计算机应用 ›› 2020, Vol. 40 ›› Issue (4): 1045-1049.DOI: 10.11772/j.issn.1001-9081.2019091610

• 人工智能 • 上一篇    下一篇

基于胶囊网络的智能交通标志识别方法

陈立潮1, 郑佳敏1, 曹建芳1,2, 潘理虎1, 张睿1   

  1. 1. 太原科技大学 计算机科学与技术学院, 太原 030024;
    2. 忻州师范学院 计算机系, 山西 忻州 034000
  • 收稿日期:2019-09-23 修回日期:2019-11-04 出版日期:2020-04-10 发布日期:2020-04-17
  • 通讯作者: 曹建芳
  • 作者简介:陈立潮(1961-),男,山西万荣人,教授,博士,CCF高级会员,主要研究方向:大数据软件工程、智能图像信息处理;郑佳敏(1994-),男,山西晋中人,硕士研究生,主要研究方向:图像处理、深度学习;曹建芳(1976-),女,山西忻州人,教授,博士,CCF高级会员,主要研究方向:数字图像理解、大数据;潘理虎(1974-),男,河南上蔡人,副教授,博士,CCF高级会员,主要研究方向:智能信息处理;张睿(1987-),男,山西太原人,讲师,博士,CCF会员,主要研究方向:智能信息处理。
  • 基金资助:
    山西省应用基础研究项目(201801D221179)。

Intelligent traffic sign recognition method based on capsule network

CHEN Lichao1, ZHENG Jiamin1, CAO Jianfang1,2, PAN Lihu1, ZHANG Rui1   

  1. 1. College of Computer Science and Technology, Taiyuan University of Science and Technology, Taiyuan Shanxi 030024, China;
    2. Computer Department, Xinzhou Teachers University, Xinzhou Shanxi 034000, China
  • Received:2019-09-23 Revised:2019-11-04 Online:2020-04-10 Published:2020-04-17
  • Supported by:
    This work is partially supported by the Applied Basic Research Project of Shanxi Province(201801D221179).

摘要: 针对卷积神经网络的标量神经元无法表达特征位置信息,对复杂的车辆行驶环境适应性差,导致交通标志识别率低的问题,提出一种基于胶囊网络的智能交通标志识别方法。首先采用超深度卷积神经网络改进特征提取部分,然后在主胶囊层引入池化层,最后采用移动指数平均法改进了动态路由算法。在GTSRB数据集上的测试结果表明,改进后的胶囊网络方法在特殊场景下的识别精度提高了10.02个百分点,相对于传统的卷积神经网络,该方法的单张图片的识别时间缩短了2.09 ms。实验结果表明,改进后的胶囊网络方法能满足准确、实时的交通标志识别要求。

关键词: 交通标志识别, 胶囊网络, 超深度卷积, 动态路由算法, 深度学习

Abstract: The scalar neurons of convolutional neural networks cannot express the feature location information,and have poor adaptability to the complex vehicle driving environment,resulting in low traffic sign recognition rate. Therefore,an intelligent traffic sign recognition method based on capsule network was proposed. Firstly,the very deep convolutional neural network was used to improve the feature extraction part. Then,a pooling layer was introduced in the main capsule layer. Finally,the movement index average method was used for improving the dynamic routing algorithm. The test results on the GTSRB dataset show that the improved capsule network method improves the recognition accuracy in special scenes by 10. 02 percentage points. Compared with the traditional convolutional neural network,the proposed method has the recognition time for single image decreased by 2. 09 ms. Experimental results show that the improved capsule network method can meet the requirement of accurate and real-time traffic sign recognition.

Key words: traffic sign recognition, capsule network, very deep convolution, dynamic routing algorithm, deep learning

中图分类号: