Skip to main content
placeholder image

Local estimation of displacement density for abnormal behavior detection

Conference Paper


Download full-text (Open Access)

Abstract


  • Detecting abnormal behavior in video sequences has become a crucial task with the development of automatic video-surveillance systems. Here, we propose an algorithm which locally models the probability distribution of objects behavioral features. A temporal Gaussian mixture with local update is introduced to estimate the local probability distribution. The update of the feature probability distribution is thus temporal and local, allowing a smooth transition for neighboring locations. The integration of local information in the estimation provides a fast adaptation along with an efficient discrimination between normal and abnormal behavior. The proposed technique is evaluated on both synthetic and real data. Synthetic data simulates different scenarios occurring in road traffic, and illustrates how the model adapts to local conditions. Real data demonstrates the ability of the system to detect abnormal behavior due to the presence of pedestrians and animals on highways. In all tested scenarios the system identifies abnormal and normal behavior correctly.

Publication Date


  • 2008

Citation


  • P. Bouttefroy, A. Bouzerdoum, S. Phung & A. Beghdadi, "Local estimation of displacement density for abnormal behavior detection," in Workshop on Machine Learning for Signal Processing, 2008, pp. 386-391.

Scopus Eid


  • 2-s2.0-58049166509

Ro Full-text Url


  • http://ro.uow.edu.au/cgi/viewcontent.cgi?article=10493&context=infopapers

Ro Metadata Url


  • http://ro.uow.edu.au/infopapers/3157

Has Global Citation Frequency


Start Page


  • 386

End Page


  • 391

Place Of Publication


  • Cancun, Mexico

Abstract


  • Detecting abnormal behavior in video sequences has become a crucial task with the development of automatic video-surveillance systems. Here, we propose an algorithm which locally models the probability distribution of objects behavioral features. A temporal Gaussian mixture with local update is introduced to estimate the local probability distribution. The update of the feature probability distribution is thus temporal and local, allowing a smooth transition for neighboring locations. The integration of local information in the estimation provides a fast adaptation along with an efficient discrimination between normal and abnormal behavior. The proposed technique is evaluated on both synthetic and real data. Synthetic data simulates different scenarios occurring in road traffic, and illustrates how the model adapts to local conditions. Real data demonstrates the ability of the system to detect abnormal behavior due to the presence of pedestrians and animals on highways. In all tested scenarios the system identifies abnormal and normal behavior correctly.

Publication Date


  • 2008

Citation


  • P. Bouttefroy, A. Bouzerdoum, S. Phung & A. Beghdadi, "Local estimation of displacement density for abnormal behavior detection," in Workshop on Machine Learning for Signal Processing, 2008, pp. 386-391.

Scopus Eid


  • 2-s2.0-58049166509

Ro Full-text Url


  • http://ro.uow.edu.au/cgi/viewcontent.cgi?article=10493&context=infopapers

Ro Metadata Url


  • http://ro.uow.edu.au/infopapers/3157

Has Global Citation Frequency


Start Page


  • 386

End Page


  • 391

Place Of Publication


  • Cancun, Mexico