Deep learning and one-class SVM based anomalous crowd detection
Version 2 2024-06-04, 06:15Version 2 2024-06-04, 06:15
Version 1 2019-10-31, 11:48Version 1 2019-10-31, 11:48
conference contribution
posted on 2024-06-04, 06:15 authored by M Yang, Sutharshan RajasegararSutharshan Rajasegarar, SM Erfani, C Leckie© 2019 IEEE. Anomalous event detection in videos is an important and challenging task. This paper proposes a deep representation approach to the problem, which extracts and represents features in an unsupervised way. This algorithm can detect anomalous activity like standing statically and loitering among a crowd of people. Our proposed framework is a two-channel scheme by using feature channels extracted from the appearance and foreground of the original video. Two hybrid deep learning architectures SDAE-DBN-PSVM (a four-layer Stacked Denoising Auto-encoder with three-layer Deep Belief Nets and Plane-based one class SVM) are implemented for these two channels to learn the high-level feature representation automatically and produce two anomaly scores. Finally, a fusion scheme is proposed for combining anomaly scores and detecting anomalous events. Experimental results on a large real-world dataset (MCG) and two benchmark datasets (UCSD and Subway) demonstrate the effectiveness of this approach. Furthermore, quantitative analyses of the effects of the amount of training data and the illumination conditions of the video on the accuracy of anomaly detection are presented.
History
Volume
2019-JulyLocation
Budapest, HungaryStart date
2019-07-14End date
2019-07-19ISBN-13
9781728119854Language
engPublication classification
E1 Full written paper - refereedTitle of proceedings
IJCNN 2019 : International Joint Conference on Neural NetworksEvent
Neural Networks. International Joint Conference (2019 : Budapest, Hungary)Issue
N-19570Publisher
IEEEPlace of publication
Piscataway, N.J.Usage metrics
Categories
No categories selectedLicence
Exports
RefWorksRefWorks
BibTeXBibTeX
Ref. managerRef. manager
EndnoteEndnote
DataCiteDataCite
NLMNLM
DCDC