Adding License Plate Recognition to your Physical Security Strategy

Managing and monitoring access to a facility is an ongoing challenge. Beyond access cards and visitor management, security teams are using license plate recognition (or automatic number-plate recognition) as a tool to manage and reduce risk. From controlling parking ramp access to managing parking passes, corporate, government and educational institutions are embracing license plate recognition.

We asked Mark Bonde, Parallel Technologies’ expert on physical security to provide an overview of what is license plate recognition and where/how the technology can be applied to support an organization’s physical security strategy.

What is License Plate Recognition?

License Plate Recognition (LPR) is a technology that uses optical character recognition on images to read vehicle registration plates to create vehicle location data. The optical camera recognition technology decodes the license plate numbers and letters into machine-readable characters. Once the image of the vehicle’s license plate is captured, the system either stores it for later review or runs the image through software to compare it to a database of stored license plate numbers. If the database finds a match, the system can be set up to send alerts or trigger other actions such as opening a gate.

What are the most common applications for LPR?

According to Bonde, “LPR provides organizations with greater visibility into the vehicles that are entering a facility as well as a method for alerting people of potential situations.” LPR is being implemented to address a wide range of situations. Situations that are well suited for LPR capture include those where you need to:

• Secure an area, such as a military base or research facility
• Monitor gated entrances & parking lot traffic
• Track a known suspect’s vehicle
• Find a stolen vehicle or car involved in a crime
• Enforce parking laws, including time limits or reserved spaces

The most common application we are being asked about is parking lot management. In these situations, the system is used to manage contract parkers and determine the number of vehicles in the lot/open spaces. In this situation, a camera is placed at the entrance gate, and it automatically records an approaching car’s license plate to compare it to a database of approved visitors. If that license plate is on the database, the gate will open and allow the car to enter.

The system is also being integrated into corporate systems to alert management when VIP clients arrive at the facility. From a security standpoint, LPR can also be configured to provide staff with an alert when terminated employees or individuals with restraining orders enter the facility grounds in their vehicles.

What are the camera requirements for LPR?

For license plate recognition to work, the camera component of the system needs a few specific things need to be carefully considered:

•  The angle of the camera is extremely important, because a camera that’s installed too high cannot see the license plate
•  The distance from the camera to the car must also be considered – while a camera may be able to zoom in quite a distance, you want to minimize the distance between camera and car
• Lighting is also important – whether you install lights or use the camera’s built-in IR, you will need additional lighting at night
• The speed of oncoming traffic must be low enough that the camera has enough time to focus on the license plate, often times no more than 35 mph depending on the angle, distance, and lighting.

How is the technology evolving?

The growth of LPR systems is driven by the advances in video camera technology but also the the strides made in deep learning analytics.

According to Bonde, “the application of deep learning analytics is rapidly increasing the technology itself and ultimately is making the camera analytics more reliable. Major improvements in just the past 18 months have made this technology more powerful than ever.”

If you are interested in learning more about LPR, please contact Mark Bonde at mbonde@ptnet.com.