DocumentCode :
3537307
Title :
Anomaly detection in videos: A dynamical systems approach
Author :
Surana, Amit ; Nakhmani, Arie ; Tannenbaum, Allen
Author_Institution :
United Technol. Res. Center, East Hartford, CT, USA
fYear :
2013
fDate :
10-13 Dec. 2013
Firstpage :
6489
Lastpage :
6495
Abstract :
We demonstrate a dynamical system framework based on motion patterns for detecting anomalous individual and group behavior in complex videos. We first describe a framework based on trajectory modeling, in which coarse statistical models are used to capture global motion patterns, and are employed in change detection to identify anomalous behavior at the object level. Our multi-target tracking framework combines geometric active contours with particle filtering to effectively deal with occlusions and clutter in the environment. In crowded scenes, however, such object level representation can become extremely unreliable: to deal with this we instead use of low-level motion features (e.g., optical flow) to capture group behavior. To keep the problem tractable, we utilize a subspace system identification method based on the Hankel matrix to extract relevant low order dynamics of these noisy features. The spectral properties of the Hankel matrix encode useful information about the dynamics, and can detect anomalous group behavior. In order to efficiently compute these spectral properties, we employ a randomized algorithm for singular value decomposition. Both approaches are demonstrated to robustly detect anomalous behavior in realistic indoor and outdoor videos.
Keywords :
Hankel matrices; feature extraction; object detection; particle filtering (numerical methods); singular value decomposition; Hankel matrix; anomalous individual behavior; anomaly detection; coarse statistical models; complex videos; dynamical system framework; geometric active contours; global motion patterns; group behavior; low level motion features; low order dynamics; multitarget tracking framework; particle filtering; singular value decomposition; spectral properties; subspace system identification method; trajectory modeling; Estimation; Hidden Markov models; Integrated optics; Monte Carlo methods; Robustness; Tracking; Videos;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Decision and Control (CDC), 2013 IEEE 52nd Annual Conference on
Conference_Location :
Firenze
ISSN :
0743-1546
Print_ISBN :
978-1-4673-5714-2
Type :
conf
DOI :
10.1109/CDC.2013.6760916
Filename :
6760916
Link To Document :
بازگشت