利用神经网络识别车辆牌照字符

Vehicle Plate Recognition Using Neural Network

由于受光照条件、牌照自身清洁程度等因素影响 ,车牌识别系统往往达不到满意的字符识别率 .结合神经网自适应的特点 ,本文利用基于自适应谐振理论 (AdaptiveResonanceTheory ,ART)构成的自组织神经网络进行字符识别 。

A satisfactory recognition rate can not be obtained sometimes because of some factors affected by light conditions and plate's cleanness in vehicle recognition systems.In this paper,an approach is presented to recognize vehicle characters based on adaptive neural network constructed by ART,combined with the neural network's adaptive feature,which effectively improves recognition rate.