Speed Cameras, ANPR and Project Columbus - 2/9/2010

- The expansion of automated checkpoints around the UK

Data Protection expert Chris Pounder of Amberhawk Training[1] has warned that moves by UK local authorities to remove speed cameras could lead to an increase in Automatic Number Plate Recognition or ANPR cameras. In a recent blog post 'Data Protection and surveillance: swapping the speed camera for ANPR?'[2] Pounder suggests that as speed cameras are removed, more accidents could occur so that over time, there will be increased public pressure to do something to counter the rising accident rate, and he says: "ANPR installations (which only need a few cameras) will be the technological fix of choice". Pounder goes on:

In this way, specific surveillance of an accident black spot by a speed camera (which only captures the image of speeding cars breaking the law) is replaced by general surveillance of all vehicles passing the cameras (where records of date, time, driver details, location are all retained for possibly up to 5 years). Perhaps privacy advocates who are against ANPR should start having public love-ins with privacy-friendly speed cameras near their home?

Pounder warns that the CCTV industry views ANPR cameras as an expanding market, he says:

ANPR technology like CCTV will become ubiquitous and already CCTV installers are making quotes such as: “ANPR is probably the next growth product to take off in the UK, and the world; in fact, it is already beginning to be the biggest potential earner for installation companies”

The expansion of the ANPR network has been part of an agenda called Project Columbus for some time and now it seems cost-savings in speed cameras is another way to further the project. How has this come about, and where could it be going?

Spending cuts

The Secretary of State for Communities and Local Government announced reductions to road safety funding for 2010/11[3] and in June the Department for Transport wrote to all local authorities in England and Wales[4] informing them that after this financial year central government will no longer fund new fixed speed cameras. The letter states:

It is clear from the evidence that speed cameras, in the right place, can be an effective way of helping manage safety risks. But my view is that to date local partners have been over-reliant on cameras to achieve road safety outcomes. The number of cameras has increased greatly over the last decade and many motorists feel they are being unfairly and indiscriminantly targeted to generate income from fines.

The letter goes on to point out that local councils should find ways to use speed cameras more efficiently:

for camera operations, I would encourage you to examine with the police the scope for efficiency savings in back office functions, including through sharing them across several police force areas.

The over-reliance on speed cameras referred to in the Department of Transport letter is supported by a study published in 2005 that looked at the impact of cameras. However, whenever police and/or government criticise camera technology it is always worth looking a little closer to see whether there are other factors at play and what the consequences could be.

GATSO benefits overstated

The current generation of speed cameras on UK roads consists predominantly of Gatsometer BV (GATSO)[5] brand cameras that take two still photographs of a speeding car as it passes over white lines painted on the road. In December 2005 the Department for Transport published 'The national safety camera programme Four-year evaluation report' prepared by UCL and PA Consulting Group[6]. The report looked at the number of personal injury collisions (PICs) and fatal or serious collisions (FSC) in 38 road safety partnerships in the UK to assess the impact of speed cameras. The headline figure used in the Executive Summary of the report was that 42% less people were killed or injured due to the installation of speed cameras:

Overall 42% fewer people were killed or seriously injured. At camera sites, there was also a reduction of over 100 fatalities per annum (32% fewer).
[The national safety camera programme Four-year evaluation, Executive Summary]

However, Appendix H contains an analysis of 216 camera sites that takes into account a statistical phenomenon (regression to mean effect) that can skew figures and found that the effect of speed cameras listed in the Executive Summary was overstated:

the average effect of these 216 cameras was a reduction of 19% in both PICs and FSCs relative to what would have been expected in the after period had the cameras not been installed.
[The national safety camera programme Four-year evaluation, page 157]

Even with the corrected statistic, the implication is that GATSO cameras reduce accidents by reducing speed - however many motorist groups stress that it is not speed that kills but dangerous driving. What speed cameras do is reinforce the ever decreasing application of discretion in modern policing in favour of no questions asked/no room for explanation strict liability.

Support for speed cameras

Last week the Royal Society for the Prevention of Accidents joined forces with the Parliamentary Advisory Council for Transport Safety (PACTS), the AA, the Association of Industrial Road Safety Officers (AIRSO), the UK’s National Cyclists’ Organisation CTC, the Institute of Road Safety Officers, Road Safety GB and others to voice concern about the switching-off of speed cameras. In a joint statement[7] the groups said:

Cameras should continue to be used where casualty statistics show they are needed.

Switching off cameras systematically would be close to creating a void in law enforcement on the road. Cameras currently account for 84 per cent of fixed penalty notices for speeding.

Cuts might also threaten many speed awareness courses that give motorists an opportunity to learn about the dangers of driving too fast.

While public spending needs to be cut, cuts must be justified by evidence. Cameras pay for themselves and currently make an important contribution to achieving compliance with the speed limit.

(See the Association of British Driver's recent press release for a rebuttal of the statements[8]).

So both within government and amongst some road safety groups there is strong support for speed cameras but also an acceptance that costs must be cut. How can both of these apparently conflicting desires be met whilst also meeting the Department of Transport suggestion of sharing back office functions across several police force areas?

Time over distance cameras

In 2009 a new generation of average speed, or time over distance cameras SPECS3[9] was given full Home Office Type Approval. SPECS are ANPR cameras, manufactured by the Speed Check Services Limited (from which they get their name SPECS - SPEed Check Services). SPECS3 is an upgrade to the existing SPECS1 cameras (given approval in 1999) that have been used to monitor speed along motorways in the UK, particularly when roadworks are taking place. SPECS3 cameras are cheaper to install because they can communicate wirelessly rather than requiring expensive fibre optic cables between cameras as required for SPECS1. Because SPECS cameras use ANPR they can be linked to a variety of databases such as the DVLA (Driver and Vehicle Licensing Agency) or the Police National Computer.

Because SPECS3 cameras are designed to work as "network nodes", where each camera is an entry and an exit camera, and able to calculate journeys between any active cameras they can be used to create speed check areas and could also be used to enforce urban 20mph zones[10]. One of the above mentioned groups in favour of cameras is the Parliamentary Advisory Council for Transport Safety (PACTS), who in 2007 produced a '20 MPH Default Speed Limit Briefing'[11] in which they suggested time over distance cameras as an alternative to traffic calming measures such as road humps, the briefing states:

Time over distance cameras offer an alternative enforcement tool. They operate by monitoring time of entry into and exit from a zone and comparing the expected travel time if all speed limits are adhered to. This allows longer stretches of road to have limits enforced than would be possible with a fixed site camera. They are also considered to be fairer, as they monitor speed over a longer distance and so avoid the complaints of ‘penalising drivers for momentary errors’, which are frequently made about fixed site cameras. Transport for London gave evidence to the Transport Select Committee’s road policing and technology inquiry that time over distance cameras would enable it to implement and enforce 20mph zones on 10,000km of residential road within 10 years – rather than the 35 years it would take to traffic calm them.

A February 2009 Times article 'Average speed cameras mean no escape for drivers'[12] expanded upon this idea of using SPEC3 cameras in residential areas, the article said:

Cameras that detect a motorist’s average speed will be deployed at all entry and exit points to residential areas as an alternative to road humps and chicanes. The Association of Chief Police Officers (ACPO) has approved a new generation of cameras that are linked wirelessly and operate in clusters, meaning that speeding drivers will be caught whichever route they take across a wide area.

What both the Times article and the PACTS Speed Limit Briefing fail to mention is that the roll-out of SPEC3 would be a massive expansion of the ANPR network in the UK as these cameras would be able to send data to the National ANPR Data Centre (NADC) in Hendon.

Project Columbus

In 2002 the Home Office launched a national ANPR programme "to deny serious and violent criminals the use of the roads". The press release accompanying the launch[13] explained the simple logic used to begin surveilling all car journeys: "Police experience has confirmed strong links between road traffic offences and other criminal conduct". This was later refined into a vision for exploiting ANPR[14]:

The Police Service, working with partner agencies, will fully exploit ANPR technology to: 'Deny Criminals the Use of the Roads'.

Of course to deny criminals use of the roads requires that they deny everyone unchecked use of the roads - as we have discussed elsewhere[15] the ANPR network is effectively a system of automated roadside checkpoints. A September 2009 British Association of Public Communication Officers (BAPCO) Journal article 'Deny the road to the criminal'[16] explains that since 2006 there has been a plan to expand the ANPR network:

To be effective, the national ANPR network has to be comprehensive and a lot of work has gone on over the last couple of years both to extend the primary network but also to work with partners using ANPR for other purposes. Since the launch of Project Columbus in 2006, a concerted effort has been made to extend the national ANPR network to cover private-sector sites including car parks, shopping centres and petrol stations.

The aim is to expand the system to 100 million 'reads' per day, all of which will be stored by the National ANPR Data Centre in Hendon. These reads will provide the time, data and place of each vehicle sighting and will be stored for five years providing a valuable source of intelligence for the future.

The article goes on to explain that many of the congestion monitoring ANPR systems installed by local councils around the country can also feed data back to the ANPR Data Centre:

An additional source of ANPR data has become available over the last couple of years as local authorities all over the country invest in journey-time measurement systems. They are doing this because the Traffic Management Act 2004, which requires local authorities to tackle congestion and disruption on their road network, is now beginning to bite. The starting point is to measure congestion, a somewhat controversial concept in its own right. The method of choice in many areas is to use ANPR on strategic routes.

Furthermore the article reveals that a new ANPR camera protocol known as Urban Traffic Management and Control (UTMC) means that: "Any local authority buying UTMC compatible cameras in future, should they be in partnership with the police, will be able to provide the data required by the National ANPR Data Centre".

Data validity

Meanwhile, the validity of the assertions that ANPR is an effective way of catching criminals has been called into question because of inaccuracies in the databases used to identify cars of interest to the police. The National Audit Office (NAO) found a third of DVLA's records could be wrong[17] and an evaluation of an ANPR pilot published in October 2004 ('Driving crime down - Denying criminals the use of the road')[18] revealed that the accuracy of DVLA data was just 40%, further noting that "Accuracy of DVLA databases declined over the study period" [page 98, Database Issues].

More recently an Independent on Sunday article in January 'The laughing policemen: 'Inaccurate' data boosts arrest rate'[19] reported that Police whistleblowers claim intelligence stored in the database behind the cameras, such as the Motor Insurance Database Application System (MIDAS), is inaccurate. The article states:

Critics claim there is "manipulation" of performance figures to make ANPR look more successful as a crime-fighting tool. The police accept it "may be out of date" and sources in Hertfordshire Police say it is "at best only 70 per cent accurate". One said: "The longest time a vehicle has taken to enter the database is 208 days from the date of the insurance commencement."

One whistleblower told the Independent: "In short, officers do not have a complete understanding of the law, use flawed databases to justify immediate seizures, fail to adequately research and evidence the basis of their belief and almost certainly knowingly seize vehicles just to satisfy service and personal performance targets".

Function creep

But whilst the data behind the systems may be full of errors there are grand visions of how this data may be interpreted that go beyond simply targeting cars flagged as vehicles of interest. Northgate Information Solutions Limited, a private company that provides ANPR systems to all police forces in England and Wales says that they "are taking ANPR beyond Roads Policing"[20]. Northgate also conduct ANPR data mining for Borders and Lothian police in Scotland[21] and their ANPR brochure 'Putting intelligence into action' [22] explains their vision of how all police forces should use the data gathered from cameras, it says:

Most police forces currently use ANPR to target specific vehicles on their hotlist. They are stopping vehicles they know they are looking for.

Our view is that where forces focus only on these hits, they are losing the potential intelligence value that exists in the 98% of reads that don’t match the hotlist.

With further analysis, this large volume of remaining data can reveal important information about the movement of vehicles. This can help identify offenders, witnesses and vehicle convoys that may be of critical importance to operational investigations.

The National CCTV Strategy

A July 2006 BT Redcare CCTV Industry News Digest[23] gave a few more details of Project Columbus, pointing out that it "is a newly revealed element of the joint ACPO/Home Office National ANPR Strategy". However recommendation 10.3 of the 2007 National CCTV Strategy[24], which relates to ANPR, gives no real indication of the scale of expansion envisioned in Project Columbus:

The ANPR partnership structure is a good example of a multi-agency approach dedicated to driving forward the ANPR strategy. The ANPR partnership mission statement declares that it will ‘work with Partner Agencies at National, Regional and Local level to share assets and data, avoid duplication and enhance operational effectiveness’.
[National CCTV Strategy, Recommendation 10.3, page 45]

ANPR and counter terrorism

Before ANPR was sold to the public as a tool for reducing road accidents, alleviating road congestion, detecting uninsured drivers or capturing criminals who drive, it was used in counter-terrorism operations in Northern Ireland. Whilst the counter-terrorism card has been little played in mainstream discussions of ANPR, it has recently reared its head after West Midlands police installed a network of 169 ANPR cameras to create a "ring of steel" encircling the Washwood Heath and Sparkbrook areas of Birmingham as part of "Project Champion, which was funded by the Association of Chief Police Officers Terrorism and Allied Matters Committee (ACPO – TAM). An active campaign 'Birmingham Spy Cameras - NO THANKS!'[25] has succeeded in getting the cameras switched off pending a review and will continue to campaign to have the cameras removed.

A July 2007 article for Australian technology website Smarthouse 'Find A Terrorist Speed Cameras Being Reviewed For Oz'[26] shows that outside the UK ANPR cameras are still seen very much as a counter terrorism tool:

The UK control centre which went live in April of 2006 is capable of processing 50 million number plates a day. ACPO national ANPR co-ordinator John Dean said that fixed ANPR cameras already exist "at strategic points" on every motorway in the UK, and that the intention was to have "good nationwide coverage." According to ACPO roads policing head Meredydd Hughes, ANPR systems are planned every 400 yards and are used to tackle more serious crime such as the arrest of the recent terrorist suspects.

It is though that the UK has the most surveillance cameras per head in the world and it has the dubious accolade of being the global leader in its use of ANPR technology. Surveillance measures introduced in the UK have implications for the rest of the world, as they look to us as a 'liberal democracy'. Each advance in surveillance in the UK has the potential to filter around the globe – the companies producing such technology are rubbing their hands with glee but ordinary citizens are not.

Regulation to tie up loose ends

Back to Chris Pounder's blog post about the swapping of speed cameras for ANPR - Pounder points out that the government's plan to introduce more regulation of CCTV is to also include guidance on the use of ANPR:

the Government has promised forthcoming legislation on CCTV and this legislation will also apply to ANPR. The legislation will determine, by law, the use and retention of CCTV/ANPR images.

With regard to CCTV cameras there is already regulation (albeit of scant use to the protection of the individual) in the form of the Data Protection Act and the Regulation of Investigatory Powers Act (RIPA), but ANPR has been an area of contention for some time. The Chief Surveillance Commissioner (who provides oversight of the conduct of covert surveillance under RIPA) has raised the issue in his Annual Reports going back to 2006. The Commissioner's 2005-2006 Annual Report [27] said:

In the normal case, where a camera is sited in an obvious position and obviously is a camera, the deployment will not be covert. The same will normally apply where adequate notice of the presence of a camera has been given. But ANPR is not a normal case, and it is arguable that, even if the presence of an ANPR camera is apparent, surveillance nevertheless remains covert if occupants of vehicles are unaware that the camera may make and record identifiable images of them. Explaining the true purpose of the equipment briefly is not easy. It is not possible to lay down rules as to what will amount to adequate notice of the presence of the camera and of its function.

Alas the commissioner, rather than recommend that such cameras should be removed because they are unlawful or an unacceptable assault on the rights and freedoms of citizens, instead suggests that the government simply makes them "legal" by creating a new law:

The unanimous view of the Commissioners is that the existing legislation is not apt to deal with the fundamental problems to which the deployment of ANPR cameras gives rise. This is probably because the current technology, or at least its very extensive use, had not been envisaged when the legislation was framed. The Commissioners are of the view that legislation is likely to be required to establish a satisfactory framework to allow for the latest technological advances. The position is complicated by the fact that the current technology can be used in a variety of different ways and at different levels of effectiveness. I am accordingly urging upon the Home Secretary the desirability of promoting such enabling legislation as may be needed.

The issue of ANPR was raised once again in the 2006-2007 and 2008-2009 Chief Surveillance Commissioner's Annual Reports but the previous government let the Association of Chief Police Officers expand the ANPR network unchecked, which at last estimate[28] stands at around 10,500 or so cameras. As that network prepares to expand still further to tackle speeding cars, criminality, terrorism, congestion, traffic calming and any number of other applications that can be dreamt up, the government want now to introduce guidelines to legitimate this automated checkpoint system. Unless vigorously challenged this will not be about taking down cameras - far from it. It will be about the abolition of motorised freedom of movement. Hollow phrases such as the ubiquitous and tedious "Nothing to hide, nothing to fear" show that many have forgotten what freedom is. ANPR is a serious threat to the freedom of everyone - it represents complete control of the road network and the ability of the state to track or prohibit the movements of whomever they see fit[29]. The police chose the name 'Project Champion' for their 'ring of steel' of ANPR cameras around residential neigbourhoods in Birmingham. As Steve Jolly of the campaign 'Birmingham Spy Cameras - NO THANKS!' points out, in marketing lingo a project champion is someone that leads the way and sets an example to be followed. For all our sakes, we must make sure that their champion fails and that we can begin the task of rolling back the surveillance nightmare.

Endnotes:

Posted in Anti-CCTV general - 2/9/2010

email: