TIAN He-lei¹ DING Sheng² YU Chang-wei¹ ZHOU Li² (School of Instrument Science and Opto-electronics Engineering,Hefei University of Technology,Hefei 230009,China)¹ (School of Computer and Information,Hefei University of Technology,Hefei 230009,China)² Abstract In order to carry out a summary of massive surveillance video under the premise of not losing useful information,a video summarization technique based on object detection and tracking was proposed.Moving targets is detected by background subtraction in video by using Gaussian mixture model,and target tracking for detected target is done by using the idea of hierarchical correlation to get complete information of moving targets.Finally,the moving objects and video background are reassembled into the abstract video.The experimental results show that the method proposed in this paper can be used to concentrate the monitoring video.The summary of the video can completely preserve the original video,which reduces the storage space and cost.It is also convenient for relevant personnel to obtain useful information in time and improve work efficiency.

Key words: Video abstraction,Video condensation,Object detection,Object tracking,Background modeling,Hierarchical correlation

田合雷,丁胜,于长伟,周立. 基于目标检测及跟踪的视频摘要技术研究[J]. 计算机科学, 2016, 43(11): 297-299. https://doi.org/10.11896/j.issn.1002-137X.2016.11.057

TIAN Helei, DING Sheng, YU Changwei and ZHOU Li. Research of Video Abstraction Based on Object Detection and Tracking[J]. Computer Science, 2016, 43(11): 297-299. https://doi.org/10.11896/j.issn.1002-137X.2016.11.057

参考文献

[1] Kumar N S,Shobha G,Balaji S.Key frame extraction algorithm for video abstraction applications in underwater videos[C]∥Underwater Technology(UT),2015.IEEE,2015:1-5
[2] Song G H,Ji Q G,Lu Z M,et al.A novel video abstraction method based on fast clustering of the regions of interest in key frames[J].AEU-International Journal of Electronics and Communications,2014,68(8):783-794
[3] Wang Ya-pei,Li Ren-wang,Liu Xiang.Extraction Method ofSurveillance Vudeo Synopsis Combines Objects and Keyframes[J].Industrial Control Computer,2015(3):11-13(in Chinese) 王亚沛,李仁旺,刘翔.对象和关键帧相结合的监控视频摘要提取方法[J].工业控制计算机,2015(3):11-13
[4] Zhang P,Zhuo T,Zhang Y,et al.Real-time Tracking-by-Lear-ning with High-order Regularization Fusion for Big Video Abstraction[J].Signal Processing,2015,124(c):246-258
[5] Liu Shou-da.Video Synopsis Based on Multi-target Trackingand Trajectory Combinatorial Optimization[D].Xiamen:Xiamen University,2014(in Chinese) 刘守达.基于多目标跟踪及轨迹组合优化的视频摘要[D].厦门:厦门大学,2014
[6] Han Jan-kang.Moving Area Detection and Tracking Based Vi-deo Condensation[D].Beijing:Beijing University of Posts and Telecommunications,2012(in Chinese) 韩建康.基于运动检测及跟踪的视频浓缩方法研究[D].北京:北京邮电大学,2012
[7] Felzenszwalb P,Girshick R,McAllester D,et al.Object dete-ction with discriminatively trained part based models[J].Pattern Analysis and Machine Intelligence(PAMI),2010,32(9):1627-1645
[8] Dalal N,Triggs B.Histograms of oriented gradients for human detection [C]∥Computer Vision and Pattern Recognition(CVPR).San Diego,CA,2005,1:886-893
[9] Wu B,Nevatia R.Detection and tracking of multiple,partially occluded humans by bayesian combination of edgelet based part detectors[J].International Journal of Computer Vision,2007,75(2):247-266
[10] Song Xue-hua,Chen Yu,Geng Jian-feng.Moving object detection based in improved Gaussian mixture background model [J].Computer Engineering and Design,2010,31(21):4646-4649(in Chinese) 宋雪桦,陈瑜,耿剑锋.基于改进的混合高斯背景模型的运动目标检测[J].计算机工程与设计,2010,1(21):4646-4649
[11] Zivkovic Z.Improved Adaptive Gaussian Mixture Model forBackground Subtraction[C]∥Proceedings of the International Conference on Recongnition Pattern,Amsterdam University.Netherlands,2004,2:23-26
[12] Huang C,Li Y,Nevatia R.Multiple target tracking by learning-based hierarchical association of detection responses[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2013,35(4):898-910
[13] Wang Jiang-feng.Researches on Object Tracking and Event Detection Based on Tracklet Association[D].Changsha: National University of Defense Technology,2011(in Chinese) 王江峰.基于轨迹片段关联的目标跟踪与事件检测方法研究[D].长沙:国防科学技术大学,2011
[14] Ristani E,Tomasi C.Tracking Multiple People Online and inReal Time[M]∥Computer Vision--ACCV 2014.Springer International Publishing,2015:444-459

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

基于目标检测及跟踪的视频摘要技术研究

Research of Video Abstraction Based on Object Detection and Tracking

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

Metrics

本文评价

推荐阅读 0