An Australian computer scientist has declared CCTV camera systems meet the definition of being 'legally blind' when it comes to precise identification of criminals. Billions of dollars have been spent worldwide on systems that should come with a sign showing a white cane and dark sunglasses attached.

Continue reading »...

CCTV is 'legally blind' - academic

An academic paper that provides a technical assessment of the effectiveness of surveillance cameras - CCTV - for identifying people has concluded that the cameras are legally blind.

"Surveillance cameras, as they are currently used, are almost useless for the identification of people. [Indeed], surveillance cameras ... arguably meet the definition of being legally blind," says Peter Kovesi of the School of Computer Science and Software Engineering at the University of Western Australia.

He is a computer scientist who has spent "many frustrating years attempting to assist police with the enhancement of numerous surveillance video images".

According to the Chair of the Australian Privacy Foundation, Prof Roger Clarke, the paper is highly valuable support for the efforts of people who are endeavouring to force proper evaluations of CCTV proposals and installed schemes, and withdrawal of those that cannot be demonstrated to be justified".

(NOTE: these are large files to download, due to numerous images)
Kovesi P. (2009) 'Video Surveillance: Legally Blind? ' Proc. Aust. Pattern Recognition Society conference (DICTA09) pp. 204-211, IEEE Computer Society, December 2009, at http://www.csse.uwa.edu.au/~pk/Research/pkpapers/legallyblind.pdf

PowerPoint slides at: http://www.csse.uwa.edu.au/~pk/Research/pkpapers/LegallyBlind.ppt