| 摘要: |
| 针对卷积神经网络（CNN）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CNN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，防止梯度弥散. |
| 关键词: 卷积神经网络(CNN) 交通标志识别 深度学习 |
| DOI：10.3969/J.ISSN.1000-5137.2018.05.014 |
| 分类号:TP391.41 |
| 基金项目: |
|
| Traffic sign recognition algorithm based on improved convolution neural network |
|
ZHU Yongjia, ZHANG Jing
|
|
College of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 200234, China
|
| Abstract: |
| For the problem of low traffic sign recognition rate due to gradient diffusion in convolution neural network(CNN), an improved convolution neural network was proposed. The experiment results showed that the improved method could increase the recognition accuracy effectively and prevent gradient vanishing. |
| Key words: convolution neural network (CNN) traffic sign recognition deep learning |
Editorial Board