Licence Plate Recognition Research Based On Image Sequences

The License Plate Recognition System(LPR) is an important part in the modern Intelligence Traffic Management,which is produced accord -ing to the requirement of the modern traffic development.The purpose of this topic is referencing the person distinguishes's characteristic to addvance the accuracy and adaptability of recognition by recognise and fuse the result of the image sequences.This paper include three contents which are locating of license plate, character segmentation and character recognition.1.License Plate Location is an important part in the LPR. The accurate of license plate location affects directly the following parts of the recognition.This thesis improves the traditional algrithm based on the vehicle edge examination. The research fuses the edge inforamtion and color information of the vehcile license plate.The results of the experiment show that the new algrithm can improve the accurate rate.At the same time, the connection information of the serial vehicle images has been used to estimate the following plate position,which reduces the time of location rapidly.2.Character segmentation affects the preciseness of the character recgnition directly after locating license plate. The thesis adopts the projection method,scanning the plate which has been extracted,finding the lower part of the wave to pick out the row character,single characeter and adjust the single character position. At the same time, make use of the connection of the serial vehicle images are used to instruct character segmentation of the next frame vehicle image.3.Character recognition is the kernel part of the whole system, which includes the recognition of the chinses characters,letters and digits. This thesis makes use of the connection of the serial vehicle images to fuse the results of the vehicle plate numbers,which can improve the accurate rate of the character recognition.The vehicle LPR system is designed and implemented based on the research work mentioned above. The experiments' results show that some improvement and innovation has been come true compared with the traditonal algorithm.On the other hand ,the purpose of the paper has been reached.