Councilman Phil Mendelson questions the effectiveness and cost of D.C. Government plans to give police access to 5262 surveillance cameras through a network managed by the city’s Homeland Security and Emergency Management Agency (HSEMA). Mr. Mendelson will schedule public hearings on the project, called Video Interoperability for Public Safety (VIPS).
Closed-circuit TV surveillance (CCTV) is an expensive technology, of limited use in preventing violent crime. It does seem to reduce property crime rates, but is chiefly used after the fact, in identifying and convicting offenders. CCTV is expensive to install and labor-intensive if real-time monitoring is required.
HSEMA has little experience with CCTV, and currently operates only four (4) cameras. Is this agency coordinating the VIPS effort because it has $9 million in unspent grant funds?
While D.C. Chief Cathy Lanier maintains that Metropolitan Police Department (MPD) cameras have reduced violent crime in “hot spots,” every systematic study of CCTV and violent crime prevention has shown otherwise, including studies cited by the U.S. Department of Justice and British sources. Perhaps the results cited by Chief Lanier are due to other measures taken simultaneously. When it comes to preventing crime, even improved street lighting works as well as CCTV surveillance.
Software companies currently market products to assist video monitoring, but automation of real-time video surveillance is a chimera. According to an email from Professor Larry S. Davis of the Computer Vision Laboratory (and Chair of the Department of Computer Science) at the University of Maryland, College Park:
If they plan any type of online monitoring of these cameras in crowded urban environments using automation, then they will be disappointed (in both price and performance). Today’s commercially available surveillance video analysis systems are best suited for detecting tripwire constraints (people entering forbidden areas), and not on complex activity analysis of people. And, as you correctly conclude, if they plan to monitor a large fraction of these cameras manually they will need a small army of operators. Typically, data are recorded from such cameras at low frame rates and archived for forensic purposes, and a small number of cameras are monitored in critical areas in real time.
The VIPS project is being sold to the public as a crime-prevention measure, not a crime-solving one. If the cameras are meant to capture records of crime for use in identifying and convicting perpetrators, few live-access monitors are needed, just image analysts. If crime prevention is the goal, the city needs to re-program the VIPS money, hire more police officers, and put them on the street, not in front of video screens.
Image by Mike Licht. Download a copy here. Creative Commons license. Credit: Mike Licht, NotionsCapital.com
April 27, 2008 at 5:26 am
I am all for CCTV it sure is a great deterrent for those wanting to break the law. Expensive yes but is needed. Our prisons can not accommodate all the those convicted of wrong doings and are released before their sentence has been carried out. CCTV will cut out these sentences being imposed so therefore no worries over criminals let out to commit further crimes. (Nip it in the bud)
April 27, 2008 at 7:42 pm
Des:
Read more closely. Despite all the tax money UK citizens like you have paid, CCTV does not prevent violent crime and is even less effective in the USA. It also means one pays civil servants to watch television instead of hiring more police officers to patrol the streets.
May 8, 2008 at 3:49 pm
[...] this sound like the VIPS system proposed for Washington [...]
May 27, 2008 at 1:39 pm
[...] occur, like when people run,” says Gaithersburg Police Chief John King. Experts in the field conclude meaningful real-time crime-prevention video analysis is beyond the scope of current software. The [...]