Traffic Sign Recognition With Hinge Loss Trained Convolutional Neural Networks

被引：218

作者：

Jin, Junqi ^{[1
]}

Fu, Kun ^{[1
]}

Zhang, Changshui ^{[1
]}

机构：

[1] Tsinghua Univ, Tsinghua Natl Lab Informat Sci & Technol, State Key Lab Intelligent Technol & Syst, Dept Automat, Beijing 100084, Peoples R China

来源：

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS | 2014年 / 15卷 / 05期

基金：

中国国家自然科学基金;

关键词：

Convolutional neural networks (CNNs); hinge loss; stochastic gradient descent (SGD); traffic sign recognition (TSR);

D O I：

10.1109/TITS.2014.2308281

中图分类号：

TU [建筑科学];

学科分类号：

0813 ;

摘要：

Traffic sign recognition (TSR) is an important and challenging task for intelligent transportation systems. We describe the details of our model's architecture for TSR and suggest a hinge loss stochastic gradient descent (HLSGD) method to train convolutional neural networks (CNNs). Our CNN consists of three stages (70-110-180) with 1 162 284 trainable parameters. The HLSGD is evaluated on the German Traffic Sign Recognition Benchmark, which offers a faster and more stable convergence and a state-of-the-art recognition rate of 99.65%. We write a graphics processing unit package to train several CNNs and establish the final classifier in an ensemble way.

引用

页码：1991 / 2000

页数：10

共 26 条

[11]

Le QV, 2013, INT CONF ACOUST SPEE, P8595, DOI 10.1109/ICASSP.2013.6639343

[12] Efficient backprop [J].