快速检索:      
引用本文:
【打印本页】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
过刊浏览    高级检索
本文已被:浏览 154次   下载   
分享到: 微信 更多
基于卷积神经网络的交通标志识别方法
朱永佳,张静
作者单位E-mail
朱永佳 上海师范大学 shanghai_yongjia@163.com 
张静 上海师范大学 jannety@shnu.edu.cn 
摘要:
随着人们对智能交通系统(Intelligent Transport Systems)的需求越来越大,交通标志识别的研究变得越来越重要.卷积神经网络(CNN)是深度学习的一个重要网络模型,由于其自动提取图片的特征使得在图像识别中受到广泛应用.针对卷积神经网络在交通标志识别过程中出现的梯度弥散而引起的识别率低的问题,给出了基于改进卷积神经网络结构的交通标志识别方法,实验结果表明:该方法能够有效提高识别精度和防止梯度弥散.
关键词:  卷积神经网络  交通标志识别  深度学习
DOI:
分类号:
基金项目:
Traffic Sign Recognition Algorithm Based on Convolution Neural Network
zhuyongjia,zhangjing
Abstract:
With increasing needs of people on the Intelligent Transport Systems,traffic sign recognition becomes more and more important.Convolution neural network (CNN) is an important network model of deep learning. Because of its automatic extraction of image features, it is widely used in image recognition.For the problem of gradient diffusion in Convolution Neural Network,the traffic sign recognition method based on convolution neural network is given and experiment results show that this method can effectively improve the recognition accuracy and prevent gradient dispersion.
Key words:  Convolution neural network  traffic sign recognition  deep learning