Effective Crowd Anomaly Detection Through Spatio-temporal Texture Analysis

Yu Hao, Zhijie Xu, Ying Liu, Jing Wang, Jiulun Fan

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Abnormal crowd behaviors in high density situations can pose great danger to public safety. Despite the extensive installation of closed-circuit television (CCTV) cameras, it is still difficult to achieve real-time alerts and automated responses from current systems. Two major breakthroughs have been reported in this research. Firstly, a spatial-temporal texture extraction algorithm is developed. This algorithm is able to effectively extract video textures with abundant crowd motion details. It is through adopting Gabor-filtered textures with the highest information entropy values. Secondly, a novel scheme for defining crowd motion patterns (signatures) is devised to identify abnormal behaviors in the crowd by employing an enhanced gray level co-occurrence matrix model. In the experiments, various classic classifiers are utilized to benchmark the performance of the proposed method. The results obtained exhibit detection and accuracy rates which are, overall, superior to other techniques.

LanguageEnglish
Pages27-39
Number of pages13
JournalInternational Journal of Automation and Computing
Volume16
Issue number1
Early online date27 Sep 2018
DOIs
Publication statusPublished - Feb 2019

Fingerprint

Texture Analysis
Anomaly Detection
Texture
Textures
Gray Level Co-occurrence Matrix
Information Entropy
Motion
Matrix Models
Video cameras
Classifiers
Entropy
Signature
Safety
Camera
Classifier
Benchmark
Real-time
Closed
Networks (circuits)
Experiment

Cite this

Hao, Yu ; Xu, Zhijie ; Liu, Ying ; Wang, Jing ; Fan, Jiulun. / Effective Crowd Anomaly Detection Through Spatio-temporal Texture Analysis. In: International Journal of Automation and Computing. 2019 ; Vol. 16, No. 1. pp. 27-39.
@article{9081169d43174478940ea198c022a439,
title = "Effective Crowd Anomaly Detection Through Spatio-temporal Texture Analysis",
abstract = "Abnormal crowd behaviors in high density situations can pose great danger to public safety. Despite the extensive installation of closed-circuit television (CCTV) cameras, it is still difficult to achieve real-time alerts and automated responses from current systems. Two major breakthroughs have been reported in this research. Firstly, a spatial-temporal texture extraction algorithm is developed. This algorithm is able to effectively extract video textures with abundant crowd motion details. It is through adopting Gabor-filtered textures with the highest information entropy values. Secondly, a novel scheme for defining crowd motion patterns (signatures) is devised to identify abnormal behaviors in the crowd by employing an enhanced gray level co-occurrence matrix model. In the experiments, various classic classifiers are utilized to benchmark the performance of the proposed method. The results obtained exhibit detection and accuracy rates which are, overall, superior to other techniques.",
keywords = "Crowd Behavior, Spatial-Temporal Texture, Gray Level Co-occurrence Matrix, Information Entropy",
author = "Yu Hao and Zhijie Xu and Ying Liu and Jing Wang and Jiulun Fan",
year = "2019",
month = "2",
doi = "10.1007/s11633-018-1141-z",
language = "English",
volume = "16",
pages = "27--39",
journal = "International Journal of Automation and Computing",
issn = "1476-8186",
publisher = "Chinese Academy of Sciences",
number = "1",

}

Effective Crowd Anomaly Detection Through Spatio-temporal Texture Analysis. / Hao, Yu; Xu, Zhijie; Liu, Ying; Wang, Jing; Fan, Jiulun.

In: International Journal of Automation and Computing, Vol. 16, No. 1, 02.2019, p. 27-39.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Effective Crowd Anomaly Detection Through Spatio-temporal Texture Analysis

AU - Hao, Yu

AU - Xu, Zhijie

AU - Liu, Ying

AU - Wang, Jing

AU - Fan, Jiulun

PY - 2019/2

Y1 - 2019/2

N2 - Abnormal crowd behaviors in high density situations can pose great danger to public safety. Despite the extensive installation of closed-circuit television (CCTV) cameras, it is still difficult to achieve real-time alerts and automated responses from current systems. Two major breakthroughs have been reported in this research. Firstly, a spatial-temporal texture extraction algorithm is developed. This algorithm is able to effectively extract video textures with abundant crowd motion details. It is through adopting Gabor-filtered textures with the highest information entropy values. Secondly, a novel scheme for defining crowd motion patterns (signatures) is devised to identify abnormal behaviors in the crowd by employing an enhanced gray level co-occurrence matrix model. In the experiments, various classic classifiers are utilized to benchmark the performance of the proposed method. The results obtained exhibit detection and accuracy rates which are, overall, superior to other techniques.

AB - Abnormal crowd behaviors in high density situations can pose great danger to public safety. Despite the extensive installation of closed-circuit television (CCTV) cameras, it is still difficult to achieve real-time alerts and automated responses from current systems. Two major breakthroughs have been reported in this research. Firstly, a spatial-temporal texture extraction algorithm is developed. This algorithm is able to effectively extract video textures with abundant crowd motion details. It is through adopting Gabor-filtered textures with the highest information entropy values. Secondly, a novel scheme for defining crowd motion patterns (signatures) is devised to identify abnormal behaviors in the crowd by employing an enhanced gray level co-occurrence matrix model. In the experiments, various classic classifiers are utilized to benchmark the performance of the proposed method. The results obtained exhibit detection and accuracy rates which are, overall, superior to other techniques.

KW - Crowd Behavior

KW - Spatial-Temporal Texture

KW - Gray Level Co-occurrence Matrix

KW - Information Entropy

UR - http://www.scopus.com/inward/record.url?scp=85054196111&partnerID=8YFLogxK

U2 - 10.1007/s11633-018-1141-z

DO - 10.1007/s11633-018-1141-z

M3 - Article

VL - 16

SP - 27

EP - 39

JO - International Journal of Automation and Computing

T2 - International Journal of Automation and Computing

JF - International Journal of Automation and Computing

SN - 1476-8186

IS - 1

ER -