On the essence of unsupervised detection of anomalous motion in surveillance videos

Abdullah A. Abuolaim, Wee Kheng Leow, Jagannadan Varadarajan, Narendra Ahuja

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

An important application in surveillance is to apply computerized methods to automatically detect anomalous activities and then notify the security officers. Many methods have been proposed for anomaly detection with varying degree of accuracy. They can be characterized according to the approach adopted, which is supervised or unsupervised, and the features used. Unfortunately, existing literature has not elucidated the essential ingredients that make the methods work as they do, despite the fact that tests have been conducted to compare the performance of various methods. This paper attempts to fill this knowledge gap by studying the videos tested by existing methods and identifying key components required by an effective unsupervised anomaly detection algorithm. Our comprehensive test results show that an unsupervised algorithm that captures the key components can be relatively simple and yet perform equally well or better compared to existing methods.

Original languageEnglish (US)
Title of host publicationComputer Analysis of Images and Patterns - 17th International Conference, CAIP 2017, Proceedings
EditorsAnders Heyden, Michael Felsberg, Norbert Kruger
PublisherSpringer-Verlag
Pages160-171
Number of pages12
ISBN (Print)9783319646886
DOIs
StatePublished - Jan 1 2017
Event17th International Conference on Computer Analysis of Images and Patterns, CAIP 2017 - Ystad, Sweden
Duration: Aug 22 2017Aug 24 2017

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10424 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other17th International Conference on Computer Analysis of Images and Patterns, CAIP 2017
CountrySweden
CityYstad
Period8/22/178/24/17

Fingerprint

Video Surveillance
Anomalous
Motion
Anomaly Detection
Surveillance

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Abuolaim, A. A., Leow, W. K., Varadarajan, J., & Ahuja, N. (2017). On the essence of unsupervised detection of anomalous motion in surveillance videos. In A. Heyden, M. Felsberg, & N. Kruger (Eds.), Computer Analysis of Images and Patterns - 17th International Conference, CAIP 2017, Proceedings (pp. 160-171). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10424 LNCS). Springer-Verlag. https://doi.org/10.1007/978-3-319-64689-3_13

On the essence of unsupervised detection of anomalous motion in surveillance videos. / Abuolaim, Abdullah A.; Leow, Wee Kheng; Varadarajan, Jagannadan; Ahuja, Narendra.

Computer Analysis of Images and Patterns - 17th International Conference, CAIP 2017, Proceedings. ed. / Anders Heyden; Michael Felsberg; Norbert Kruger. Springer-Verlag, 2017. p. 160-171 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10424 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abuolaim, AA, Leow, WK, Varadarajan, J & Ahuja, N 2017, On the essence of unsupervised detection of anomalous motion in surveillance videos. in A Heyden, M Felsberg & N Kruger (eds), Computer Analysis of Images and Patterns - 17th International Conference, CAIP 2017, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10424 LNCS, Springer-Verlag, pp. 160-171, 17th International Conference on Computer Analysis of Images and Patterns, CAIP 2017, Ystad, Sweden, 8/22/17. https://doi.org/10.1007/978-3-319-64689-3_13
Abuolaim AA, Leow WK, Varadarajan J, Ahuja N. On the essence of unsupervised detection of anomalous motion in surveillance videos. In Heyden A, Felsberg M, Kruger N, editors, Computer Analysis of Images and Patterns - 17th International Conference, CAIP 2017, Proceedings. Springer-Verlag. 2017. p. 160-171. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-64689-3_13
Abuolaim, Abdullah A. ; Leow, Wee Kheng ; Varadarajan, Jagannadan ; Ahuja, Narendra. / On the essence of unsupervised detection of anomalous motion in surveillance videos. Computer Analysis of Images and Patterns - 17th International Conference, CAIP 2017, Proceedings. editor / Anders Heyden ; Michael Felsberg ; Norbert Kruger. Springer-Verlag, 2017. pp. 160-171 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{0fd5270bf9154a65bc9e22973e721415,
title = "On the essence of unsupervised detection of anomalous motion in surveillance videos",
abstract = "An important application in surveillance is to apply computerized methods to automatically detect anomalous activities and then notify the security officers. Many methods have been proposed for anomaly detection with varying degree of accuracy. They can be characterized according to the approach adopted, which is supervised or unsupervised, and the features used. Unfortunately, existing literature has not elucidated the essential ingredients that make the methods work as they do, despite the fact that tests have been conducted to compare the performance of various methods. This paper attempts to fill this knowledge gap by studying the videos tested by existing methods and identifying key components required by an effective unsupervised anomaly detection algorithm. Our comprehensive test results show that an unsupervised algorithm that captures the key components can be relatively simple and yet perform equally well or better compared to existing methods.",
author = "Abuolaim, {Abdullah A.} and Leow, {Wee Kheng} and Jagannadan Varadarajan and Narendra Ahuja",
year = "2017",
month = "1",
day = "1",
doi = "10.1007/978-3-319-64689-3_13",
language = "English (US)",
isbn = "9783319646886",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer-Verlag",
pages = "160--171",
editor = "Anders Heyden and Michael Felsberg and Norbert Kruger",
booktitle = "Computer Analysis of Images and Patterns - 17th International Conference, CAIP 2017, Proceedings",

}

TY - GEN

T1 - On the essence of unsupervised detection of anomalous motion in surveillance videos

AU - Abuolaim, Abdullah A.

AU - Leow, Wee Kheng

AU - Varadarajan, Jagannadan

AU - Ahuja, Narendra

PY - 2017/1/1

Y1 - 2017/1/1

N2 - An important application in surveillance is to apply computerized methods to automatically detect anomalous activities and then notify the security officers. Many methods have been proposed for anomaly detection with varying degree of accuracy. They can be characterized according to the approach adopted, which is supervised or unsupervised, and the features used. Unfortunately, existing literature has not elucidated the essential ingredients that make the methods work as they do, despite the fact that tests have been conducted to compare the performance of various methods. This paper attempts to fill this knowledge gap by studying the videos tested by existing methods and identifying key components required by an effective unsupervised anomaly detection algorithm. Our comprehensive test results show that an unsupervised algorithm that captures the key components can be relatively simple and yet perform equally well or better compared to existing methods.

AB - An important application in surveillance is to apply computerized methods to automatically detect anomalous activities and then notify the security officers. Many methods have been proposed for anomaly detection with varying degree of accuracy. They can be characterized according to the approach adopted, which is supervised or unsupervised, and the features used. Unfortunately, existing literature has not elucidated the essential ingredients that make the methods work as they do, despite the fact that tests have been conducted to compare the performance of various methods. This paper attempts to fill this knowledge gap by studying the videos tested by existing methods and identifying key components required by an effective unsupervised anomaly detection algorithm. Our comprehensive test results show that an unsupervised algorithm that captures the key components can be relatively simple and yet perform equally well or better compared to existing methods.

UR - http://www.scopus.com/inward/record.url?scp=85028510966&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85028510966&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-64689-3_13

DO - 10.1007/978-3-319-64689-3_13

M3 - Conference contribution

AN - SCOPUS:85028510966

SN - 9783319646886

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 160

EP - 171

BT - Computer Analysis of Images and Patterns - 17th International Conference, CAIP 2017, Proceedings

A2 - Heyden, Anders

A2 - Felsberg, Michael

A2 - Kruger, Norbert

PB - Springer-Verlag

ER -