摘要
提出了一种应用BP神经网络识别汽车牌照字符的方法,重点讨论了关于BP神经网络学习过程初始权值的选取、隐含层节点数的确定和权值学习算法的改进问题,实验结果表明:该方法用于车牌识别具有较快的收敛速度和较高的识别精度。
A BP algorithm is proposed to automatically recognize the characters on a car license plate. The discussion is emphasized on the selection of original weight, the decision of the number of hide nodes and the learning algorithm of weight. Experiments indicate that the method can greatly improve the learning speed and the precision of recognition.
出处
《电子产品可靠性与环境试验》
2008年第5期71-74,共4页
Electronic Product Reliability and Environmental Testing
基金
太原市科技兴市专项基金(081222071)项目资助
