论文部分内容阅读
限速标志识别系统是智能交通系统的一个重要组成部分,它能有效地辅助司机安全驾驶.针对限速标志的数字字符识别问题,提出一种基于超网络模型的模式识别方法.首先介绍了超网络计算模型及其分类原理;然后采用颜色分割和形状分析相结合的方法对限速标志进行定位,并提取出限速数字字符特征;最后以限速字符的特征向量为训练样本对超网络模型进行演化学习.本文使用超网络模型对限速标志20、40、60、80 km/h进行识别.实验结果表明,基于超网络模型的道路限速标志识别系统最快只需3次迭代便可以完成对样本的学习,识别率为96.15%.和其它传统模式识别方法相比,该模型具有学习时间短、识别率高的优点,为解决现实应用中的道路限速标志识别问题提供了可能.
Speed limit sign recognition system is an important part of ITS, which can effectively assist the drivers to drive safely.Aiming at the problem of digital character recognition of speed limit sign, this paper proposes a pattern recognition method based on super-network model.First, Network computing model and its classification principle, and then use the method of color segmentation and shape analysis to locate the speed limit sign and extract the character of speed-limited digital character. Finally, using the character vector of speed-limited character as the training sample, The author uses the super-network model to identify the speed limit signs of 20, 40, 60 and 80 km / h.The experimental results show that the speed limit identification system based on the super-network model can only need three iterations at the earliest Compared with other traditional pattern recognition methods, this model has the advantages of short learning time and high recognition rate, which makes it possible to solve the problem of road speed restriction sign recognition in practical applications.