Research On License Plate Recognition System

The number of vehicles in China today is increasing day by day, and how to give an effective management for vehicles has become the main problem for traffic management agencies. Vehicle license plate which is a unique tag for each vehicle plays a huge role in the management of the vehicle. Vehicle license plate recognition system mainly relies on image processing technology, plays an important role on various roads in the city, community, and other public places and other fields.This paper introduced a prototype system method of license plate recognition based on the research of license plate location, license plate character segmentation and license plate character recognition technology. In the plate preprocessing module, it has been done that getting rid of noise of non-plate area by the image gray processing and a series of operations. In the plate location module, the paper puts forward an algorithm of plate edge recognition by wavelet decomposition and mathematical morphology. After the binaryzation and tilt correction, license plate characters were segmented by vertical projection. Finally, license plate character was recognized by the combining method of more templates matching and secondary breakdown, and the matching degree of characters which are waited recognition and standard module is evaluated by improved Hausdorff distance.After experimental verification, the license plate recognition system based on this algorithm can achieve a higher license plate recognition rate.