There currently are about 330,000 police response vehicles in the United States. A great number of these vehicles are equipped with the latest in crime-fighting equipment – i.e., systems ranging from LED light bars to mobile data terminals and/or laptop computers. Most of these systems are much more than helpful; they are essential. But they also are costly. Every department nationwide, in fact, can expect to spend ten to fifteen thousand dollars (or more) – in addition to the purchase price and annual maintenance and upkeep costs – for the various crime-fighting systems installed on each of its vehicles.
Cost is one reason, but not the only one, why departments must ensure they are effectively using the equipment in the vehicle to get maximum benefit from the taxpayer dollars invested. In several specialized areas of law enforcement, particularly in the forensic sciences, technology and innovation have been at the forefront in identifying criminals and linking suspects to crimes. State-of-the-art communications and record-management systems assist patrol officers with basic communications and record keeping, obviously. Occasionally, though, creative thinking “outside the box” has led to important operational breakthroughs in crime fighting and elsewhere in the law-enforcement field.
Today, so-called ALPR (automated license plate recognition) systems give police officers the ability to capture, optically scan, and recognize vehicle registration plates through the use of an optical character recognition (OCR) system, and then to compare the information recorded to the information in a known database. However, relatively few law-enforcement agencies have attempted to use ALPR technology – for one simple reason: its extremely high cost. The cost of most current ALPR systems ranges from twenty thousand dollars to sixty thousand dollars, and many of the systems available are somewhat limited in their capabilities. Usually, the target vehicle must be at a specific location; the system’s camera must be mounted on a stable, secure base; and the vehicle must not be in motion. Because of these limitations, ALPR systems have been used primarily in gate-controlled parking lots and at tollbooths.
More Capability at Lower Cost
OCR is not new – several corporations and a number of government agencies have been using OCR technology for the last thirty years to scan and record mail, tax records, insurance information, and medical records. OCR technology is now used for a number of other purposes as well, and – of equal or greater importance – its cost is going down. Thanks to miniaturization and other improvements, systems fitted with customized hardware and mainframe computers that once sold for hundreds of thousands of dollars are now available at a much lower cost. Today, for example, Windows-based personal computers using off-the shelf cameras are able to read – at a rate of 10 envelopes per second or higher – large numbers of marked-up envelopes, find and recognize the addresses on each, and sort them to the appropriate Zip-code delivery area. Even consumer OCR systems have filtered down to the home personal computer, for under one hundred dollars, as the key component of an all-in-one printer/copier/fax machine.
In addition, a number of breakthrough law-enforcement applications are starting to materialize. The ability to quickly identify stolen vehicles, or vehicles registered to wanted persons (including suspected terrorists) – or registered to missing or endangered persons – could quickly assist in the apprehension of thieves or terrorists, the identification of Alzheimer victims, and help in various other caretaker functions police officers are often called on to perform.
There are two problems, though. The first is that current technology relies on the officer to call in, or type in, the numbers and letters on a registration plate to see if it falls into one of the categories mentioned. The second is that, for the police officer to call it in, his or her attention must be first be drawn to the vehicle for any of several reasons – e.g., suspicious activity, broken locks or windows, a hanging registration plate, or the car being parked in an inappropriate location. But if the vehicle seems to be in good condition and is operating within the law, there is little reason for an inquiry.
ALPR systems, however, do not have to sort and discriminate and/or make judgment calls. They see every plate that passes within a specific point of the camera, and compare all registration plates (which, in legal terms, are public documents) against a database of stolen vehicles and/or vehicles wanted for other reasons.
There are now about 75 or more companies worldwide that advertise ALPR systems. Some of these companies are private branding – i.e., distributing similar systems under several different names – but most are encumbered by similar limitations when it comes to using them for law-enforcement applications: a relatively high cost, a requirement for stationary mounting, and the use of customized hardware.
A Prototype of Other Breakthroughs to Come
A few companies have gone forward from the original technology and have built systems with the ability to identify moving vehicles. However, the twenty thousand dollar plus price tag is still too costly for most departments, and cramming another computer into an already crowded patrol vehicle is sometimes just not practical.
Enter into the law-enforcement ALPR race one Richard Coburn, the retired founder of one of the leading companies in the world in high-speed optical recognition, Scan Optics Inc. – which already supplies large-scale OCR systems to the postal system and several large insurance companies. Coburn asked a simple question: What does the mobile data terminal or laptop now carried in most police cars do when the officer is patrolling? There are three somewhat similar answers to that question: (1) Nothing; (2) Wait for an incoming message; or (3) Wait for the officer to stop patrolling and start typing.
Coburn’s idea was simple: use the processing power of the computer, which is already in the police vehicle – and stays there, even when the officer is not – to carry out the ALPR function automatically. The only other equipment needed would be a screen-saver of some type to run in the background and an off-the shelf-camera (or one already installed in the vehicle to record traffic stops and/or for other law-enforcement purposes).
Drawing on 30 plus years of experience, Coburn worked with Ronald Gocht, a former development engineer with Scan Optics but now retired, to create a software package that could operate in the Windows operating system to provide exactly what police departments need: a relatively low-cost (less than two thousand dollars) ALPR system that can scan a lane of moving traffic, identify all of the registration plates in its field of vision, and compare them to a known database on the police vehicle’s existing mobile data terminal/laptop and camera.
Although his SentryTec ALPR is only in the prototype stage, Coburn already can demonstrate its use in real time, on real cars. According to Coburn, the SentryTec ALPR still needs final software engineering to make it an operational police system. Future versions of the system, he said, will be available on Windows-based pocket PCs for use by officers on foot patrol, on bikes, or mounted on horses.
Similar creative ideas – from inside the law-enforcement community, from municipal administrators, and from the commercial world – using new technology or existing technology converted from other industries, will continue to change the way police work is done for many years to come. For the present, though, agency administrators must carefully weigh the costs of purchasing and operating ALPR systems and compare those costs with the benefits provided to their communities. Interestingly, to make an equipment purchase more appealing, several manufacturers have been able to demonstrate a positive cash flow, within just a few short months, from the use of such systems simply by adding the vehicle registration plates of delinquent taxpayers to the ALPR database.