Anomaly detection in moving object

Xiaolei Li, Jiawei Han, Sangkyum Kim, Hector Gonzalez

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

With recent advances in sensory and mobile computing technology, many interesting applications involving moving objects have emerged. One of them is identification of suspicious movements : an important problem in homeland security. The objects in question can be vehicles, airplanes, or ships; however, due to the sheer volume of data and the complexities within, manual inspection of the moving objects would require too much manpower. Thus, an automated or semi-automated solution to this problem would be very helpful. That said, it is challenging to develop a method that can efficiently and effectively detect anomalies. The problem is exacerbated by the fact that anomalies may occur at arbitrary levels of abstraction and be associated with multiple granularity of spatiotemporal features. In this study, we propose a new framework named ROAM (Rule- and Motif-based Anomaly Detection in M oving Objects). In ROAM, object trajectories are expressed using discrete pattern fragments called motifs. Associated features are extracted to form a hierarchical feature space, which facilitates a multi-resolution view of the data. We also develop a general-purpose, rule-based classifier which explores the structured feature space and learns effective rules at multiple levels of granularity. Such rules are easily readable and can be easily provided to humans to aid better inspection of moving objects.

Original languageEnglish (US)
Title of host publicationIntelligence and Security Informatics
Subtitle of host publicationTechniques and Applications
EditorsHsinchun Chen, Christopher Yang
Pages357-381
Number of pages25
DOIs
StatePublished - Jul 3 2008

Publication series

NameStudies in Computational Intelligence
Volume135
ISSN (Print)1860-949X

Fingerprint

Inspection
Mobile computing
National security
Ships
Classifiers
Trajectories
Aircraft

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Li, X., Han, J., Kim, S., & Gonzalez, H. (2008). Anomaly detection in moving object. In H. Chen, & C. Yang (Eds.), Intelligence and Security Informatics: Techniques and Applications (pp. 357-381). (Studies in Computational Intelligence; Vol. 135). https://doi.org/10.1007/978-3-540-69209-6_19

Anomaly detection in moving object. / Li, Xiaolei; Han, Jiawei; Kim, Sangkyum; Gonzalez, Hector.

Intelligence and Security Informatics: Techniques and Applications. ed. / Hsinchun Chen; Christopher Yang. 2008. p. 357-381 (Studies in Computational Intelligence; Vol. 135).

Research output: Chapter in Book/Report/Conference proceedingChapter

Li, X, Han, J, Kim, S & Gonzalez, H 2008, Anomaly detection in moving object. in H Chen & C Yang (eds), Intelligence and Security Informatics: Techniques and Applications. Studies in Computational Intelligence, vol. 135, pp. 357-381. https://doi.org/10.1007/978-3-540-69209-6_19
Li X, Han J, Kim S, Gonzalez H. Anomaly detection in moving object. In Chen H, Yang C, editors, Intelligence and Security Informatics: Techniques and Applications. 2008. p. 357-381. (Studies in Computational Intelligence). https://doi.org/10.1007/978-3-540-69209-6_19
Li, Xiaolei ; Han, Jiawei ; Kim, Sangkyum ; Gonzalez, Hector. / Anomaly detection in moving object. Intelligence and Security Informatics: Techniques and Applications. editor / Hsinchun Chen ; Christopher Yang. 2008. pp. 357-381 (Studies in Computational Intelligence).
@inbook{f9fa7c119da240509642af9a4315277d,
title = "Anomaly detection in moving object",
abstract = "With recent advances in sensory and mobile computing technology, many interesting applications involving moving objects have emerged. One of them is identification of suspicious movements : an important problem in homeland security. The objects in question can be vehicles, airplanes, or ships; however, due to the sheer volume of data and the complexities within, manual inspection of the moving objects would require too much manpower. Thus, an automated or semi-automated solution to this problem would be very helpful. That said, it is challenging to develop a method that can efficiently and effectively detect anomalies. The problem is exacerbated by the fact that anomalies may occur at arbitrary levels of abstraction and be associated with multiple granularity of spatiotemporal features. In this study, we propose a new framework named ROAM (Rule- and Motif-based Anomaly Detection in M oving Objects). In ROAM, object trajectories are expressed using discrete pattern fragments called motifs. Associated features are extracted to form a hierarchical feature space, which facilitates a multi-resolution view of the data. We also develop a general-purpose, rule-based classifier which explores the structured feature space and learns effective rules at multiple levels of granularity. Such rules are easily readable and can be easily provided to humans to aid better inspection of moving objects.",
author = "Xiaolei Li and Jiawei Han and Sangkyum Kim and Hector Gonzalez",
year = "2008",
month = "7",
day = "3",
doi = "10.1007/978-3-540-69209-6_19",
language = "English (US)",
isbn = "9783540692072",
series = "Studies in Computational Intelligence",
pages = "357--381",
editor = "Hsinchun Chen and Christopher Yang",
booktitle = "Intelligence and Security Informatics",

}

TY - CHAP

T1 - Anomaly detection in moving object

AU - Li, Xiaolei

AU - Han, Jiawei

AU - Kim, Sangkyum

AU - Gonzalez, Hector

PY - 2008/7/3

Y1 - 2008/7/3

N2 - With recent advances in sensory and mobile computing technology, many interesting applications involving moving objects have emerged. One of them is identification of suspicious movements : an important problem in homeland security. The objects in question can be vehicles, airplanes, or ships; however, due to the sheer volume of data and the complexities within, manual inspection of the moving objects would require too much manpower. Thus, an automated or semi-automated solution to this problem would be very helpful. That said, it is challenging to develop a method that can efficiently and effectively detect anomalies. The problem is exacerbated by the fact that anomalies may occur at arbitrary levels of abstraction and be associated with multiple granularity of spatiotemporal features. In this study, we propose a new framework named ROAM (Rule- and Motif-based Anomaly Detection in M oving Objects). In ROAM, object trajectories are expressed using discrete pattern fragments called motifs. Associated features are extracted to form a hierarchical feature space, which facilitates a multi-resolution view of the data. We also develop a general-purpose, rule-based classifier which explores the structured feature space and learns effective rules at multiple levels of granularity. Such rules are easily readable and can be easily provided to humans to aid better inspection of moving objects.

AB - With recent advances in sensory and mobile computing technology, many interesting applications involving moving objects have emerged. One of them is identification of suspicious movements : an important problem in homeland security. The objects in question can be vehicles, airplanes, or ships; however, due to the sheer volume of data and the complexities within, manual inspection of the moving objects would require too much manpower. Thus, an automated or semi-automated solution to this problem would be very helpful. That said, it is challenging to develop a method that can efficiently and effectively detect anomalies. The problem is exacerbated by the fact that anomalies may occur at arbitrary levels of abstraction and be associated with multiple granularity of spatiotemporal features. In this study, we propose a new framework named ROAM (Rule- and Motif-based Anomaly Detection in M oving Objects). In ROAM, object trajectories are expressed using discrete pattern fragments called motifs. Associated features are extracted to form a hierarchical feature space, which facilitates a multi-resolution view of the data. We also develop a general-purpose, rule-based classifier which explores the structured feature space and learns effective rules at multiple levels of granularity. Such rules are easily readable and can be easily provided to humans to aid better inspection of moving objects.

UR - http://www.scopus.com/inward/record.url?scp=45949111626&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=45949111626&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-69209-6_19

DO - 10.1007/978-3-540-69209-6_19

M3 - Chapter

AN - SCOPUS:45949111626

SN - 9783540692072

T3 - Studies in Computational Intelligence

SP - 357

EP - 381

BT - Intelligence and Security Informatics

A2 - Chen, Hsinchun

A2 - Yang, Christopher

ER -