This was one of a series of articles I wrote on police ANPR. It eventually led to a Guardian page lead story in August 2012 – based on the information gleaned from the eventual failure of a Freedom of Information request to Devon and Cornwall Constabulary, which revealed a “patchy” national system – and an article for MATTER published in August 2013 which I co-edited.
The cameras mentioned on the M5 were still in place at the end of August 2013. Following the loss of the FoI case there is, of course, no way of being sure if they are run by Devon and Cornwall Constabulary (whose area they are in).
Automatic numberplate recognition (ANPR) is almost inescapable in cities and on major roads in Britain. There are thousands of cameras equipped with software to read numberplates automatically.
Many of these cameras are less of an intrusion on privacy than may be feared: the 6,600 ANPR traffic monitoring cameras run by the Highways Agency and Trafficmaster do not transmit numberplates. But according to the Association of Chief Police Officers (Acpo), there are 10,502 ANPR cameras collecting data for police forces – including cameras run by councils, which pass data on – which transmit full registration numbers, and in some cases photos of drivers and passengers, with the former held for two years at the National ANPR Data Centre (NADC).
The centre, which is run by the National Policing Improvement Agency on behalf of Acpo, currently records 10m-14m numberplate spottings a day from nearly all forces in England and Wales. The fact that the system has been designed to handle 50m spottings suggests there is plenty of potential for growth.
NADC, which becomes fully operational later this year, stores both the numberplate and the digital images from which the data has been derived for two years. Acpo says there are no plans to extend this retention period to five years, a figure quoted in earlier police documents on plans for the centre.
“Acpo and the NPIA are currently working with the Information Commissioner’s Office to ensure that data retention is appropriate and proportionate,” says a spokesperson, adding that some forces also store photographs of drivers and front seat passengers for up to a year, although on local systems rather than the national database.
As with the National DNA Database, the legal data controllers are the chief constables of the forces which gathered the data.
ANPR cameras have been in use for more than two decades. According to Denying Criminals Use of the Roads, a 2004 document from Acpo’s National ANPR User Group, the technology was developed in the 1980s when covert devices were used for counter-terrorism. City of London Police installed such cameras in a ring around the city in 1997, and by 1999 15 forces were using the technology. By 2002, all in England and Wales had adopted it.
Some forces have since moved from using a few mobile units to a network of fixed cameras. Transport for London passes the ANPR data from its congestion charge cameras around central London to the Metropolitan Police. In 2008, Greater Manchester Police said it was setting up 12 fixed cameras around the city centre to record 660,000 journeys through the area every day. A similar ring of police cameras is in place around the centre of Birmingham.
Senior police officers say the technology is important in tackling crime, and it has been used in some high profile court cases, including terrorism trials. It can also be used for vehicle related crime: as well as storing the data, the NADC analyses it, An example is looking for the same plate apparently moving from one part of the country to another at an impossible speed, implying a cloned plate.
It is also used in more controversial circumstances: ANPR systems have led to police officers stopping a hire car which was used to steal petrol weeks earlier, someone whose insurer failed to remove her from a database of uninsured drivers, and anti-war campaigners whose vehicle had been added to a watchlist after police noted it at protests.
Leaving aside its proportionality, there is question over how effective ANPR can be in much of the country. Department for Transport research says that road vehicles in Britain cover around 866m miles every day, implying a driver would currently be captured once every 72 miles. However, most of the cameras seem to be in cities, with very few elsewhere.
The locations of the cameras are, in most cases, a secret. GC made Freedom of Information requests to four non-metropolitan forces to disclose the location of their fixed ANPR cameras and their spending on the technology. All refused on the former, and just one force provided information on the latter.
Thames Valley Police replied that it had spent £1.94m on ANPR by October 2009, and that this will have reached £2.6m by March 2010. It has requested further funds for the following three years, although these have not yet been allocated. Last August, it awarded a framework contract, worth up to £10m, to five firms for ANPR equipment and services.
Arrests and recoveries
Thames Valley, which describes itself as one of the leading forces in the country in its use of ANPR, recently opened a unit to monitor the cameras’ output 24 hours a day. “In the last two months, the unit has been directly responsible for 53 arrests, the recovery of 26 stolen motor vehicles and recovery of property to the value of almost a quarter of a million pounds,” says ANPR manager John Knight. “It is anticipated these figures will continue to rise.”
He adds that the camera network reduces the need for on-the-ground surveillance, in turn cutting costs and risk to officers, along with data analysis of suspects’ movements. “It is difficult to show this in financial savings, but Thames Valley Police is fully committed to the use of ANPR and believe it is an invaluable tool in the fight against crime,” Knight says.
Last year, the force said it had 47 fixed ANPR systems in place, and planned to add 34 more, each costing £35,000. But even when it has completed this, the force will have 81 fixed ANPR systems for all of Berkshire, Buckinghamshire and Oxfordshire, a large chunk of south-east England covering 2.1m people. Particularly outside cities, anyone who does a little visual research should be able to avoid driving past such cameras if they wish – presumably including criminals.
One set of ANPR cameras in the Thames Valley area provide an example of how easy it can be to avoid being spotted: they are installed on the main road through a town, but are the only roadside cameras of any kind in the area, and are easy to get around.
Observation suggests that the same may also be true on many motorways. ANPR cameras used for traffic monitoring only need to capture a sample of numberplates to calculate average speeds, so they tend to monitor just one lane. (They also have a distinctive colour and shape.) They are very common, whereas multiple-lane camera installations – presumably used by the police to capture traffic in every lane – are very rare, with just one set visible on the 100 miles of the M5 between Cheltenham and Exeter.
Except when used in rings such as around big cities, the fixed ANPR cameras used by the police are reliant to some extent on the position of highly visible equipment remaining a secret. Security experts call this ‘security by obscurity,’ with many arguing that a system which relies on poorly guarded secrecy is likely to be of limited value, which will fall as the ‘secret’ spreads.
Police secrecy over locations and, in many cases, spending makes the usefulness of fixed ANPR cameras hard to assess. But in at least one area, that assessment is starting to be made: in December, Thames Valley Police Authority asked its force to justify the planned further spending on ANPR on value for money grounds. The force was planning to produce a report on this during January.
guardian.co.uk © Guardian News & Media Limited 2010