License Plate Recognition – A Tutorial
This web page provides an overview of LPR systems. The description starts ‘technical’ then proceeds to the ‘market’ view.
(1) What is LPR?
LPR (License Plate Recognition) is an image-processing technology used to identify vehicles by their license plates. This technology is used in various security and traffic applications, such as the access-control system featured in the following animation:
(see tutorial In the above example: while the vehicle approaches the gate, the LPR unit automatically “reads” the license plate registration number, compares to a predefined list and opens the gate if there is a match.
(2) Technology Highlights:
This technology is gaining popularity in security and traffic installations. The technology concept assumes that all vehicles already have the identity displayed (the plate!) so no additional transmitter or responder is required to be installed on the car.The system uses illumination (such as Infra-red) and a camera to take the image of the front or rear of the vehicle, then an image-processing software analyzes the images and extracts the plate information. This data is used for enforcement, data collection, and (as in the access-control system featured above) can be used to open a gate if the car is authorized or keep a time record on the entry or exit for automatic payment calculations.
The LPR system significant advantage is that the system can keep an image record of the vehicle which is useful in order to fight crime and fraud (“an image is worth a thousand words”). An additional camera can focus on the driver face and save the image for security reasons. Additionally, this technology does not need any installation per car (such as in all the other technologies that require a transmitter added on each car or carried by the driver).
(3) Other Names
LPR is also called in different references as:
- Automatic Vehicle Identification (AVI)
- Car Plate Recognition (CPR)
- Automatic Number Plate Recognition (ANPR)
- Car Plate Reader (CPR)
- Optical Character Recognition (OCR) for Cars
(4) Plate and images
LPR units are based on images of the front and/or rear plates.The following example shows a typical installation of an access-control system in the gate of a secured office compound.
The car, with the plate number: W856RKX, is just exiting from the secured area and is picked up by the LPR system which identifies it as an authorized car and opens the gate. A welcome display (which includes the vehicle number and driver name) is displayed in the large outdoor display in the back of the car.
A typical image captured by the Infra-red illumination and camera unit is shown in the following image (this image is of the same car above). The image information is read by the LPR unit and is analyzed and recognized automatically. Note that the image is monochrome since the infra-red spectrum is above the normal color spectrum.
Since the vehicle plates are based on different Country standards, they usually different in form, shape and material. Therefore the LPR systems are Country specific and are adapted to the Country where they are installed and used (in the above case – a UK installation). You can see a large collection of plates of different Countries – download some recognition demoes with the link at the bottom of this page.
Some plates have additional information written or attached to the plate, such as state identification (as in USA plates in upper smaller row), province initials (such as in China upper row), plate issue date lable (such as in USA plates), vehicle type (such as in Korea in an upper row), Country flag (as in Europe on the left side), and more. For most applications this additional information is not required (and is usually not identified due to the complexity, variation, quality and resolution).
(5) What’s in an image
The above image of the front side of the car, shown in a typical format, is composed of 256 grey levels ranging from black (grey level 0) to white (grey level 255). For a typical format there are 768 X 288 pixels (PICture ELements), or about 0.2 Million elements. This vast amount of information is processed by the recognition software in order to automatically locate and read the plate.
The following image shows a zoom into the above image sample. The initial image starts from the raw data, then repeatedly zooms up (with factor x2) until the pixel level.
The computer processing needs to work on the global information (the entire image) for detecting the plate, then zoom into the data in order to handle the small details and finally extract the registration data. The end result of the recognition process is a string (in the above case, “W859RKX”). This transformation of the image data into a result string is actually a very large compression of the original raw data (1:31600 !).
(6) Does it Work?
Early LPR systems sufferred from a low recognition rate, lower than required by practical systems. The external effects (sun and headlights, bad plates, wide number of plates types) and the limited level of the recognition software and vision hardware yielded low quality systems.However, recent improvements in the software and hardware have made the LPR systems much more reliable and wide spread. You can now find these systems in numerous installations and the number of systems are growing exponentially, efficiently automating more and more tasks in different market segments. In many cases the LPR unit is added as retrofit in addition to existing solutions, such as a magnetic card reader or ticket dispenser/reader, in order to add more functionality to the existing facility.
Even if the recognition is not absolute, the application that depends on the recognition results can compensate the errors and produce a virtually flawless system. For example, when comparing the recognition of the entry time of a car to the exit time in order to establish the parking time, the match (of entry verses exit) can allow some small degree of error without making a mistake. This intelligent integration can overcome some of the LPR flaws and yield dependable and fully automatic systems.
(7) Elements of typical LPR systems
LPR systems normally consist of the following units:
- Camera(s) – that take the images of the car (front or rear side)
- Illumination – a controlled light that can bright up the plate, and allow day and night operation. In most cases the illumination is Infra-Red (IR) which is invisible to the driver.
- Frame grabber – an interface board between the camera and the PC, allows the software to read the image information
- Computer – normally a PC running Windows or Linux. It runs the LPR application which controls the system, reads the images, analyzes and identifies the plate, and interfaces with other applications and systems.
- Software – the application and the recognition package. Usually the recognition package is supplied as a DLL (Dynamic Link Library).
- Hardware – various input/output boards used to interface the external world (such as control boards and networking boards)
- Database – the events are recorded on a local database or transmitted over the network. The data includes the recognition results and (optionally) the vehicle or drver-face image file
The following illustration shows a typical configuration of a LPR system (for example, for 2-lanes-in and 2-lanes-out access control system). The system (“SeeLane”) is a typical example of such system.
The SeeLane application runs as a background Windows application in the PC (shown in the center), and interfaces to a set of SeeCarHead camera/illumination units (one for each vehicle) which are interfaced by the frame grabber. The application controls the sensors and controls via an I/O card that is connected thru a terminal block to the inputs and outputs.
The application displays the results and can also send them via serial communication and via DDE messages to other application(s). It writes the information to local database or to optional remote databases (via the network).
(see tutorial http://www.licenseplaterecognition.com/