基于视频粒子流和FTLE场的人群运动分割算法

doi:10.3724/SP.J.1087.2012.00252

计算机应用 ›› 2012, Vol. 32 ›› Issue (01): 252-255.DOI: 10.3724/SP.J.1087.2012.00252

基于视频粒子流和FTLE场的人群运动分割算法

童超,章东平,陈非予

中国计量学院信息工程学院，杭州 310018

收稿日期:2011-06-27 修回日期:2011-08-10 发布日期:2012-02-06 出版日期:2012-01-01
通讯作者: 章东平
作者简介:童超(1987-)，男，浙江衢州人，硕士研究生，主要研究方向：计算机视觉；章东平(1970-)，男，江西波阳人，副教授，博士，主要研究方向:图像处理、视频分析；陈非予(1987-)，男，浙江绍兴人，硕士研究生，主要研究方向：计算机视觉。
基金资助:
浙江省自然科学基金资助项目(Y1110506);浙江省科技计划项目(2010C31010);浙江省网络通信技术与应用重点实验室资助项目(2011Z1001);上海市信息安全综合管理技术研究重点实验室资助项目(AGK2011002)

Crowd motion segmentation algorithm based on video particle flow and FTLE field

TONG Chao,ZHANG Dong-ping,CHEN Fei-yu

College of Information Engineering, China Jiliang University, Hangzhou Zhejiang 310018, China

Received:2011-06-27 Revised:2011-08-10 Online:2012-02-06 Published:2012-01-01
Contact: ZHANG Dong-ping

摘要/Abstract

摘要： 针对复杂视频监控场景中不同运动行为的人群分割，提出了将视频粒子流和有限时间李雅普诺夫指数(FTLE)场相结合的人群运动分割算法。首先利用视频粒子流来表示长周期的粒子运动估计，通过最小化包含粒子外观匹配一致性和粒子间形变的能量函数，来优化每个粒子的轨迹；接着求解粒子流图的空间梯度，并构造FTLE场；最后利用FTLE场中的拉格朗日相干结构把流图分割成运动特性不同的区域。实验结构表明，算法能从拥挤复杂的视频监控场景中有效地分割出不同运动特性的群体，且具有较好的鲁棒性。

关键词: 视频粒子, 粒子流图, 李雅普诺夫指数, 运动分割, 拉格朗日相干结构

Abstract: To segment moving crowd with different dynamics in complex video surveillance scenes, this paper proposed a crowd motion segmentation algorithm which was based on video particle flow and Finite Time Lyapunov Exponent (FTLE) field. Firstly, video particle flow was used to represent the long-range particle motion estimation. To optimize these particles trajectories, an energy function containing point-based appearance matching and distortion between the particles was minimized. Then the spatial gradient of the particle flow map was solved and the FTLE field was constructed. Finally, the Lagrangian Coherent Structure (LCS) in the FTLE field was used to divide flow into regions of qualitatively different dynamics. The experimental results show that the proposed algorithm can effectively segment crowd flow with different dynamics in complex video surveillance scenes, and it has strong robustness.

Key words: particle viedo, particle flow, Lyapunov exponent, motion segmentation, Lagrangian Coherent Structure (LCS)

中图分类号:

TP391.413

童超章东平陈非予. 基于视频粒子流和FTLE场的人群运动分割算法[J]. 计算机应用, 2012, 32(01): 252-255.

TONG Chao ZHANG Dong-ping CHEN Fei-yu. Crowd motion segmentation algorithm based on video particle flow and FTLE field[J]. Journal of Computer Applications, 2012, 32(01): 252-255.

参考文献

[1]

MEHRAN R, MOORE B E, SHAH M. A streakline representation of flow in crowded scenes ［C］// Proceedings of the 11th European Conference on Computer Vision. Berlin: Springer-Verlag, 2010: 439-452.

[2]

NGUYEN H T, JI Q, SMEULDERS A W M. Spatio-temporal context for robust multitarget tracking ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007, 29(1): 52-64.

[3]

KUMAR M P, TORR P H S, ZISSERMAN A. Learning layered motion segmentations of video ［J］. International Journal of Computer Vision, 2008, 76(3): 301-319.

[4]

TU P, SEBASTIAN T, DORETTO G, et al. Unified crowd segmentation ［C］// Proceedings of the 10th European Conference on Computer Vision. Berlin: Springer-Verlag, 2008: 691-704.

[5]

ZHAO T, NEVATIA R. Bayesian human segmentation in crowded situations ［C］// IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington, DC: IEEE Computer Society, 2003: 459-466.

[6]

TU P, RITTSCHER J. Crowd segmentation through emergent labeling ［C］// European Conference on Computer Vision Workshop on Statistical Methods in Video Processing. Berlin: Springer-Verlag, 2004: 187-198.

[7]

BROSTOW G J, CIPOLLA R. Unsupervised Bayesian detection of independent motion in crowds ［C］// IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington, DC: IEEE Computer Society, 2006: 594-601.

[8]

CHAN A B, VASCONCELOS N. Mixtures of dynamic textures［C］// Proceedings of the Tenth IEEE International Conference on Computer Vision. Washington, DC: IEEE Computer Society, 2005: 641-647.

[9]

ALI S, SHAH M. A Lagrangian particle dynamics approach for crowd flow segmentation and stability analysis ［C］// IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington, DC: IEEE Computer Society, 2007: 1-6.

[10]

SAND P, TELLER S. Particle video: Long-range motion estimation using point trajectories ［C］// IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington, DC: IEEE Computer Society, 2006: 2195-2202.

[11]

LISCHINSKI D. Chapter incremental Delaunay triangulation ［M］// HECKBERT P. Graphics gems IV. Boston: Academic Press, 1994.

[12]

BROX T, BRUHN A, PAPENBERG N, et al. High accuracy optical flow estimation based on a theory for warping ［C］// European Conference on Computer Vision. Berlin: Springer-Verlag, 2004: 25-36.

基于视频粒子流和FTLE场的人群运动分割算法

Crowd motion segmentation algorithm based on video particle flow and FTLE field

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 6

编辑推荐

Metrics

[1]	吴杰祺, 李晓宇, 袁晓彤, 刘青山. 利用坐标下降实现并行稀疏子空间聚类[J]. 计算机应用, 2016, 36(2): 372-376.
[2]	张运楚宋世军张汝敏郝建林. 高斯混合背景模型的适应能力研究[J]. 计算机应用, 2011, 31(03): 706-709.
[3]	王新马义德徐志坚李涟凤. 基于脉冲耦合神经网络的混沌控制[J]. 计算机应用, 2009, 29(12): 3277-3279.
[4]	刘玉兰彭思龙. 基于图像块差分和Mean Shift算法的运动目标检测[J]. 计算机应用, 2008, 28(8): 2017-2020.
[5]	李全民张运楚. 自适应混合高斯背景模型的改进[J]. 计算机应用, 2007, 27(8): 2014-2017.
[6]	谭晓军，沈伟，郭志豪. 一种基于视频的道路交通流量监测方法[J]. 计算机应用, 2005, 25(05): 1215-1218.