Motion Pattern Study and Analysis from Video Monitoring Trajectory

Kai KANG; Weibin LIU; Weiwei XING

doi:10.1587/transinf.E97.D.1574

IEICE TRANSACTIONS on Information

Motion Pattern Study and Analysis from Video Monitoring Trajectory

Kai KANG, Weibin LIU, Weiwei XING

Full Text Views

0

Cite this

Summary :

This paper introduces an unsupervised method for motion pattern learning and abnormality detection from video surveillance. In the preprocessing steps, trajectories are segmented based on their locations, and the sub-trajectories are represented as codebooks. Under our framework, Hidden Markov Models (HMMs) are used to characterize the motion pattern feature of the trajectory groups. The state of trajectory is represented by a HMM and has a probability distribution over the possible output sub-trajectories. Bayesian Information Criterion (BIC) is introduced to measure the similarity between groups. Based on the pairwise similarity scores, an affinity matrix is constructed which indicates the distance between different trajectory groups. An Adaptable Dynamic Hierarchical Clustering (ADHC) tree is proposed to gradually merge the most similar groups and form the trajectory motion patterns, which implements a simpler and more tractable dynamical clustering procedure in updating the clustering results with lower time complexity and avoids the traditional overfitting problem. By using the HMM models generated for the obtained trajectory motion patterns, we may recognize motion patterns and detect anomalies by computing the likelihood of the given trajectory, where a maximum likelihood for HMM indicates a pattern, and a small one below a threshold suggests an anomaly. Experiments are performed on EIFPD trajectory datasets from a structureless scene, where pedestrians choose their walking paths randomly. The experimental results show that our method can accurately learn motion patterns and detect anomalies with better performance.

Publication: IEICE TRANSACTIONS on Information Vol.E97-D No.6 pp.1574-1582

Publication Date: 2014/06/01

Publicized

Online ISSN: 1745-1361

DOI: 10.1587/transinf.E97.D.1574

Type of Manuscript: PAPER

Category: Pattern Recognition

Authors

Kai KANG
  Beijing Jiaotong University,Beijing Key Laboratory of Advanced Information Science and Network Technology
Weibin LIU
  Beijing Jiaotong University,Beijing Key Laboratory of Advanced Information Science and Network Technology
Weiwei XING
  Beijing Jiaotong University

Keyword

visual monitoring, trajectory clustering, Hidden Markov Models, hierarchical clustering, abnormal detection

Cite this

Copy

Kai KANG, Weibin LIU, Weiwei XING, "Motion Pattern Study and Analysis from Video Monitoring Trajectory" in IEICE TRANSACTIONS on Information, vol. E97-D, no. 6, pp. 1574-1582, June 2014, doi: 10.1587/transinf.E97.D.1574.
Abstract: This paper introduces an unsupervised method for motion pattern learning and abnormality detection from video surveillance. In the preprocessing steps, trajectories are segmented based on their locations, and the sub-trajectories are represented as codebooks. Under our framework, Hidden Markov Models (HMMs) are used to characterize the motion pattern feature of the trajectory groups. The state of trajectory is represented by a HMM and has a probability distribution over the possible output sub-trajectories. Bayesian Information Criterion (BIC) is introduced to measure the similarity between groups. Based on the pairwise similarity scores, an affinity matrix is constructed which indicates the distance between different trajectory groups. An Adaptable Dynamic Hierarchical Clustering (ADHC) tree is proposed to gradually merge the most similar groups and form the trajectory motion patterns, which implements a simpler and more tractable dynamical clustering procedure in updating the clustering results with lower time complexity and avoids the traditional overfitting problem. By using the HMM models generated for the obtained trajectory motion patterns, we may recognize motion patterns and detect anomalies by computing the likelihood of the given trajectory, where a maximum likelihood for HMM indicates a pattern, and a small one below a threshold suggests an anomaly. Experiments are performed on EIFPD trajectory datasets from a structureless scene, where pedestrians choose their walking paths randomly. The experimental results show that our method can accurately learn motion patterns and detect anomalies with better performance.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E97.D.1574/_p

Copy

@ARTICLE{e97-d_6_1574,
author={Kai KANG, Weibin LIU, Weiwei XING, },
journal={IEICE TRANSACTIONS on Information},
title={Motion Pattern Study and Analysis from Video Monitoring Trajectory},
year={2014},
volume={E97-D},
number={6},
pages={1574-1582},
abstract={This paper introduces an unsupervised method for motion pattern learning and abnormality detection from video surveillance. In the preprocessing steps, trajectories are segmented based on their locations, and the sub-trajectories are represented as codebooks. Under our framework, Hidden Markov Models (HMMs) are used to characterize the motion pattern feature of the trajectory groups. The state of trajectory is represented by a HMM and has a probability distribution over the possible output sub-trajectories. Bayesian Information Criterion (BIC) is introduced to measure the similarity between groups. Based on the pairwise similarity scores, an affinity matrix is constructed which indicates the distance between different trajectory groups. An Adaptable Dynamic Hierarchical Clustering (ADHC) tree is proposed to gradually merge the most similar groups and form the trajectory motion patterns, which implements a simpler and more tractable dynamical clustering procedure in updating the clustering results with lower time complexity and avoids the traditional overfitting problem. By using the HMM models generated for the obtained trajectory motion patterns, we may recognize motion patterns and detect anomalies by computing the likelihood of the given trajectory, where a maximum likelihood for HMM indicates a pattern, and a small one below a threshold suggests an anomaly. Experiments are performed on EIFPD trajectory datasets from a structureless scene, where pedestrians choose their walking paths randomly. The experimental results show that our method can accurately learn motion patterns and detect anomalies with better performance.},
keywords={},
doi={10.1587/transinf.E97.D.1574},
ISSN={1745-1361},
month={June},}

Copy

TY - JOUR
TI - Motion Pattern Study and Analysis from Video Monitoring Trajectory
T2 - IEICE TRANSACTIONS on Information
SP - 1574
EP - 1582
AU - Kai KANG
AU - Weibin LIU
AU - Weiwei XING
PY - 2014
DO - 10.1587/transinf.E97.D.1574
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E97-D
IS - 6
JA - IEICE TRANSACTIONS on Information
Y1 - June 2014
AB - This paper introduces an unsupervised method for motion pattern learning and abnormality detection from video surveillance. In the preprocessing steps, trajectories are segmented based on their locations, and the sub-trajectories are represented as codebooks. Under our framework, Hidden Markov Models (HMMs) are used to characterize the motion pattern feature of the trajectory groups. The state of trajectory is represented by a HMM and has a probability distribution over the possible output sub-trajectories. Bayesian Information Criterion (BIC) is introduced to measure the similarity between groups. Based on the pairwise similarity scores, an affinity matrix is constructed which indicates the distance between different trajectory groups. An Adaptable Dynamic Hierarchical Clustering (ADHC) tree is proposed to gradually merge the most similar groups and form the trajectory motion patterns, which implements a simpler and more tractable dynamical clustering procedure in updating the clustering results with lower time complexity and avoids the traditional overfitting problem. By using the HMM models generated for the obtained trajectory motion patterns, we may recognize motion patterns and detect anomalies by computing the likelihood of the given trajectory, where a maximum likelihood for HMM indicates a pattern, and a small one below a threshold suggests an anomaly. Experiments are performed on EIFPD trajectory datasets from a structureless scene, where pedestrians choose their walking paths randomly. The experimental results show that our method can accurately learn motion patterns and detect anomalies with better performance.
ER -

IEICE TRANSACTIONS on Information