Behavior pattern extraction by trajectory analysis

Abstract

Trajectory clustering and behavior pattern extraction are the foundations of research into activity perception of objects in motion. In this paper, a new framework is proposed to extract behavior patterns through trajectory analysis. Firstly, we introduce directional trimmed mean distance (DTMD), a novel method used to measure similarity between trajectories. DTMD has the attributes of anti-noise, self-adaptation and the capability to determine the direction for each trajectory. Secondly, we use a hierarchical clustering algorithm to cluster trajectories. We design a length-weighted linkage rule to enhance the accuracy of trajectory clustering and reduce problems associated with incomplete trajectories. Thirdly, the motion model parameters are estimated for each trajectory’s classification, and behavior patterns for trajectories are extracted. Finally, the difference between normal and abnormal behaviors can be distinguished.

Keywords

trajectory clustering / directional trimmed mean distance (DTMD) / behavior pattern extraction

References

[1] Wang X, Tieu K, Crimson E. Learning scene models by trajectory analysis. In: Proceedings of ECCV, 2006: 110–123

[2] Howarth R J, Buxton H. An analogical representation of space and time. Image and Vision Computing, 1992, 10(7): 467–478

[3] Brand M, Kettnaker V. Discovery and segmentation of activities in video. IEEE Transactions on Behavior pattern Analysis and Machine Intelligence, 2000, 22(8): 844–851

[4] Kim Z, Malik J. Fast vehicle detection with probabilistic feature grouping and its application to vehicle tracking. In Proceedings of IEEE Int’l Conf. Computer Vision, 2003: 524–531

[5] Veeraraghavan H, Masoud O, Papanikolopoulos N P. Computer vision algorithms for intersection monitoring. IEEE Transactions on Intelligent Transportation Systems, 2003, 4(2): 78–89

[6] Stauffer C, W Eric L. Grimson. Learning behavior patterns of activity using real-time tracking. IEEE Transactions on Behavior pattern Analysis and Machine Intelligence, 2000, 22(8): 747–753

[7] Johnson N, Hogg D. Learning the distribution of object trajectories for event recognition. Image and Vision Computing, 1996, 14(8): 583–592

[8] Sumpter N, Bulpitt A J. Learning spatio-temporal patterns for predicting object behaviour. Image and Vision Computing, 2000, 18(9): 697–704

[9] Hu W, Xie D, Tan T N, Maybank S. Learning activity patterns using fuzzy self-organizing neural network. IEEE Transactions on Systems, Man, and Cybernetics. Part B, Cybernetics, 2004, 34(3): 1618–1626

[10] Melo J, Naftel A, Bernardino A, Santos-Victor J. Detection and classification of highway lanes using vehicle motion trajectories. IEEE Transactions on Intelligent Transportation Systems, 2006, 7(2): 188–200

[11] Kim ZReal time object tracking based on dynamic feature grouping with background subtraction. CVPR. 2008: 1–8

[12] Buzan D, Sclaroff S, Kollios G. Extraction and clustering of motion trajectories in video. In: Proceedings of the 17th International Conference on Behavior Pattern Recognition, London. 2004: 521–524

[13] HuttenlocherD P, Klanderman G, RucklidgeW J. Comparing images using the Hausdorff distance. IEEE Transactions on Behavior pattern Analysis and Machine Intelligence, 2006, 15(9): 850–863

[14] Dubuisson M P, Jain A K. A modified Hausdorff distance for object matching. In: Proceedings of 12th International Conference on Behavior Pattern Recognition, Jerusalem, Israe. 1994: 566–568

[15] Atev S, Masoud O, Papanikolopoulos N. Learning traffic behavior patterns at intersections by spectral clustering of motion trajectories. In: Proceedings of the 2006 IEEE/RSJ International Conference on Intelligent Robots and Systems. 2006: 4851–4856

[16] Fu Z, Hu W, Tan T. Similiarity based vehicle trajectory clustering and anomaly detection. In: Proceedings of the 2005 IEEE International Conference on Image processing. 2005: 602–605

RIGHTS & PERMISSIONS

2014 Higher Education Press and Springer-Verlag Berlin Heidelberg