MAT-CNN-SOPC: Motionless Analysis of Traffic Using Convolutional Neural Networks on System-On-a-Programmable-Chip

Somdip Dey; Grigorios Kalliatakis; Sangeet Saha; Amit Kumar Singh; Shoaib Ehsan; Klaus McDonald-Maier

doi:10.1109/AHS.2018.8541406

MAT-CNN-SOPC: Motionless Analysis of Traffic Using Convolutional Neural Networks on System-On-a-Programmable-Chip

Somdip Dey, Grigorios Kalliatakis, Sangeet Saha, Amit Kumar Singh, Shoaib Ehsan, Klaus McDonald-Maier

University of Essex

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

15 Citations (Scopus)

Abstract

Intelligent Transportation Systems (ITS) have become an important pillar in modern 'smart city' framework which demands intelligent involvement of machines. Traffic load recognition can be categorized as an important and challenging issue for such systems. Recently, Convolutional Neural Network (CNN) models have drawn considerable amount of interest in many areas such as weather classification, human rights violation detection through images, due to its accurate prediction capabilities. This work tackles real-life traffic load recognition problem on System-On-a-Programmable-Chip (SOPC) platform and coin it as MAT-CNN-SOPC, which uses an intelligent retraining mechanism of the CNN with known environments. The proposed methodology is capable of enhancing the efficacy of the approach by 2.44x in comparison to the state-of-art and proven through experimental analysis. We have also introduced a mathematical equation, which is capable of quantifying the suitability of using different CNN models over the other for a particular application based implementation.

Original language	English
Title of host publication	2018 NASA/ESA Conference on Adaptive Hardware and Systems, AHS 2018
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	291-298
Number of pages	8
ISBN (Electronic)	9781538677537, 9781538677520
ISBN (Print)	9781538677544
DOIs	https://doi.org/10.1109/AHS.2018.8541406
Publication status	Published - 22 Nov 2018
Externally published	Yes
Event	2018 NASA/ESA Conference on Adaptive Hardware and Systems - Edinburgh, United Kingdom Duration: 6 Aug 2018 → 9 Aug 2018

Publication series

Name	Proceedings of Conference on Adaptive Hardware and Systems
Publisher	IEEE
ISSN (Print)	1939-7003
ISSN (Electronic)	2471-769X

Conference

Conference	2018 NASA/ESA Conference on Adaptive Hardware and Systems
Abbreviated title	AHS 2018
Country/Territory	United Kingdom
City	Edinburgh
Period	6/08/18 → 9/08/18

UN SDGs

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

SDG 11 Sustainable Cities and Communities

Access to Document

10.1109/AHS.2018.8541406Licence: Unspecified

Cite this

Dey, S., Kalliatakis, G., Saha, S., Singh, A. K., Ehsan, S., & McDonald-Maier, K. (2018). MAT-CNN-SOPC: Motionless Analysis of Traffic Using Convolutional Neural Networks on System-On-a-Programmable-Chip. In 2018 NASA/ESA Conference on Adaptive Hardware and Systems, AHS 2018 (pp. 291-298). Article 8541406 (Proceedings of Conference on Adaptive Hardware and Systems). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/AHS.2018.8541406

Dey, Somdip ; Kalliatakis, Grigorios ; Saha, Sangeet et al. / MAT-CNN-SOPC : Motionless Analysis of Traffic Using Convolutional Neural Networks on System-On-a-Programmable-Chip. 2018 NASA/ESA Conference on Adaptive Hardware and Systems, AHS 2018. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 291-298 (Proceedings of Conference on Adaptive Hardware and Systems).

@inproceedings{6b2ef54cb8764d2787b2444f60076791,

title = "MAT-CNN-SOPC: Motionless Analysis of Traffic Using Convolutional Neural Networks on System-On-a-Programmable-Chip",

abstract = "Intelligent Transportation Systems (ITS) have become an important pillar in modern 'smart city' framework which demands intelligent involvement of machines. Traffic load recognition can be categorized as an important and challenging issue for such systems. Recently, Convolutional Neural Network (CNN) models have drawn considerable amount of interest in many areas such as weather classification, human rights violation detection through images, due to its accurate prediction capabilities. This work tackles real-life traffic load recognition problem on System-On-a-Programmable-Chip (SOPC) platform and coin it as MAT-CNN-SOPC, which uses an intelligent retraining mechanism of the CNN with known environments. The proposed methodology is capable of enhancing the efficacy of the approach by 2.44x in comparison to the state-of-art and proven through experimental analysis. We have also introduced a mathematical equation, which is capable of quantifying the suitability of using different CNN models over the other for a particular application based implementation.",

keywords = "Convolutional neural network (CNN), system-on-a-programmable-chip (SOPC), traffic analysis, traffic density, transfer learning",

author = "Somdip Dey and Grigorios Kalliatakis and Sangeet Saha and Singh, \{Amit Kumar\} and Shoaib Ehsan and Klaus McDonald-Maier",

note = "Funding Information: This work is supported by the UK Engineering and Physical Sciences Research Council EPSRC [EP/R02572X/1 and EP/P017487/1]. Funding Information: This work is supported by the UK Engineering and Physical Sciences Research Council EPSRC [EP/R02572X/1 and EP/P017487/1] and the authors would like to thank the people associated with National Centre for Nuclear Robotics (NCNR) and Extreme Environments for their support and feedback. Somdip would also like to thank everyone from the Embedded and Intelligent Systems Laboratory at the University of Essex for their feedback on this project. Publisher Copyright: {\textcopyright} 2018 IEEE. Copyright: Copyright 2019 Elsevier B.V., All rights reserved.; 2018 NASA/ESA Conference on Adaptive Hardware and Systems, AHS 2018 ; Conference date: 06-08-2018 Through 09-08-2018",

year = "2018",

month = nov,

day = "22",

doi = "10.1109/AHS.2018.8541406",

language = "English",

isbn = "9781538677544",

series = "Proceedings of Conference on Adaptive Hardware and Systems",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "291--298",

booktitle = "2018 NASA/ESA Conference on Adaptive Hardware and Systems, AHS 2018",

address = "United States",

}

Dey, S, Kalliatakis, G, Saha, S, Singh, AK, Ehsan, S & McDonald-Maier, K 2018, MAT-CNN-SOPC: Motionless Analysis of Traffic Using Convolutional Neural Networks on System-On-a-Programmable-Chip. in 2018 NASA/ESA Conference on Adaptive Hardware and Systems, AHS 2018., 8541406, Proceedings of Conference on Adaptive Hardware and Systems, Institute of Electrical and Electronics Engineers Inc., pp. 291-298, 2018 NASA/ESA Conference on Adaptive Hardware and Systems, Edinburgh, United Kingdom, 6/08/18. https://doi.org/10.1109/AHS.2018.8541406

MAT-CNN-SOPC: Motionless Analysis of Traffic Using Convolutional Neural Networks on System-On-a-Programmable-Chip. / Dey, Somdip; Kalliatakis, Grigorios; Saha, Sangeet et al.
2018 NASA/ESA Conference on Adaptive Hardware and Systems, AHS 2018. Institute of Electrical and Electronics Engineers Inc., 2018. p. 291-298 8541406 (Proceedings of Conference on Adaptive Hardware and Systems).

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

TY - GEN

T1 - MAT-CNN-SOPC

T2 - 2018 NASA/ESA Conference on Adaptive Hardware and Systems

AU - Dey, Somdip

AU - Kalliatakis, Grigorios

AU - Saha, Sangeet

AU - Singh, Amit Kumar

AU - Ehsan, Shoaib

AU - McDonald-Maier, Klaus

N1 - Funding Information: This work is supported by the UK Engineering and Physical Sciences Research Council EPSRC [EP/R02572X/1 and EP/P017487/1]. Funding Information: This work is supported by the UK Engineering and Physical Sciences Research Council EPSRC [EP/R02572X/1 and EP/P017487/1] and the authors would like to thank the people associated with National Centre for Nuclear Robotics (NCNR) and Extreme Environments for their support and feedback. Somdip would also like to thank everyone from the Embedded and Intelligent Systems Laboratory at the University of Essex for their feedback on this project. Publisher Copyright: © 2018 IEEE. Copyright: Copyright 2019 Elsevier B.V., All rights reserved.

PY - 2018/11/22

Y1 - 2018/11/22

N2 - Intelligent Transportation Systems (ITS) have become an important pillar in modern 'smart city' framework which demands intelligent involvement of machines. Traffic load recognition can be categorized as an important and challenging issue for such systems. Recently, Convolutional Neural Network (CNN) models have drawn considerable amount of interest in many areas such as weather classification, human rights violation detection through images, due to its accurate prediction capabilities. This work tackles real-life traffic load recognition problem on System-On-a-Programmable-Chip (SOPC) platform and coin it as MAT-CNN-SOPC, which uses an intelligent retraining mechanism of the CNN with known environments. The proposed methodology is capable of enhancing the efficacy of the approach by 2.44x in comparison to the state-of-art and proven through experimental analysis. We have also introduced a mathematical equation, which is capable of quantifying the suitability of using different CNN models over the other for a particular application based implementation.

AB - Intelligent Transportation Systems (ITS) have become an important pillar in modern 'smart city' framework which demands intelligent involvement of machines. Traffic load recognition can be categorized as an important and challenging issue for such systems. Recently, Convolutional Neural Network (CNN) models have drawn considerable amount of interest in many areas such as weather classification, human rights violation detection through images, due to its accurate prediction capabilities. This work tackles real-life traffic load recognition problem on System-On-a-Programmable-Chip (SOPC) platform and coin it as MAT-CNN-SOPC, which uses an intelligent retraining mechanism of the CNN with known environments. The proposed methodology is capable of enhancing the efficacy of the approach by 2.44x in comparison to the state-of-art and proven through experimental analysis. We have also introduced a mathematical equation, which is capable of quantifying the suitability of using different CNN models over the other for a particular application based implementation.

KW - Convolutional neural network (CNN)

KW - system-on-a-programmable-chip (SOPC)

KW - traffic analysis

KW - traffic density

KW - transfer learning

UR - https://www.scopus.com/pages/publications/85059937576

U2 - 10.1109/AHS.2018.8541406

DO - 10.1109/AHS.2018.8541406

M3 - Conference contribution

AN - SCOPUS:85059937576

SN - 9781538677544

T3 - Proceedings of Conference on Adaptive Hardware and Systems

SP - 291

EP - 298

BT - 2018 NASA/ESA Conference on Adaptive Hardware and Systems, AHS 2018

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 6 August 2018 through 9 August 2018

ER -

Dey S, Kalliatakis G, Saha S, Singh AK, Ehsan S, McDonald-Maier K. MAT-CNN-SOPC: Motionless Analysis of Traffic Using Convolutional Neural Networks on System-On-a-Programmable-Chip. In 2018 NASA/ESA Conference on Adaptive Hardware and Systems, AHS 2018. Institute of Electrical and Electronics Engineers Inc. 2018. p. 291-298. 8541406. (Proceedings of Conference on Adaptive Hardware and Systems). doi: 10.1109/AHS.2018.8541406

MAT-CNN-SOPC: Motionless Analysis of Traffic Using Convolutional Neural Networks on System-On-a-Programmable-Chip

Abstract

Publication series

Conference

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this