Autonomous road potholes detection on video

Jia Juang Koh; Timothy Tzen Vun Yap; Hu Ng; Vik Tor Goh; Hau Lee Tong; Chiung Ching Ho; Thiam Yong Kuek

doi:10.1007/978-981-13-2622-6_14

Autonomous road potholes detection on video

Jia Juang Koh, Timothy Tzen Vun Yap^*, Hu Ng, Vik Tor Goh, Hau Lee Tong, Chiung Ching Ho, Thiam Yong Kuek

^*Corresponding author for this work

School of Mathematical & Computer Sciences

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

10 Citations (Scopus)

Abstract

This research work explores the possibility of using deep learning to produce an autonomous system for detecting potholes on video to assist in road monitoring and maintenance. Video data of roads was collected using a GoPro camera mounted on a car. Region-based Fully Convolutional Networks (RFCN) was employed to produce the model to detect potholes from images, and validated on the collected videos. The R-FCN model is able to achieve a Mean Average Precision (MAP) of 89% and a True Positive Rate (TPR) of 89% with no false positive.

Original language	English
Title of host publication	Computational Science and Technology
Editors	Rayner Alfred, Ag Asri Ag Ibrahim, Yuto Lim, Patricia Anthony
Publisher	Springer
Pages	137-143
Number of pages	7
ISBN (Electronic)	9789811326226
ISBN (Print)	9789811326219
DOIs	https://doi.org/10.1007/978-981-13-2622-6_14
Publication status	Published - 2019
Event	5th International Conference on Computational Science and Technology 2018 - Kota Kinabalu, Malaysia Duration: 29 Aug 2018 → 30 Aug 2018

Publication series

Name	Lecture Notes in Electrical Engineering
Volume	481
ISSN (Print)	1876-1100
ISSN (Electronic)	1876-1119

Conference

Conference	5th International Conference on Computational Science and Technology 2018
Abbreviated title	ICCST 2018
Country/Territory	Malaysia
City	Kota Kinabalu
Period	29/08/18 → 30/08/18

Keywords

Deep learning
Machine learning
Object identification
Road surface defects
Video data

ASJC Scopus subject areas

Industrial and Manufacturing Engineering

Access to Document

10.1007/978-981-13-2622-6_14

Cite this

@inproceedings{c1f4d23e903148cdb2dad7d7cf237260,

title = "Autonomous road potholes detection on video",

abstract = "This research work explores the possibility of using deep learning to produce an autonomous system for detecting potholes on video to assist in road monitoring and maintenance. Video data of roads was collected using a GoPro camera mounted on a car. Region-based Fully Convolutional Networks (RFCN) was employed to produce the model to detect potholes from images, and validated on the collected videos. The R-FCN model is able to achieve a Mean Average Precision (MAP) of 89% and a True Positive Rate (TPR) of 89% with no false positive.",

keywords = "Deep learning, Machine learning, Object identification, Road surface defects, Video data",

author = "Koh, {Jia Juang} and Yap, {Timothy Tzen Vun} and Hu Ng and Goh, {Vik Tor} and Tong, {Hau Lee} and Ho, {Chiung Ching} and Kuek, {Thiam Yong}",

note = "Funding Information: Acknowledgements. Financial support from the Ministry of Higher Education, Malaysia, under the Fundamental Research Grant Scheme with grant number FRGS/1/2015/SG07/MMU/02/1, as well as the Multimedia University Capex Fund with Project ID MMUI/CAPEX170008, are gratefully acknowledged. Publisher Copyright: {\textcopyright} Springer Nature Singapore Pte Ltd. 2019.; 5th International Conference on Computational Science and Technology 2018, ICCST 2018 ; Conference date: 29-08-2018 Through 30-08-2018",

year = "2019",

doi = "10.1007/978-981-13-2622-6_14",

language = "English",

isbn = "9789811326219",

series = "Lecture Notes in Electrical Engineering",

publisher = "Springer",

pages = "137--143",

editor = "Rayner Alfred and Ibrahim, {Ag Asri Ag} and Yuto Lim and Patricia Anthony",

booktitle = "Computational Science and Technology",

}

Koh, JJ, Yap, TTV, Ng, H, Goh, VT, Tong, HL, Ho, CC & Kuek, TY 2019, Autonomous road potholes detection on video. in R Alfred, AAA Ibrahim, Y Lim & P Anthony (eds), Computational Science and Technology. Lecture Notes in Electrical Engineering, vol. 481, Springer, pp. 137-143, 5th International Conference on Computational Science and Technology 2018, Kota Kinabalu, Malaysia, 29/08/18. https://doi.org/10.1007/978-981-13-2622-6_14

Autonomous road potholes detection on video. / Koh, Jia Juang; Yap, Timothy Tzen Vun; Ng, Hu et al.
Computational Science and Technology. ed. / Rayner Alfred; Ag Asri Ag Ibrahim; Yuto Lim; Patricia Anthony. Springer, 2019. p. 137-143 (Lecture Notes in Electrical Engineering; Vol. 481).

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

TY - GEN

T1 - Autonomous road potholes detection on video

AU - Koh, Jia Juang

AU - Yap, Timothy Tzen Vun

AU - Ng, Hu

AU - Goh, Vik Tor

AU - Tong, Hau Lee

AU - Ho, Chiung Ching

AU - Kuek, Thiam Yong

N1 - Funding Information: Acknowledgements. Financial support from the Ministry of Higher Education, Malaysia, under the Fundamental Research Grant Scheme with grant number FRGS/1/2015/SG07/MMU/02/1, as well as the Multimedia University Capex Fund with Project ID MMUI/CAPEX170008, are gratefully acknowledged. Publisher Copyright: © Springer Nature Singapore Pte Ltd. 2019.

PY - 2019

Y1 - 2019

N2 - This research work explores the possibility of using deep learning to produce an autonomous system for detecting potholes on video to assist in road monitoring and maintenance. Video data of roads was collected using a GoPro camera mounted on a car. Region-based Fully Convolutional Networks (RFCN) was employed to produce the model to detect potholes from images, and validated on the collected videos. The R-FCN model is able to achieve a Mean Average Precision (MAP) of 89% and a True Positive Rate (TPR) of 89% with no false positive.

AB - This research work explores the possibility of using deep learning to produce an autonomous system for detecting potholes on video to assist in road monitoring and maintenance. Video data of roads was collected using a GoPro camera mounted on a car. Region-based Fully Convolutional Networks (RFCN) was employed to produce the model to detect potholes from images, and validated on the collected videos. The R-FCN model is able to achieve a Mean Average Precision (MAP) of 89% and a True Positive Rate (TPR) of 89% with no false positive.

KW - Deep learning

KW - Machine learning

KW - Object identification

KW - Road surface defects

KW - Video data

UR - http://www.scopus.com/inward/record.url?scp=85053267717&partnerID=8YFLogxK

U2 - 10.1007/978-981-13-2622-6_14

DO - 10.1007/978-981-13-2622-6_14

M3 - Conference contribution

AN - SCOPUS:85053267717

SN - 9789811326219

T3 - Lecture Notes in Electrical Engineering

SP - 137

EP - 143

BT - Computational Science and Technology

A2 - Alfred, Rayner

A2 - Ibrahim, Ag Asri Ag

A2 - Lim, Yuto

A2 - Anthony, Patricia

PB - Springer

T2 - 5th International Conference on Computational Science and Technology 2018

Y2 - 29 August 2018 through 30 August 2018

ER -

Autonomous road potholes detection on video

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this