More and more often major retailers and private car park owners report of their car parks being abused by individuals who use their premises to park their cars over extended periods of time when they are, in fact, not visiting the companies that own the car parks. This is especially relevant when the car parks belong to airports, hospitals and private residential developments. It’s essential that car parks belonging to these establishments are managed effectively whilst remaining sufficiently versatile to accommodate the multiple needs of the car park user.
This type of car park abuse leads to a reduction in car parking spaces being available, and as a result leads to a loss in revenue for the retailers. For many years the only way of tracking cars using car parking facilities was to have a car parking attendant manually log the cars using the site. However, this was a costly and ineffective way of keeping track of vehicles using the car park illegally.
This is when the automatic number plate recognition or ANPR system in the UK came into being. Invented in the UK back in 1976, the system was not widely used until the 1990s. The technology involves automatically reading vehicle number plates through CCTV cameras at the entrance and exit of the car park. This information is then used to determine which vehicles have been on the site longer than they should be. Once a car which has overstayed its allotted time is identified a case report is compiled along with associated images, and a Parking Charge Notice is sent through the post to the owner of the vehicle.
The ANPR system in the UK uses six primary algorithms in order to identify a number plate.
- Plate localisation – enabling locating the plate and isolating it on the picture.
- Plate orientation and sizing – compensates for the angle of the plate and adjusts the dimensions to the size that’s needed.
- Normalisation – adjusts the image for brightness and contrast.
- Character segmentation – searches the plates for individual characters.
- Optical character recognition.
- Syntactical/Geometrical analysis – check characters and positions against the rules for different countries.
Difficulties
There are several potential difficulties that ANPR software must deal with in order for it to be effective.
Bad resolution, often resulting from the plate being at too far a distance, but sometimes due to a low quality camera.
Blurry images, especially from motion blur.
Bad lighting and low contrast caused by reflection, shadows or overexposure.
An object obscuring the plate, sometimes a tow bar, or some dirt on the plate itself.
Number plates which are different at the front and the back due to towed trailers or campers.
Whilst a few of these issues can be corrected by the software, it is normally a question of adjusting the hardware in order to work out the solutions. Increasing the height of the camera, for example, is often an effective way of dealing with obscuring objects.
Enterprise-level car park management system providers such as Century Fire and Security can design and install a software driven ANPR server suite with multiple cameras and integrate this with automated gates and barriers. Find out more about versatile ANPR solutions from Century Fire and Security.
