Automatic License Plate Recognition (ALPR)

All around the world the traffic police would stop drivers on the road to check if their road tax has been renewed. Unfortunately this sometimes causes delays for the people stopped and may even slow down the general traffic. Furthermore, for the police involved a big portion of their time is wasted just doing these checks. In this work we propose a new strategy for automating the monitoring of road-tax for drivers on the road and propose to notify the ones who need to renew their road tax by a formal e-mail message. The idea is to combine the license plate detection and recognition process with the driver’s information stored in a database and an e-mail sending function. The first task is the acquisition of frontal images for approaching cars. Then the application will estimate the rotation angle on each frame and make it horizontal by reverse transforming it. Subsequently, the position of the license plate in the frame will be detected and it will be segmented out. Once the plate region is cropped out the next step will be to segment out the individual alpha numeric characters. This task is achieved by looking at a vertical projection profile. Later the characters are traced to obtain their associated stroke based (8-connectivity) chain codes. Finally features are extracted from characters and chain codes and character recognition is carried out based on the collected features. Our hybrid license plate detection and recognition system was tested using 150 test images. With two-letter and three-digit notation of North Cyprus license plates this would mean around 750 characters. Simulation results have showed that our proposed system was 93.33% successful in detecting and segmenting out the license plates. For detected plates the character segmentation accuracy was 100% and the recognition accuracy was 96.42%. Finally our ALPR system was found 92.85% successful in recognizing vehicle license plates. Note that only the cases where all characters were correctly detected have been considered as correctly recognized plate. After obtaining the plate number a database is searched to locate the owner of the vehicle. The name obtained will then be used to check the road tax statue of the driver and if necessary an automatic e-mail message will be sent out using the e-mail address of the driver. The message will be informing the driver about what he/she should do and also at the same time warn him/her about what would happen if he does not attend in a given time period. Keywords: License Plate Detection, Character Segmentation, Vertical Projection Profile, Stroke Analysis, Chain Code, Feature Extraction