Improving road damage maintenance in South Africa using deep learning

Devesh Mothilall; Terence Van Zyl

doi:10.1109/ACDSA59508.2024.10467363

Improving road damage maintenance in South Africa using deep learning

Devesh Mothilall
, Terence Van Zyl

Institute for Intelligent Systems

University of Johannesburg

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

4 Citations (Scopus)

Abstract

The preservation of the road infrastructure is one of the essential factors for a safe, economical and sustainable transport system. Manually collecting data is tedious. This field is intended to benefit from the advancement of artificial intelligence technologies. Advances in deep learning enable the automatic detection of road damage from the collected road images. This work proposes to use an Indian subset of the Road Damage Dataset (RDD) 2022, which has a plethora of images of streets worldwide. The data is processed and labelled. Then, a YOLOv5s model is trained and validated. The model is evaluated against 1959 test images, and the results are tabulated and discussed. The proposed approach has the F1 results of 41% for road damage data collected from the RDD 2022 India subset. In the future, it is recommended to use the broader Global RDD 2022 data to train more robust models with higher accuracy.

Original language	English
Title of host publication	International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9798350394528
DOIs	https://doi.org/10.1109/ACDSA59508.2024.10467363
Publication status	Published - 2024
Event	2024 International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024 - Victoria, Seychelles Duration: 1 Feb 2024 → 2 Feb 2024

Publication series

Name	International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024

Conference

Conference	2024 International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024
Country/Territory	Seychelles
City	Victoria
Period	1/02/24 → 2/02/24

UN SDGs

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

SDG 9 Industry, Innovation, and Infrastructure

Keywords

Machine learning
RDD2022_India
Road Damage Detection
Road Infrastructure
sustainable transport
YOLO5

ASJC Scopus subject areas

Information Systems
Software
Information Systems and Management
Health Informatics
Artificial Intelligence
Computer Science Applications

Access to Document

10.1109/ACDSA59508.2024.10467363

Cite this

Mothilall, D., & Van Zyl, T. (2024). Improving road damage maintenance in South Africa using deep learning. In International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024 (International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ACDSA59508.2024.10467363

Mothilall, Devesh ; Van Zyl, Terence. / Improving road damage maintenance in South Africa using deep learning. International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024. Institute of Electrical and Electronics Engineers Inc., 2024. (International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024).

@inproceedings{d7bd4e999fe74f29807cff794c604dec,

title = "Improving road damage maintenance in South Africa using deep learning",

abstract = "The preservation of the road infrastructure is one of the essential factors for a safe, economical and sustainable transport system. Manually collecting data is tedious. This field is intended to benefit from the advancement of artificial intelligence technologies. Advances in deep learning enable the automatic detection of road damage from the collected road images. This work proposes to use an Indian subset of the Road Damage Dataset (RDD) 2022, which has a plethora of images of streets worldwide. The data is processed and labelled. Then, a YOLOv5s model is trained and validated. The model is evaluated against 1959 test images, and the results are tabulated and discussed. The proposed approach has the F1 results of 41\% for road damage data collected from the RDD 2022 India subset. In the future, it is recommended to use the broader Global RDD 2022 data to train more robust models with higher accuracy.",

keywords = "Machine learning, RDD2022\_India, Road Damage Detection, Road Infrastructure, sustainable transport, YOLO5",

author = "Devesh Mothilall and \{Van Zyl\}, Terence",

note = "Publisher Copyright: {\textcopyright} 2024 IEEE.; 2024 International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024 ; Conference date: 01-02-2024 Through 02-02-2024",

year = "2024",

doi = "10.1109/ACDSA59508.2024.10467363",

language = "English",

series = "International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024",

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

booktitle = "International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024",

address = "United States",

}

Mothilall, D & Van Zyl, T 2024, Improving road damage maintenance in South Africa using deep learning. in International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024. International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024, Institute of Electrical and Electronics Engineers Inc., 2024 International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024, Victoria, Seychelles, 1/02/24. https://doi.org/10.1109/ACDSA59508.2024.10467363

Improving road damage maintenance in South Africa using deep learning. / Mothilall, Devesh; Van Zyl, Terence.
International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024. Institute of Electrical and Electronics Engineers Inc., 2024. (International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024).

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

TY - GEN

T1 - Improving road damage maintenance in South Africa using deep learning

AU - Mothilall, Devesh

AU - Van Zyl, Terence

PY - 2024

Y1 - 2024

N2 - The preservation of the road infrastructure is one of the essential factors for a safe, economical and sustainable transport system. Manually collecting data is tedious. This field is intended to benefit from the advancement of artificial intelligence technologies. Advances in deep learning enable the automatic detection of road damage from the collected road images. This work proposes to use an Indian subset of the Road Damage Dataset (RDD) 2022, which has a plethora of images of streets worldwide. The data is processed and labelled. Then, a YOLOv5s model is trained and validated. The model is evaluated against 1959 test images, and the results are tabulated and discussed. The proposed approach has the F1 results of 41% for road damage data collected from the RDD 2022 India subset. In the future, it is recommended to use the broader Global RDD 2022 data to train more robust models with higher accuracy.

AB - The preservation of the road infrastructure is one of the essential factors for a safe, economical and sustainable transport system. Manually collecting data is tedious. This field is intended to benefit from the advancement of artificial intelligence technologies. Advances in deep learning enable the automatic detection of road damage from the collected road images. This work proposes to use an Indian subset of the Road Damage Dataset (RDD) 2022, which has a plethora of images of streets worldwide. The data is processed and labelled. Then, a YOLOv5s model is trained and validated. The model is evaluated against 1959 test images, and the results are tabulated and discussed. The proposed approach has the F1 results of 41% for road damage data collected from the RDD 2022 India subset. In the future, it is recommended to use the broader Global RDD 2022 data to train more robust models with higher accuracy.

KW - Machine learning

KW - RDD2022_India

KW - Road Damage Detection

KW - Road Infrastructure

KW - sustainable transport

KW - YOLO5

UR - https://www.scopus.com/pages/publications/85189943693

U2 - 10.1109/ACDSA59508.2024.10467363

DO - 10.1109/ACDSA59508.2024.10467363

M3 - Conference contribution

AN - SCOPUS:85189943693

T3 - International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024

BT - International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2024 International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024

Y2 - 1 February 2024 through 2 February 2024

ER -

Mothilall D, Van Zyl T. Improving road damage maintenance in South Africa using deep learning. In International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024. Institute of Electrical and Electronics Engineers Inc. 2024. (International Conference on Artificial Intelligence, Computer, Data Sciences, and Applications, ACDSA 2024). doi: 10.1109/ACDSA59508.2024.10467363

Improving road damage maintenance in South Africa using deep learning

Abstract

Publication series

Conference

UN SDGs

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this