Harnessing ANPR technology in policing

Sharon Beshenivsky had always wanted to be a police officer. After being a community police officer, in May 2005, she became a British police constable in Bradford South, New Yorkshire. On 18 November 2005, which was her daughter's fourth birthday, she definitely would not have known she would wear the uniform for the final time. While on a patrol, she and her colleague thought they were attending a routine call. When they arrived at the scene, they faced three men. One of them was armed with a gun, the gunman fired at them, police constable Sharon got fatally shot in the chest. The miscreants immediately fled from the scene. The CCTV at the scene which links to an Automatic Number Plate Recognition (ANPR) system helped trace the car, resulting in arrest and subsequent conviction of six suspects. This is the first documented case of an ANPR being used to solve a murder case. Back in 1981, ANPR had helped the police make the first arrest through detection of a car.

Automatic number-plate recognition (ANPR) is a technology that helps detect, deter and disrupt criminality, including organised crime groups and terrorists. ANPR makes use of optical character recognition on images to read vehicle registration plates and generate vehicle location data. ANPR can use existing closed-circuit television, road-rule enforcement cameras, or cameras specially devised for the task. ANPR is being used by police forces around the world for law enforcement purposes and to check the licence and registration status of the vehicles. Highway agencies use it for electronic toll collection on pay-per-use roads. Companies configure some ANPRs to even store a photograph of the driver. ANPR systems commonly use infrared lighting to allow the camera to take pictures during both day and night. The invention of ANPR had come about through the Police Scientific Development Branch in the UK in 1976.

Typically, the system works in the following way: the ANPR system with its software 'reads' vehicle registration plate from digital images, captured through cameras located either in a mobile unit, alongside the road or in-built in traffic vehicles or via CCTV and then the optical character recognition (OCR) system kicks in to extract the alphanumerics of the license plate. There are two types of ANPR systems, the first kind transmits all the images captured on the roads to a remote computer location which performs the OCR at some later point of time while the second system performs the total process of capturing and extracting the alphanumerics at the lane location in real time. In the latter, the ANPR captures and extracts the alphanumeric, date-time, lane identification, and any other information required in approximately 250 milliseconds. It transmits this information to a remote computer for further processing or to store it at the lane for later retrieval. ANPR systems come with the capability to cross-reference the data against a variety of databases. Once the data gets cross-checked against the various databases such as the road or motor vehicle data base or police database such as CCTNS or intelligence database a process that takes around 1.5 seconds to complete, it is able to generate information about the vehicle, its registered owner and driver. When the information supplied via the ANPR system alerts officers to an offence or relevant intelligence on a vehicle, the officers stop the vehicle to investigate further. Some ANPR systems can check up to 3,000 number plates per hour per lane, even at speeds of up to 150 kms per hour.

During the 1990s, significant advances in technology took automatic number-plate recognition (ANPR) systems from limited, expensive, hard to set up, fixed based applications to simple "point and shoot" mobile ones. This they could achieve by creating software that ran on cheaper PC based, non-specialist hardware. Further scaled-down components at more cost-effective price points led to a record number of deployments by law enforcement agencies around the world. Smaller cameras with the ability to read license plates at higher speeds, along with smaller, more durable processors that fit in the trunks of police vehicles, allowed law enforcement officers to patrol daily with the benefit of license plate reading in real time.

We need different ANPR cameras in different situations. Highway patrol would require forward-looking cameras that span multiple lanes with capabilities to read license plates at very high speeds. City patrols would need shorter range, lower focal length cameras for capturing plates on parked cars. Parking lots with perpendicularly parked cars would require a specialized camera with a short focal length.

We can also use ANPR to enforce speed limits. ANPR is being used for speed limit enforcement in Australia, Austria, Belgium, UAE, France, Italy, India, Spain, South Africa, the UK, and Kuwait. The ability of ANPR to record the time and location of a vehicle means that if we locate an ANPR-enabled camera at another location along a stretch of the highway, it is possible to estimate the time taken for the vehicle to have travelled between the two points. In case the ANPR finds an over speeding vehicle an auto-generated challan gets despatched to the offending driver by accessing the vehicle registration database.

 ANPR system can help in taking to task errant motorists who jump signals, do not wear seat belts, overspeeding, violate helmet rules and other road safety rules. When I was holding the charge of traffic and road safety in Tamil Nadu, I had the opportunity of studying the traffic management system in Kerala where the ANPR cameras watch for over speeding and other violations in the state highways . I also visited KELTRON Unit in Trivandrum, which manufactured the Traffic Enforcement system installed in Kerala including Speed Limit Violation Detection System (SLVD), Red Light Violation Detection System (RLVD) and All Vehicle Automatic Number Plate Recognition (ANPR) System, and related hardware. The Tamil Nadu Government has entrusted KELTRON with installation of a similar speed enforcement system in Tamil Nadu highways in three stretches. Cities such as Delhi, Bengaluru, Mumbai, Hyderabad etc. have already implemented intelligent traffic management systems.

In the UK, a network of nearly 8,000 cameras is capturing between 25 and 30 million ANPR records daily. They stockpile these records for up to two years in the National ANPR Data Center, which UK law enforcement authorities can gain access to, analyse and use as evidence as part of investigations. In 2012, the UK Parliament promulgated the Protection of Freedoms Act. Under this Act, the Home Office published a code of practice in 2013 for the use of surveillance cameras, including ANPR, by government and the police. The aim of the code is to help ensure policing by consent.

Roughly 71% of all US police departments use an ANPR. Mobile ANPR is a significant component of predictive policing strategies and intelligence gathering, and for recovery of stolen vehicles, identification of wanted felons. Authorities in the US allow ANPR systems to collect and store the captured licence plate data indefinitely and use images, dates, times and GPS coordinates to place a suspect at a scene. They can match successfully recognised plates against databases including "wanted person,", missing person, a gang, suspected terrorist, and sex offender lists. In 1998, a Washington, D.C. police officer was found resorting to extortion by blackmailing the owners of vehicles parked near a gay bar. In 2015, the Los Angeles Police Department hit upon the idea of sending letters to the residential addresses of all cars that enter areas of high prostitution.

Australia has a mobile ANPR system called MANPR with three infrared cameras fitted to its Highway Patrol fleet capable of identifying unregistered and stolen vehicles, disqualified or suspended drivers and persons with outstanding warrants. On 11 March 2008, the Federal Constitutional Court of Germany decreed that some parts of the laws permitting the use of ANPR in Germany violated the right to privacy. Vehicle
registration plates in Saudi Arabia use a white background, but several vehicle types may have a different background. France has 250 fixed cameras that are being used to levy an eco-tax on lorries with over 3.5 tonnage. Capital Ankara in Turkey has a system with two cameras per lane, one for plate recognition and another for speed detection.

Efforts of Department of Homeland Security in the USA to curb terrorism has resulted in adoption of ANPR for susceptible facilities such as airports, sea-ports, embassies, military, government buildings,  schools etc. ANPR is being used for perimeter security and access control in major US agencies. IP based surveillance cameras besides ANPR also have facial recognition, object tracking and recording systems for the purpose of watching suspicious behaviour and identification of suspected criminals and terrorists. Massive ANPR networks are being installed by cities such as Boston, London and New York City to protect the cities against acts of terrorism, and to protect places of public gatherings.
Critics are describing ANPR as mass surveillance by raising concerns about privacy, high error rates, misidentification and government watching movements of citizens. The extent to which ANPR could become an intelligence or surveillance tool would depend on the databases used with the designated cameras. If the authorities link ANPR cameras only to traffic enforcement databases, there need not be any concern, but mass surveillance comes to the fore when they link it to intelligence or other databases such as Aadhaar.

ANPR may help police solve crimes and enforce traffic effectively but only soul recognition technology can help police personnel who see the worst of society and get caught up in the everyday struggle of catching bad guys realise they are more than physical beings. The technology of inner work can help them recognise that they are not mere physical beings but spiritual beings having human existence. The physical world cannot end the despair and hopelessness of police personnel; only the spiritual world can by giving them a purpose and a reason for being.

(Dr Jayanth K. Murali is ADGP (L&0), Tamil Nadu.Feedback can be sent to www.jayanthmurali.com)