Deep learning approach for U.S. traffic sign recognition

Emmanuel B. Nuakoh; Kaushik Roy; Xiaohong Yuan; Albert Esterline

doi:10.1145/3342999.3343016

Deep learning approach for U.S. traffic sign recognition

Emmanuel B. Nuakoh, Kaushik Roy, Xiaohong Yuan, Albert Esterline

Industrial and systems engineering with North Carolina A&T State University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Advanced Driver Assistant Systems (ADAS) have seen massive improvements in recent years; from detecting pedestrians, road lanes, traffic signs and signals, and vehicles to recognizing and tracking traffic signs. Traffic sign recognition systems are used to detect and classify the traffic signs. This research is focused on the classification aspect of the ADAS; i.e. identifying the class a traffic sign belongs to. Most of the current ADAS that use the U.S. traffic signs are limited to speed limit signs recognition only. This work seeks to expand the corridors of U.S. traffic signs recognition to cover all the publicly available classes. The research adopts the VGGNet architecture modified to classify U.S. traffic signs provided by the LISA TS benchmark. The original VGGNet was used to classify The German Traffic Sign Recognition Benchmark and reported an accuracy of 98.7%. This work recorded a validation accuracy of 99.04%.

Original language	English
Title of host publication	Proceedings of the 2019 3rd International Conference on Deep Learning Technologies, ICDLT 2019
Publisher	Association for Computing Machinery
Pages	47-50
Number of pages	4
ISBN (Electronic)	9781450371605
DOIs	https://doi.org/10.1145/3342999.3343016
State	Published - Jul 5 2019
Externally published	Yes
Event	3rd International Conference on Deep Learning Technologies, ICDLT 2019 - Xiamen, China Duration: Jul 5 2019 → Jul 7 2019

Publication series

Name	ACM International Conference Proceeding Series

Conference

Conference	3rd International Conference on Deep Learning Technologies, ICDLT 2019
Country/Territory	China
City	Xiamen
Period	07/5/19 → 07/7/19

Keywords

Computer Vision
Convolutional Neural Network
Deep Learning
Deep Neural Networks
Traffic Sign Recognition

Access to Document

10.1145/3342999.3343016

Cite this

@inproceedings{99159ab400ff4eb39b1d0c7ae62c5576,

title = "Deep learning approach for U.S. traffic sign recognition",

abstract = "Advanced Driver Assistant Systems (ADAS) have seen massive improvements in recent years; from detecting pedestrians, road lanes, traffic signs and signals, and vehicles to recognizing and tracking traffic signs. Traffic sign recognition systems are used to detect and classify the traffic signs. This research is focused on the classification aspect of the ADAS; i.e. identifying the class a traffic sign belongs to. Most of the current ADAS that use the U.S. traffic signs are limited to speed limit signs recognition only. This work seeks to expand the corridors of U.S. traffic signs recognition to cover all the publicly available classes. The research adopts the VGGNet architecture modified to classify U.S. traffic signs provided by the LISA TS benchmark. The original VGGNet was used to classify The German Traffic Sign Recognition Benchmark and reported an accuracy of 98.7\%. This work recorded a validation accuracy of 99.04\%.",

keywords = "Computer Vision, Convolutional Neural Network, Deep Learning, Deep Neural Networks, Traffic Sign Recognition",

author = "Nuakoh, \{Emmanuel B.\} and Kaushik Roy and Xiaohong Yuan and Albert Esterline",

note = "Publisher Copyright: {\textcopyright} 2019 Association for Computing Machinery.; 3rd International Conference on Deep Learning Technologies, ICDLT 2019 ; Conference date: 05-07-2019 Through 07-07-2019",

year = "2019",

month = jul,

day = "5",

doi = "10.1145/3342999.3343016",

language = "English",

series = "ACM International Conference Proceeding Series",

publisher = "Association for Computing Machinery",

pages = "47--50",

booktitle = "Proceedings of the 2019 3rd International Conference on Deep Learning Technologies, ICDLT 2019",

}

Nuakoh, EB, Roy, K , Yuan, X & Esterline, A 2019, Deep learning approach for U.S. traffic sign recognition. in Proceedings of the 2019 3rd International Conference on Deep Learning Technologies, ICDLT 2019. ACM International Conference Proceeding Series, Association for Computing Machinery, pp. 47-50, 3rd International Conference on Deep Learning Technologies, ICDLT 2019, Xiamen, China, 07/5/19. https://doi.org/10.1145/3342999.3343016

Deep learning approach for U.S. traffic sign recognition. / Nuakoh, Emmanuel B.; Roy, Kaushik ; Yuan, Xiaohong et al.
Proceedings of the 2019 3rd International Conference on Deep Learning Technologies, ICDLT 2019. Association for Computing Machinery, 2019. p. 47-50 (ACM International Conference Proceeding Series).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Deep learning approach for U.S. traffic sign recognition

AU - Nuakoh, Emmanuel B.

AU - Roy, Kaushik

AU - Yuan, Xiaohong

AU - Esterline, Albert

PY - 2019/7/5

Y1 - 2019/7/5

N2 - Advanced Driver Assistant Systems (ADAS) have seen massive improvements in recent years; from detecting pedestrians, road lanes, traffic signs and signals, and vehicles to recognizing and tracking traffic signs. Traffic sign recognition systems are used to detect and classify the traffic signs. This research is focused on the classification aspect of the ADAS; i.e. identifying the class a traffic sign belongs to. Most of the current ADAS that use the U.S. traffic signs are limited to speed limit signs recognition only. This work seeks to expand the corridors of U.S. traffic signs recognition to cover all the publicly available classes. The research adopts the VGGNet architecture modified to classify U.S. traffic signs provided by the LISA TS benchmark. The original VGGNet was used to classify The German Traffic Sign Recognition Benchmark and reported an accuracy of 98.7%. This work recorded a validation accuracy of 99.04%.

AB - Advanced Driver Assistant Systems (ADAS) have seen massive improvements in recent years; from detecting pedestrians, road lanes, traffic signs and signals, and vehicles to recognizing and tracking traffic signs. Traffic sign recognition systems are used to detect and classify the traffic signs. This research is focused on the classification aspect of the ADAS; i.e. identifying the class a traffic sign belongs to. Most of the current ADAS that use the U.S. traffic signs are limited to speed limit signs recognition only. This work seeks to expand the corridors of U.S. traffic signs recognition to cover all the publicly available classes. The research adopts the VGGNet architecture modified to classify U.S. traffic signs provided by the LISA TS benchmark. The original VGGNet was used to classify The German Traffic Sign Recognition Benchmark and reported an accuracy of 98.7%. This work recorded a validation accuracy of 99.04%.

KW - Computer Vision

KW - Convolutional Neural Network

KW - Deep Learning

KW - Deep Neural Networks

KW - Traffic Sign Recognition

UR - https://www.scopus.com/pages/publications/85071554622

U2 - 10.1145/3342999.3343016

DO - 10.1145/3342999.3343016

M3 - Conference contribution

T3 - ACM International Conference Proceeding Series

SP - 47

EP - 50

BT - Proceedings of the 2019 3rd International Conference on Deep Learning Technologies, ICDLT 2019

PB - Association for Computing Machinery

T2 - 3rd International Conference on Deep Learning Technologies, ICDLT 2019

Y2 - 5 July 2019 through 7 July 2019

ER -

Deep learning approach for U.S. traffic sign recognition

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this