2) Properly Position the Camera
Ideally position the cameras at vehicle "bottlenecks" or by control points such as barriers where vehicles have less opportunity to travel at excessively high speed or at angles to the camera. Consider traffic management issues to ensure clear line of sight from camera to license plate at all times. The camera should not be positioned too high above the road and the shallower the angle, the better. Ideally the license plate should be viewed from not more than 35 degrees. Consider the position of the sun in relation to your camera during different parts of the day. Be aware of the sun's position during different parts of the year - summer vs. winter.
3) One camera, One Lane
Some installations have failed because a single camera has been set-up to cover more than one lane (eg. in and out from a car park). Although the camera may be dedicated to capture license plates, by attempting to cover too wide an area image size and quality will not be good enough for consistent reliable capture on a 24 hour basis, particularly if the system is to be used with intelligent software.
4) The Importance of Infrared
Infrared illumination plays a vital part in all license plate capture systems, not just at night, but during the day as well. With license plate capture systems, visible light is cut-filtered and only IR light is used by the camera. This solves the serious problems caused by glare on the license plates from factors such as headlights, sunlight and reflections from wet roads. Therefore, good quality IR lighting is essential to provide good quality license plate images -- day and night.
5) Recording and Compression
Avoid compressing images too much on your DVR. While compression is fine for many applications, it cuts the file size of each image to save on memory. As a result, compression reduces the resolution of the image and can render the captured license plate unusable. The degree of compression acceptable will depend on factors such as the intended use of the license plate image and the speed of traffic. Simple review of pictures by the human eye will allow a certain reduction in quality. The simple test is to play back the recorded images to determine if they are fit for the purpose.
6. Capture vs. Recognition
Although these two terms are often confused and sometimes interchanged, they are altogether different. Capture refers to the process by which the CCTV camera obtains the image and is primarily concerned with delivering a high quality image of the license plate. It does not refer to recognition or interpretation of the image in an intelligent manner. Recognition - as in Automatic License Plate Recognition (ALPR) systems - refers to the intelligent software process of reading, recognizing and analyzing the captured CCTV image and using optical character recognition techniques to create an alphanumeric data string of the characters on the license plate. The data is saved digitally and may be delivered to a remote location using the internet, wireless broadband or other transmission system. The digital record can then be searched for, reviewed, compared to existing databases and used to trigger commands such as opening barriers. Although ALPR is still very new technology and generally reserved for high-end applications, the implications of ALPR are game-changing in nature.
Getting it right
In conclusion, no matter what your requirement, it is essential to have high quality images of the license plate on a 24/7 basis. This requires a combination of good installation and high-quality equipment.
The essential first step for the installer is to clearly define with the customer the purpose of capturing the license plate and the level of intelligence required. As a general rule the higher the intelligence the higher the final cost. It is always easy to over-specify, but considering your aim first will help reduce unnecessary costs as the project progresses.