[From Fla. police cameras analyze thousands of license tags via Sun-Sentinel]
ALPR data provides police with the location of a specific car. This data is typically used to identify stolen vehicles, vehicles associated with an Amber Alert, parking tickets, and congestion charges.
One issue the article raises is how long should this information be stored (and who should have access to it over the long term). After all, the Broward Sheriff’s Office has 35 systems located on patrols cars and fixed locations and scans an average of 300,000 plates a month. The resulting data over a period of months and years can provide significant details about the use of a car.
As the article notes, some cities are storing this data permanently. They may also be sharing this information. This means any police officer can type in a license plate and then have a location history of the vehicle.
There is no bright line between the value of this data for later investigative police work versus its encroachment on privacy. From a gut level, I believe this information should be deleted after 30 days or at most 1 year (depending on what its being used for, e.g. finding stolen cars versus congestion charges) unless it is needed for an investigation. I also think access to this information should be limited, e.g., reasonable suspicion requirement.
rshah ALPR
[From A Chance For Input On 3,000 New Police Cameras- City Limits: News for NYC's Nonprofit, Policy and Activist World]
Ali Winston has another good story on the cameras in NYC. (The details of this network came about from a lawsuit filed by the NY Civil Liberties Union.)
Modeled on London’s 10,000 camera system, called the “Ring of Steel,” the Lower Manhattan Security Initiative (LMSI) – which has attained the same nickname – will consist of 3,000 networked cameras monitoring 1.7 miles of area south of Canal Street. One-third of the cameras will be city-owned, with the other two-thirds belonging to private businesses, termed “stakeholders” in the guidelines. Automated license plate readers and environmental sensors will also provide data to the police. The cameras will be monitored by officers at a coordination center on Broadway, which opened last November. The system is expected to cost $89 million in local and federal funds. [More info at Danger Room]
The NYC network is a study in contrast with Chicago.
First, the NYC network is much more geographically focused system. Using my previous estimates, if lower manhattan is 4 square miles, then it would need about 5000 cameras to ensure a camera is every 50 yards. So the network is highly likely to complete saturate the area. In contrast, Chicago is far from evenly covering the city and even the downtown area which has a much higher concentration of cameras is not near this density.
Second, we know something about the NYC network. Through the NYCLU and NYPD, the public knows about the camera system and is being asked for their input. (The NYPD is asking for comments on its privacy guidelines, something Chicago has never done).
Third, the NYPD is putting into place policies for how the surveillance technologies (cameras and ALPR) will be used. The policies make it seem the focus of these technologies is counterterrorism and not crime prevention. As a result, there are procedural safeguards to using the surveillance data for other purposes. In Chicago, the city has never publicly stated what its policies are regarding security and privacy of surveillance data.
From my view, I am just heartened that the NYPD is publicly developing such a policy.
rshah Other Cities, Policy
[From Traffic cameras could help wipe out city's projected deficit :: CHICAGO SUN-TIMES :: City Hall]
Insurenet proposes that Chicago could make “well in excess of $100 million. We think at least $200 million.” All they would have to do is use traffic cameras to collect license plate numbers and then match the numbers with those of uninsured motorists. A traffic camera would then read a license plate, if it was uninsured, it would then send a ticket in the mail. To accomplish this, Chicago would also need to compel insurance companies to report the names and license plates of insured motorists into the National Law Enforcement Telecommunications System (NLETS), the information-sharing network that links federal, state and local law enforcement agencies.
Does anybody see any problems in this scheme? The readers of the Sun-Times offered quite a few.
1. The numbers don’t seem to add up. Frankly, I couldn’t understand the figures from the newspaper article. But basically, the city would need on the order of 700,000 tickets per year to generate this revenue. Thats a lot of tickets.
2. The levied fines seem disproportional. Fining someone that doesn’t have insurance on the car a $1000? What are the odds that they could either afford to pay or be willing to pay the fine?
This brings me to something called unintended consequences.
People without insurance (they know who they are) will now have a bullseye on their car (aka the license plate). They will have a $1000 incentive to make sure they do not receive a fine. They may try to disguise their plate or avoid traffic cameras. Or they might be willing to go down the slippery slope of illegality by telling the city an incorrect address for their car registration or making sure the license plate on their car isn’t theirs. This is a slope the city doesn’t want people to venture. Once citizens muck with the connection between license plates and their registration, it throws the whole enforcement mechanism for many sorts of issues awry.
I think fines for uninsured motorists are a good use of traffic cameras. But lets use it as a tool to push people to buy insurance and not as a way to unfairly punish people, who will likely become even harder to catch.
rshah ALPR, Red Light Cameras
[From Intelligence officials looking to cull terror info from huge stashes of surveillance tape -- chicagotribune.com]
A research program, Video Analysis and Content Extraction, has been under way since 2001 by the Office of Incisive Analysis, part of the government’s Intelligence Advanced Research Projects Activity. It is spending about $30 million and has several interesting projects related to this blog.
An overall program overview is available here. The VACE program provides funding for innovative, creative, high-risk research to achieve significant advancements in video content extraction technologies. Some listed examples include:
• VACE-funded face detection technology was transferred to NSA, who is providing additional funding to move to a PDA.
• VACE UAV technology transferred to the Joint Warfare Analysis Center enables UAV exploitation.
• Many VACE funded video research technologies are feeding the commercial market. Technologies are being integrated into existing commercial products, such as InforMedia, TerraSight, VideoFOCUS, MARVEL, AlertVideo, Digital Video Manager by companies including CMU, IBM, Honeywell, Salient Stills and Sarnoff Corporation. For example video resolution enhancement technology was transferred to the FBI Forensics via VideoFOCUS diminished time for decision making.
Another online publication, Intelligence, lists the companies by funding area for the phase II research in 2003-2005. There are also many powerpoint presentations online by Dennis Moellman who heads the program, for those that are seeking more information.
rshah General
[From Photo Enforced: License Plate Hunter Cameras]
A post describing ELSAG North America’s Automatic License Plate Recognition (ALPR) technology.
A network of fixed LPR systems, which can be mounted to structures such as bridges and overpasses, as well as mobile LPR systems, which are mounted to police vehicles that can capture up to 3,000 license plates per hour, help keep a tight watch on cities, ports, borders and other sensitive areas. The MPH-900® LPR technology is being used by hundreds of agencies all across the United States to assist with interdiction, the capturing criminal intelligence data.
rshah ALPR, Vendors
[From Study Questions Whether Cameras Cut Crime - City Room Blog - NYTimes.com]
David Greenberg and Jeffrey Roush have published a study on the effectiveness of cameras in a privately owned apartment complex. They found:
CTV may be moderately effective in preventing minor crimes or in diverting them to distant areas. Its effects on more serious crime could not be gauged precisely from this research because there was so little of it prior to the introduction of electronic monitoring.
This is another study that shows cameras are at best lukewarm for reducing crime. However, as the post on DC noted, everyday citizens don’t believe this. I believe surveillance will continue growing rapidly in the US despite the lack of support for a connection between cameras and crime.
rshah Policy
[From How useless are the D.C. Police Department's CCTV crime cameras? - Cover Story - Washington City Paper]
An investigative piece on the cameras in DC. It offers a nice bit of history on the DC cameras which has grown to 5,200 cameras which can be monitored live. I once considered DC a model for other cities based on its openness in regards to polices with cameras. However, as the article shows, the city has become more secretive and aggressive in its use of cameras.
The article discusses the lack of effectiveness of the DC camera network.
In the fall of 2007, in response to a FOIA request by the ACLU, the police admitted that surveillance footage had never been used to make an arrest from the start of the program to March 2007 (when the request was filed). Hughes subsequently insisted the cameras have been useful since then—they provided evidence that contributed to two arrests.
The department’s 2007 annual camera report, released early in 2008, says investigators viewed images 532 times and recovered 144 useful bits of video. One camera captured images the report says became “vital evidence” leading to the arrest of a murder suspect.
The department has not yet released an annual camera report for 2008—not that the year didn’t provide some camera moments worth reporting.
The final bit I liked was the view of citizens for the cameras:
The department’s cameras are good for at least one very significant thing: public relations. Residents routinely demand camera installment near shady alleys and troubled side streets in the wake of crime waves. Last year, the mother of a man who’d been shot to death demanded that the mayor apologize for the city’s failure to catch the perp. If the city can pay for traffic cams, it can pay for crime cams, the woman said.
“What’s the difference in the price to put a camera up to catch someone speeding and putting a camera up to catch a killer?” she asked the mayor.
“Regular citizens want those crime cameras up,” says D.C. police union boss Kris Baumann, who never hesitates to criticize the department. Because citizens like cameras, Baumann says rank-and-file cops like them, too. A new camera and its accompanying signage give citizens a tangible piece of evidence that the department is trying to help the neighborhood.
rshah Other Cities
Recent Comments