Real-Time Monitoring of Road Networks for Pavement Damage Detection Based on Preprocessing and Neural Networks

Nataliya Shakhovska; Vitaliy Yakovyna; Maksym Mysak; Stergios-Aristoteles Mitoulis; Sotirios Argyroudis; Yuriy Syerov

doi:10.3390/bdcc8100136

Real-Time Monitoring of Road Networks for Pavement Damage Detection Based on Preprocessing and Neural Networks

Nataliya Shakhovska, Vitaliy Yakovyna, Maksym Mysak, Stergios-Aristoteles Mitoulis, Sotirios Argyroudis, Yuriy Syerov^*

^*Corresponding author for this work

Civil Engineering

Research output: Contribution to journal › Article › peer-review

43 Downloads (Pure)

Abstract

This paper presents a novel multi-initialization model for recognizing road surface damage, e.g. potholes and cracks, on video using convolutional neural networks (CNNs) in real-time for fast damage recognition. The model is trained by the latest Road Damage Detection dataset, which includes four types of road damage. In addition, the CNN model is updated using pseudo-labeled images from semi-learned methods to improve the performance of the pavement damage detection technique. This study describes the use of the YOLO architecture and optimizes it according to the selected parameters, demonstrating high efficiency and accuracy. The results obtained can enhance the safety and efficiency of road pavement and, hence, its traffic quality and contribute to decision-making for the maintenance and restoration of road infrastructure.

Original language	English
Article number	136
Number of pages	22
Journal	Big Data and Cognitive Computing
Volume	8
Issue number	10
DOIs	https://doi.org/10.3390/bdcc8100136
Publication status	Published - 11 Oct 2024

Keywords

neural networks
classification
damage detection
machine learning
data preprocessing
YOLO architecture
convolutional neural network
pavement

Access to Document

10.3390/bdcc8100136Licence: Creative Commons: Attribution (CC BY)

ShakhovskaN2024RealFinal published version, 10.7 MBLicence: Creative Commons: Attribution (CC BY)

ZEBAI: Innovative methodologies to design Zero-Emission and cost-effective Buildings based on Artificial Intelligence
Mitoulis, S. (Principal Investigator), Faramarzi, A. (Co-Investigator), Freer, M. (Co-Investigator), Fraga, B. (Co-Investigator) & Sharifi, S. (Co-Investigator)
UKRI Horizon Europe Underwriting Innovate UK, European Commission
1/01/24 → 31/12/28
Project: Research

Cite this

@article{7f02f6bbd7844e6fa0d5c18fb775154c,

title = "Real-Time Monitoring of Road Networks for Pavement Damage Detection Based on Preprocessing and Neural Networks",

abstract = "This paper presents a novel multi-initialization model for recognizing road surface damage, e.g. potholes and cracks, on video using convolutional neural networks (CNNs) in real-time for fast damage recognition. The model is trained by the latest Road Damage Detection dataset, which includes four types of road damage. In addition, the CNN model is updated using pseudo-labeled images from semi-learned methods to improve the performance of the pavement damage detection technique. This study describes the use of the YOLO architecture and optimizes it according to the selected parameters, demonstrating high efficiency and accuracy. The results obtained can enhance the safety and efficiency of road pavement and, hence, its traffic quality and contribute to decision-making for the maintenance and restoration of road infrastructure.",

keywords = "neural networks, classification, damage detection, machine learning, data preprocessing, YOLO architecture, convolutional neural network, pavement",

author = "Nataliya Shakhovska and Vitaliy Yakovyna and Maksym Mysak and Stergios-Aristoteles Mitoulis and Sotirios Argyroudis and Yuriy Syerov",

year = "2024",

month = oct,

day = "11",

doi = "10.3390/bdcc8100136",

language = "English",

volume = "8",

journal = "Big Data and Cognitive Computing",

issn = "2504-2289",

publisher = "MDPI",

number = "10",

}

TY - JOUR

T1 - Real-Time Monitoring of Road Networks for Pavement Damage Detection Based on Preprocessing and Neural Networks

AU - Shakhovska, Nataliya

AU - Yakovyna, Vitaliy

AU - Mysak, Maksym

AU - Mitoulis, Stergios-Aristoteles

AU - Argyroudis, Sotirios

AU - Syerov, Yuriy

PY - 2024/10/11

Y1 - 2024/10/11

N2 - This paper presents a novel multi-initialization model for recognizing road surface damage, e.g. potholes and cracks, on video using convolutional neural networks (CNNs) in real-time for fast damage recognition. The model is trained by the latest Road Damage Detection dataset, which includes four types of road damage. In addition, the CNN model is updated using pseudo-labeled images from semi-learned methods to improve the performance of the pavement damage detection technique. This study describes the use of the YOLO architecture and optimizes it according to the selected parameters, demonstrating high efficiency and accuracy. The results obtained can enhance the safety and efficiency of road pavement and, hence, its traffic quality and contribute to decision-making for the maintenance and restoration of road infrastructure.

AB - This paper presents a novel multi-initialization model for recognizing road surface damage, e.g. potholes and cracks, on video using convolutional neural networks (CNNs) in real-time for fast damage recognition. The model is trained by the latest Road Damage Detection dataset, which includes four types of road damage. In addition, the CNN model is updated using pseudo-labeled images from semi-learned methods to improve the performance of the pavement damage detection technique. This study describes the use of the YOLO architecture and optimizes it according to the selected parameters, demonstrating high efficiency and accuracy. The results obtained can enhance the safety and efficiency of road pavement and, hence, its traffic quality and contribute to decision-making for the maintenance and restoration of road infrastructure.

KW - neural networks

KW - classification

KW - damage detection

KW - machine learning

KW - data preprocessing

KW - YOLO architecture

KW - convolutional neural network

KW - pavement

U2 - 10.3390/bdcc8100136

DO - 10.3390/bdcc8100136

M3 - Article

SN - 2504-2289

VL - 8

JO - Big Data and Cognitive Computing

JF - Big Data and Cognitive Computing

IS - 10

M1 - 136

ER -

Real-Time Monitoring of Road Networks for Pavement Damage Detection Based on Preprocessing and Neural Networks

Abstract

Keywords

Access to Document

Fingerprint

Projects

ZEBAI: Innovative methodologies to design Zero-Emission and cost-effective Buildings based on Artificial Intelligence

Cite this