A real-time automatic pothole detection system using convolution neural networks

Detecting a pothole can help prevent damage to your vehicle and potentially prevent an accident. Different techniques, including machine learning, deep learning models, sensor methods, stereo vision, the internet of things (IoT), and black-box cameras, have already been applied to address the proble...

Full description

Saved in:
Bibliographic Details
Main Author: Bharat, Ricardo (author)
Other Authors: Ikotun, Abiodun M (author), Ezugwu, Absalom E. (author), Abualigah, Laith (author), Shehab, Mohammad (author), Abu Zitar, Raed (author)
Published: 2023
Subjects:
Online Access:https://depot.sorbonne.ae/handle/20.500.12458/1424
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1857415063521984512
author Bharat, Ricardo
author2 Ikotun, Abiodun M
Ezugwu, Absalom E.
Abualigah, Laith
Shehab, Mohammad
Abu Zitar, Raed
author2_role author
author
author
author
author
author_facet Bharat, Ricardo
Ikotun, Abiodun M
Ezugwu, Absalom E.
Abualigah, Laith
Shehab, Mohammad
Abu Zitar, Raed
author_role author
dc.creator.none.fl_str_mv Bharat, Ricardo
Ikotun, Abiodun M
Ezugwu, Absalom E.
Abualigah, Laith
Shehab, Mohammad
Abu Zitar, Raed
dc.date.none.fl_str_mv 2023-07-24T05:24:12Z
2023-07-24T05:24:12Z
2023
dc.format.none.fl_str_mv application/pdf
dc.identifier.none.fl_str_mv 10.54254/2755-2721/6/20230948
2755-2721
2755-273X
https://depot.sorbonne.ae/handle/20.500.12458/1424
10.54254/2755-2721/6/20230948
dc.language.none.fl_str_mv en
dc.relation.none.fl_str_mv Applied and Computational Engineering
2755-2721
dc.subject.none.fl_str_mv Pothole
Machine Learning
Convolution Neural Network
dc.title.none.fl_str_mv A real-time automatic pothole detection system using convolution neural networks
dc.type.none.fl_str_mv Controlled Vocabulary for Resource Type Genres::text::conference object::conference proceedings
description Detecting a pothole can help prevent damage to your vehicle and potentially prevent an accident. Different techniques, including machine learning, deep learning models, sensor methods, stereo vision, the internet of things (IoT), and black-box cameras, have already been applied to address the problem. However, studies have shown that machine learning and deep learning techniques successfully detect potholes. However, because most of these successful attempts are peculiar to the location of the study, we found no study which has addressed the peculiarity of potholes in South Africa using a tailored-trained deep learning model. In this study, we propose using a convolutional neural network (CNN), a type of deep learning model, to address this growing problem on South African roads. To achieve this, a CNN model was designed from scratch and trained with image samples obtained from the context of the study. The classifier was adapted to distinguish between a binary class which identifies the presence or absence of potholes. Results showed a significant performance enhancement at a classification accuracy of 92.72%. The outcome of this study showed that this machine learning approach holds great potential for addressing the challenge of potholes and road bumps in the region and abroad.
id sorbonner_a597bb6de660f79704586f0bc6f666f4
identifier_str_mv 10.54254/2755-2721/6/20230948
2755-2721
2755-273X
language_invalid_str_mv en
network_acronym_str sorbonner
network_name_str Sorbonne University Abu Dhabi repository
oai_identifier_str oai:depot.sorbonne.ae:20.500.12458/1424
publishDate 2023
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
spelling A real-time automatic pothole detection system using convolution neural networksBharat, RicardoIkotun, Abiodun MEzugwu, Absalom E.Abualigah, LaithShehab, MohammadAbu Zitar, RaedPotholeMachine LearningConvolution Neural NetworkDetecting a pothole can help prevent damage to your vehicle and potentially prevent an accident. Different techniques, including machine learning, deep learning models, sensor methods, stereo vision, the internet of things (IoT), and black-box cameras, have already been applied to address the problem. However, studies have shown that machine learning and deep learning techniques successfully detect potholes. However, because most of these successful attempts are peculiar to the location of the study, we found no study which has addressed the peculiarity of potholes in South Africa using a tailored-trained deep learning model. In this study, we propose using a convolutional neural network (CNN), a type of deep learning model, to address this growing problem on South African roads. To achieve this, a CNN model was designed from scratch and trained with image samples obtained from the context of the study. The classifier was adapted to distinguish between a binary class which identifies the presence or absence of potholes. Results showed a significant performance enhancement at a classification accuracy of 92.72%. The outcome of this study showed that this machine learning approach holds great potential for addressing the challenge of potholes and road bumps in the region and abroad.2023-07-24T05:24:12Z2023-07-24T05:24:12Z2023Controlled Vocabulary for Resource Type Genres::text::conference object::conference proceedingsapplication/pdf10.54254/2755-2721/6/202309482755-27212755-273Xhttps://depot.sorbonne.ae/handle/20.500.12458/142410.54254/2755-2721/6/20230948enApplied and Computational Engineering2755-2721oai:depot.sorbonne.ae:20.500.12458/14242024-03-10T07:13:13Z
spellingShingle A real-time automatic pothole detection system using convolution neural networks
Bharat, Ricardo
Pothole
Machine Learning
Convolution Neural Network
title A real-time automatic pothole detection system using convolution neural networks
title_full A real-time automatic pothole detection system using convolution neural networks
title_fullStr A real-time automatic pothole detection system using convolution neural networks
title_full_unstemmed A real-time automatic pothole detection system using convolution neural networks
title_short A real-time automatic pothole detection system using convolution neural networks
title_sort A real-time automatic pothole detection system using convolution neural networks
topic Pothole
Machine Learning
Convolution Neural Network
url https://depot.sorbonne.ae/handle/20.500.12458/1424