Publication Type

Journal Article

Version

publishedVersion

Publication Date

4-2017

Abstract

Anomaly detection is still a challenging task for video surveillance due to complex environments and unpredictable human behaviors. Most existing approaches train offline detectors using manually labeled data and predefined parameters, and are hard to model changing scenes. This paper introduces a neural network based model called online Growing Neural Gas (online GNG) to perform an unsupervised learning. Unlike a parameter-fixed GNG, our model updates learning parameters continuously, for which we propose several online neighbor-related strategies. Specific operations, namely neuron insertion, deletion, learning rate adaptation and stopping criteria selection, get upgraded to online modes. In the anomaly detection stage, the behavior patterns far away from our model are labeled as anomalous, for which far away is measured by a time varying threshold. Experiments are implemented on three surveillance datasets, namely UMN, UCSD Ped1/Ped2 and Avenue dataset. All datasets have changing scenes due to mutable crowd density and behavior types. Anomaly detection results show that our model can adapt to the current scene rapidly and reduce false alarms while still detecting most anomalies. Quantitative comparisons with 12 recent approaches further confirm our superiority.

Keywords

Anomaly detection, Video surveillance, Unsupervised learning

Discipline

Computer Engineering | Databases and Information Systems

Research Areas

Data Science and Engineering

Publication

Pattern Recognition

Volume

64

First Page

187

Last Page

201

ISSN

0031-3203

Identifier

10.1016/j.patcog.2016.09.016

Publisher

Elsevier

Additional URL

https://doi.org/10.1016/j.patcog.2016.09.016

Download

Share

COinS