Portable convolution neural networks for traffic sign recognition in intelligent transportation systems

Junhao Zhou; Hong Ning Dai; Hao Wang

doi:10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00032

Portable convolution neural networks for traffic sign recognition in intelligent transportation systems

Junhao Zhou, Hong Ning Dai, Hao Wang

Department of Computer Science

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

4 Citations (Scopus)

Abstract

Deep convolutional neural networks (CNN) have the strength in traffic-sign classification in terms of high accuracy. However, CNN models usually contains multiple layers with a large number of parameters consequently leading to a large model size. The bulky model size of CNN models prevents them from the wide deployment in mobile and portable devices in Intelligent Transportation Systems. In this paper, we design and develop a portable convolutional neural network (namely portable CNN) structure used for traffic-sign classification. This portable CNN model contains a stacked convolutional structure consisting of factorization and compression modules. We conducted extensive experiments to evaluate the performance of the proposed Portable CNN model. Experimental results show that our model has the advantages of smaller model size while maintaining high classification accuracy, compared with conventional CNN models.

Original language	English
Title of host publication	Proceedings - 2019 IEEE International Congress on Cybermatics, 12th IEEE International Conference on Internet of Things, 15th IEEE International Conference on Green Computing and Communications, 12th IEEE International Conference on Cyber, Physical and Social Computing and 5th IEEE International Conference on Smart Data, iThings/GreenCom/CPSCom/SmartData 2019
Publisher	IEEE
Pages	52-57
Number of pages	6
ISBN (Electronic)	9781728129808
ISBN (Print)	9781728129815
DOIs	https://doi.org/10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00032
Publication status	Published - Jul 2019
Event	12th IEEE International Conference on Internet of Things, 15th IEEE International Conference on Green Computing and Communications, 12th IEEE International Conference on Cyber, Physical and Social Computing and 5th IEEE International Conference on Smart Data, iThings/GreenCom/CPSCom/SmartData 2019 - Atlanta, United States Duration: 14 Jul 2019 → 17 Jul 2019 https://ieeexplore.ieee.org/xpl/conhome/8867850/proceeding

Publication series

Name	Proceedings - IEEE/ACM Int'l Conference on & Int'l Conference on Cyber, Physical and Social Computing (CPSCom) Green Computing and Communications (GreenCom)

Conference

Conference	12th IEEE International Conference on Internet of Things, 15th IEEE International Conference on Green Computing and Communications, 12th IEEE International Conference on Cyber, Physical and Social Computing and 5th IEEE International Conference on Smart Data, iThings/GreenCom/CPSCom/SmartData 2019
Country/Territory	United States
City	Atlanta
Period	14/07/19 → 17/07/19
Internet address	https://ieeexplore.ieee.org/xpl/conhome/8867850/proceeding

Scopus Subject Areas

Artificial Intelligence
Computer Networks and Communications
Renewable Energy, Sustainability and the Environment
Hardware and Architecture
Information Systems and Management
Safety, Risk, Reliability and Quality
Communication

User-Defined Keywords

Convolutional neural networks
Factorization
Intelligent Transportation Systems
Model Compression
Portable

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00032

Cite this

Zhou, J., Dai, H. N., & Wang, H. (2019). Portable convolution neural networks for traffic sign recognition in intelligent transportation systems. In Proceedings - 2019 IEEE International Congress on Cybermatics, 12th IEEE International Conference on Internet of Things, 15th IEEE International Conference on Green Computing and Communications, 12th IEEE International Conference on Cyber, Physical and Social Computing and 5th IEEE International Conference on Smart Data, iThings/GreenCom/CPSCom/SmartData 2019 (pp. 52-57). (Proceedings - IEEE/ACM Int'l Conference on & Int'l Conference on Cyber, Physical and Social Computing (CPSCom) Green Computing and Communications (GreenCom)). IEEE. https://doi.org/10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00032

Zhou, Junhao ; Dai, Hong Ning ; Wang, Hao. / Portable convolution neural networks for traffic sign recognition in intelligent transportation systems. Proceedings - 2019 IEEE International Congress on Cybermatics, 12th IEEE International Conference on Internet of Things, 15th IEEE International Conference on Green Computing and Communications, 12th IEEE International Conference on Cyber, Physical and Social Computing and 5th IEEE International Conference on Smart Data, iThings/GreenCom/CPSCom/SmartData 2019. IEEE, 2019. pp. 52-57 (Proceedings - IEEE/ACM Int'l Conference on & Int'l Conference on Cyber, Physical and Social Computing (CPSCom) Green Computing and Communications (GreenCom)).

@inproceedings{1ce8d2d878954ce697b4edb5080aec47,

title = "Portable convolution neural networks for traffic sign recognition in intelligent transportation systems",

abstract = "Deep convolutional neural networks (CNN) have the strength in traffic-sign classification in terms of high accuracy. However, CNN models usually contains multiple layers with a large number of parameters consequently leading to a large model size. The bulky model size of CNN models prevents them from the wide deployment in mobile and portable devices in Intelligent Transportation Systems. In this paper, we design and develop a portable convolutional neural network (namely portable CNN) structure used for traffic-sign classification. This portable CNN model contains a stacked convolutional structure consisting of factorization and compression modules. We conducted extensive experiments to evaluate the performance of the proposed Portable CNN model. Experimental results show that our model has the advantages of smaller model size while maintaining high classification accuracy, compared with conventional CNN models.",

keywords = "Convolutional neural networks, Factorization, Intelligent Transportation Systems, Model Compression, Portable",

author = "Junhao Zhou and Dai, {Hong Ning} and Hao Wang",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 12th IEEE International Conference on Internet of Things, 15th IEEE International Conference on Green Computing and Communications, 12th IEEE International Conference on Cyber, Physical and Social Computing and 5th IEEE International Conference on Smart Data, iThings/GreenCom/CPSCom/SmartData 2019 ; Conference date: 14-07-2019 Through 17-07-2019",

year = "2019",

month = jul,

doi = "10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00032",

language = "English",

isbn = "9781728129815",

series = "Proceedings - IEEE/ACM Int'l Conference on & Int'l Conference on Cyber, Physical and Social Computing (CPSCom) Green Computing and Communications (GreenCom)",

publisher = "IEEE",

pages = "52--57",

booktitle = "Proceedings - 2019 IEEE International Congress on Cybermatics, 12th IEEE International Conference on Internet of Things, 15th IEEE International Conference on Green Computing and Communications, 12th IEEE International Conference on Cyber, Physical and Social Computing and 5th IEEE International Conference on Smart Data, iThings/GreenCom/CPSCom/SmartData 2019",

address = "United States",

url = "https://ieeexplore.ieee.org/xpl/conhome/8867850/proceeding",

}

Zhou, J, Dai, HN & Wang, H 2019, Portable convolution neural networks for traffic sign recognition in intelligent transportation systems. in Proceedings - 2019 IEEE International Congress on Cybermatics, 12th IEEE International Conference on Internet of Things, 15th IEEE International Conference on Green Computing and Communications, 12th IEEE International Conference on Cyber, Physical and Social Computing and 5th IEEE International Conference on Smart Data, iThings/GreenCom/CPSCom/SmartData 2019. Proceedings - IEEE/ACM Int'l Conference on & Int'l Conference on Cyber, Physical and Social Computing (CPSCom) Green Computing and Communications (GreenCom), IEEE, pp. 52-57, 12th IEEE International Conference on Internet of Things, 15th IEEE International Conference on Green Computing and Communications, 12th IEEE International Conference on Cyber, Physical and Social Computing and 5th IEEE International Conference on Smart Data, iThings/GreenCom/CPSCom/SmartData 2019, Atlanta, United States, 14/07/19. https://doi.org/10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00032

Portable convolution neural networks for traffic sign recognition in intelligent transportation systems. / Zhou, Junhao; Dai, Hong Ning; Wang, Hao.
Proceedings - 2019 IEEE International Congress on Cybermatics, 12th IEEE International Conference on Internet of Things, 15th IEEE International Conference on Green Computing and Communications, 12th IEEE International Conference on Cyber, Physical and Social Computing and 5th IEEE International Conference on Smart Data, iThings/GreenCom/CPSCom/SmartData 2019. IEEE, 2019. p. 52-57 (Proceedings - IEEE/ACM Int'l Conference on & Int'l Conference on Cyber, Physical and Social Computing (CPSCom) Green Computing and Communications (GreenCom)).

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

TY - GEN

T1 - Portable convolution neural networks for traffic sign recognition in intelligent transportation systems

AU - Zhou, Junhao

AU - Dai, Hong Ning

AU - Wang, Hao

PY - 2019/7

Y1 - 2019/7

N2 - Deep convolutional neural networks (CNN) have the strength in traffic-sign classification in terms of high accuracy. However, CNN models usually contains multiple layers with a large number of parameters consequently leading to a large model size. The bulky model size of CNN models prevents them from the wide deployment in mobile and portable devices in Intelligent Transportation Systems. In this paper, we design and develop a portable convolutional neural network (namely portable CNN) structure used for traffic-sign classification. This portable CNN model contains a stacked convolutional structure consisting of factorization and compression modules. We conducted extensive experiments to evaluate the performance of the proposed Portable CNN model. Experimental results show that our model has the advantages of smaller model size while maintaining high classification accuracy, compared with conventional CNN models.

AB - Deep convolutional neural networks (CNN) have the strength in traffic-sign classification in terms of high accuracy. However, CNN models usually contains multiple layers with a large number of parameters consequently leading to a large model size. The bulky model size of CNN models prevents them from the wide deployment in mobile and portable devices in Intelligent Transportation Systems. In this paper, we design and develop a portable convolutional neural network (namely portable CNN) structure used for traffic-sign classification. This portable CNN model contains a stacked convolutional structure consisting of factorization and compression modules. We conducted extensive experiments to evaluate the performance of the proposed Portable CNN model. Experimental results show that our model has the advantages of smaller model size while maintaining high classification accuracy, compared with conventional CNN models.

KW - Convolutional neural networks

KW - Factorization

KW - Intelligent Transportation Systems

KW - Model Compression

KW - Portable

UR - http://www.scopus.com/inward/record.url?scp=85074846166&partnerID=8YFLogxK

U2 - 10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00032

DO - 10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00032

M3 - Conference proceeding

AN - SCOPUS:85074846166

SN - 9781728129815

T3 - Proceedings - IEEE/ACM Int'l Conference on & Int'l Conference on Cyber, Physical and Social Computing (CPSCom) Green Computing and Communications (GreenCom)

SP - 52

EP - 57

BT - Proceedings - 2019 IEEE International Congress on Cybermatics, 12th IEEE International Conference on Internet of Things, 15th IEEE International Conference on Green Computing and Communications, 12th IEEE International Conference on Cyber, Physical and Social Computing and 5th IEEE International Conference on Smart Data, iThings/GreenCom/CPSCom/SmartData 2019

PB - IEEE

T2 - 12th IEEE International Conference on Internet of Things, 15th IEEE International Conference on Green Computing and Communications, 12th IEEE International Conference on Cyber, Physical and Social Computing and 5th IEEE International Conference on Smart Data, iThings/GreenCom/CPSCom/SmartData 2019

Y2 - 14 July 2019 through 17 July 2019

ER -

Zhou J, Dai HN, Wang H. Portable convolution neural networks for traffic sign recognition in intelligent transportation systems. In Proceedings - 2019 IEEE International Congress on Cybermatics, 12th IEEE International Conference on Internet of Things, 15th IEEE International Conference on Green Computing and Communications, 12th IEEE International Conference on Cyber, Physical and Social Computing and 5th IEEE International Conference on Smart Data, iThings/GreenCom/CPSCom/SmartData 2019. IEEE. 2019. p. 52-57. (Proceedings - IEEE/ACM Int'l Conference on & Int'l Conference on Cyber, Physical and Social Computing (CPSCom) Green Computing and Communications (GreenCom)). doi: 10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00032

Portable convolution neural networks for traffic sign recognition in intelligent transportation systems

Abstract

Publication series

Conference

Scopus Subject Areas

User-Defined Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this