A confabulation model for abnormal vehicle events detection in wide-area traffic monitoring

Qiuwen Chen, Qinru Qiu, Qing Wu, Morgan Bishop, Mark Barnell

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

7 Scopus citations

Abstract

The advanced sensing and imaging technologies of today's digital camera systems provide the capability of monitoring traffic flows in a very large area. In order to provide continuous monitoring and prompt anomaly detection, an abstract-level autonomous anomaly detection model is developed that is able to detect various categories of abnormal vehicle events with unsupervised learning. The method is based on the cogent confabulation model, which performs statistical inference functions in a neuromorphic formulation. The proposed approach covers the partitioning of a large region, training of the vehicle behavior knowledge base and the detection of anomalies according to the likelihood-ratio test. A software version of the system is implemented to verify the proposed model. The experimental results demonstrate the functionality of the detection model and compare the system performance under different configurations.

Original languageEnglish (US)
Title of host publication2014 IEEE International Inter-Disciplinary Conference on Cognitive Methods in Situation Awareness and Decision Support, CogSIMA 2014
PublisherIEEE Computer Society
Pages216-222
Number of pages7
ISBN (Print)9781479935642
DOIs
StatePublished - Jan 1 2014
Event2014 IEEE International Inter-Disciplinary Conference on Cognitive Methods in Situation Awareness and Decision Support, CogSIMA 2014 - San Antonio, TX, United States
Duration: Mar 3 2014Mar 6 2014

Publication series

Name2014 IEEE International Inter-Disciplinary Conference on Cognitive Methods in Situation Awareness and Decision Support, CogSIMA 2014

Other

Other2014 IEEE International Inter-Disciplinary Conference on Cognitive Methods in Situation Awareness and Decision Support, CogSIMA 2014
CountryUnited States
CitySan Antonio, TX
Period3/3/143/6/14

Keywords

  • anomaly detection
  • cogent confabulation
  • intelligent transportatio
  • unsupervised learning

ASJC Scopus subject areas

  • Software

Fingerprint Dive into the research topics of 'A confabulation model for abnormal vehicle events detection in wide-area traffic monitoring'. Together they form a unique fingerprint.

  • Cite this

    Chen, Q., Qiu, Q., Wu, Q., Bishop, M., & Barnell, M. (2014). A confabulation model for abnormal vehicle events detection in wide-area traffic monitoring. In 2014 IEEE International Inter-Disciplinary Conference on Cognitive Methods in Situation Awareness and Decision Support, CogSIMA 2014 (pp. 216-222). [6816565] (2014 IEEE International Inter-Disciplinary Conference on Cognitive Methods in Situation Awareness and Decision Support, CogSIMA 2014). IEEE Computer Society. https://doi.org/10.1109/CogSIMA.2014.6816565