Using inactivity to detect unusual behavior

Dickinson, Patrick and Hunter, Andrew (2008) Using inactivity to detect unusual behavior. In: IEEE Workshop on Motion and video Computing, 2008. WMVC 2008., 8-9 January 2008, Copper Mountain, Colorado, US.

Full content URL:

Using Inactivity to Detect Unusual behavi
Pre-print. Marked 'confidential ... do not redistribute'
[img] PDF
wmvc08pld.pdf - Whole Document
Restricted to Repository staff only

Item Type:Conference or Workshop contribution (Paper)
Item Status:Live Archive


We present a novel method for detecting unusual modes
of behavior in video surveillance data, suitable for supporting home-based care of elderly patients. Our approach is based on detecting unusual patterns of inactivity.
We first learn a spatial map of normal inactivity for an observedscene, expressed as a two-dimensional mixture of Gaus-sians. The map components are used to construct a HiddenMarkov Model representing normal patterns of behavior. Athreshold model is also inferred, and unusual behavior de-tected by comparing the model likelihoods. Our learning procedures are unsupervised, and yield a highly transparent model of scene activity. We present an evaluation of our pproach, and show that it is effective in detecting unusual bhavior across a range of parameter settings.

Keywords:computer vision, Behaviour, Pattern recognition, Automated
Subjects:G Mathematical and Computer Sciences > G740 Computer Vision
Divisions:College of Science > School of Computer Science
Related URLs:
ID Code:1753
Deposited On:16 Jan 2009 09:43

Repository Staff Only: item control page