The Research And Achievement Of Character Segmentation In Vehicle Plate Recognition System

License plate recognition(LPR)is the technique to draw the character information from the vehicle image and confirm the identity of the vehicle through using computer vision,image manipulation,pattern recognition,and so on.LPR is a important study field in the ITS,and it has great apply value.LPR is divided into plate position,character segmentation,and character recognition.The thesis studies character segmentation based on LPR and proposes a new method.The License plate segmentation system had been realized based on it.There are three key points in the license plate segmentation system.They are binarization,tilt correction and character segmentation.The thesis analysis and studies the related technique in the world,then proposed novel algorithm based on license plate image.A binarization method of vertical projection combined with image morphology is brought forward.It overcome variable Unfavorable factors,such as poor image quality etc,and get a good result in the end.Consider of both accuracy and efficiency,the thesis proposes a new algorithm Shell weighting vertical projection.Compare with the pure projection,it is more efficient.It also overcome the influence of license plate's frame and solves the problem beautifully.According to projection method that is most famous,the thesis also proposes a novel character segmentation algorithm.It had been approved to get the correct result on blur sample,distorted image and the image with license plate's frame.License plate segmentation system is self-adaptive.In order to improve the accuracy,it could give the solution based on wrong segmentation automatically. Experiments have proved that the accuracy of the system could reach 96.1%,the average time cost is 52ms.It could process the images that sampled from different environment,and get a wonderful result.