PUB529 Light camera action and arrest.pdf (66.43 kB)
Light, camera, action and arrest
online resource
posted on 2008-04-22, 15:04 authored by Iain T. Darker, Alastair Gale, Leila Ward, Anastassia Blechko, Kevin PurdyGun crime is a fast growing problem in the UK and it is important to detect a
potential gun crime before an incident occurs. A possible technological and
cost-effective approach is to utilise the widespread and endemic installation of
CCTV cameras to automatically recognise individuals carrying concealed
weapons and so prompt the CCTV operator. Current machine imaging software
can identify a range of suspicious behaviours but with varying accuracy and
associated false alarms. CCTV operators learn to identify certain cues
associated with suspicious behaviour, again with varying accuracy. In a new
EPSRC supported research project, MEDUSA sets out to identify both human
and machine detected cues of individuals carrying concealed guns and merge
these into new software for use by CCTV operators. This paper concentrates on
the identification of cues associated with carrying concealed weapons and sets
out the ergonomic challenges surrounding such an approach, together with the
potential ways to overcome them.
History
School
- Science
Department
- Computer Science
Citation
DARKER, I.T. ... et al, 2007. Light, camera, action and arrest. IN: Contemporary Ergonomics 2007 : Proceedings of the International Conference on Contemporary Ergonomics (CE2007), 17-19 April 2007, Nottingham, UK, pp. 171-177Publisher
© Taylor & FrancisPublication date
2007Notes
This definitive version of this conference paper is available from http://www.taylorandfrancis.co.uk/ISBN
9780415436380;0415436389Language
- en