Cover Image

Classification of traffic accidents’ factors using TrafficRiskClassifier

The TrafficRiskClassifier model introduced in this study adopts an innovative approach that incorporates migration learning, image classification, and self-supervised learning, aiming to significantly improve the accuracy and efficiency of traffic accident risk analysis. Compared with traditional tr...

Full description

Saved in:
Bibliographic Details
Main Authors: Sun, Wei, Abdullah, Lili Nurliyana, Khalid, Fatimah binti, Sulaiman, Puteri Suhaiza
Format: Article
Language:English
Published: KeAi Communications 2024
Online Access:http://psasir.upm.edu.my/id/eprint/113360/1/113360.pdf
http://psasir.upm.edu.my/id/eprint/113360/
https://linkinghub.elsevier.com/retrieve/pii/S2046043024000492
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.upm.eprints.113360
record_format eprints
spelling my.upm.eprints.1133602024-11-19T08:08:03Z http://psasir.upm.edu.my/id/eprint/113360/ Classification of traffic accidents’ factors using TrafficRiskClassifier Sun, Wei Abdullah, Lili Nurliyana Khalid, Fatimah binti Sulaiman, Puteri Suhaiza The TrafficRiskClassifier model introduced in this study adopts an innovative approach that incorporates migration learning, image classification, and self-supervised learning, aiming to significantly improve the accuracy and efficiency of traffic accident risk analysis. Compared with traditional traffic safety analysis techniques, this model focuses on utilizing contextual information and situational data of traffic accidents to achieve a higher level of risk classification accuracy. The core of this approach is to deeply mine and analyze the detailed information in the accident environment, to provide more scientific and effective support for traffic accident risk prevention and response. Initially, by integrating migration learning with image classification techniques, the model proficiently extracts pivotal features from intricate traffic scenarios and formulates initial assessments of accident risks. Subsequently, self-supervised learning is incorporated in this study, augmenting the model's capability to comprehend and categorize accident imagery. The TrafficRiskClassifier model exhibits a generalization ability of 91.82%, 85.16%, and 80.92% on individual classification tasks, respectively, signifying its robust learning capacity and proficiency in managing unseen data. Furthermore, the TrafficRiskClassifier model delineates a functional nexus between accident risk and variables such as weather, road conditions, and personal factors, employing a polynomial regression approach. This methodology not only amplifies the predictive precision of the model but also renders it versatile across diverse scenarios. Through the analysis of various polynomial functions, the model achieves enhanced accuracy in classifying disparate risk levels. The outcomes demonstrate that the TrafficRiskClassifier model can efficaciously amalgamate contextual information within traffic scenarios, thereby achieving more precise classification of traffic accident risks, and consequently serving as an invaluable instrument for urban traffic safety management. © 2024 Tongji University and Tongji University Press KeAi Communications 2024 Article PeerReviewed text en cc_by_nc_nd_4 http://psasir.upm.edu.my/id/eprint/113360/1/113360.pdf Sun, Wei and Abdullah, Lili Nurliyana and Khalid, Fatimah binti and Sulaiman, Puteri Suhaiza (2024) Classification of traffic accidents’ factors using TrafficRiskClassifier. International Journal of Transportation Science and Technology. pp. 1-17. ISSN 2046-0430; eISSN: 2046-0449 (In Press) https://linkinghub.elsevier.com/retrieve/pii/S2046043024000492 10.1016/j.ijtst.2024.05.002
institution Universiti Putra Malaysia
building UPM Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Putra Malaysia
content_source UPM Institutional Repository
url_provider http://psasir.upm.edu.my/
language English
description The TrafficRiskClassifier model introduced in this study adopts an innovative approach that incorporates migration learning, image classification, and self-supervised learning, aiming to significantly improve the accuracy and efficiency of traffic accident risk analysis. Compared with traditional traffic safety analysis techniques, this model focuses on utilizing contextual information and situational data of traffic accidents to achieve a higher level of risk classification accuracy. The core of this approach is to deeply mine and analyze the detailed information in the accident environment, to provide more scientific and effective support for traffic accident risk prevention and response. Initially, by integrating migration learning with image classification techniques, the model proficiently extracts pivotal features from intricate traffic scenarios and formulates initial assessments of accident risks. Subsequently, self-supervised learning is incorporated in this study, augmenting the model's capability to comprehend and categorize accident imagery. The TrafficRiskClassifier model exhibits a generalization ability of 91.82%, 85.16%, and 80.92% on individual classification tasks, respectively, signifying its robust learning capacity and proficiency in managing unseen data. Furthermore, the TrafficRiskClassifier model delineates a functional nexus between accident risk and variables such as weather, road conditions, and personal factors, employing a polynomial regression approach. This methodology not only amplifies the predictive precision of the model but also renders it versatile across diverse scenarios. Through the analysis of various polynomial functions, the model achieves enhanced accuracy in classifying disparate risk levels. The outcomes demonstrate that the TrafficRiskClassifier model can efficaciously amalgamate contextual information within traffic scenarios, thereby achieving more precise classification of traffic accident risks, and consequently serving as an invaluable instrument for urban traffic safety management. © 2024 Tongji University and Tongji University Press
format Article
author Sun, Wei
Abdullah, Lili Nurliyana
Khalid, Fatimah binti
Sulaiman, Puteri Suhaiza
spellingShingle Sun, Wei
Abdullah, Lili Nurliyana
Khalid, Fatimah binti
Sulaiman, Puteri Suhaiza
Classification of traffic accidents’ factors using TrafficRiskClassifier
author_facet Sun, Wei
Abdullah, Lili Nurliyana
Khalid, Fatimah binti
Sulaiman, Puteri Suhaiza
author_sort Sun, Wei
title Classification of traffic accidents’ factors using TrafficRiskClassifier
title_short Classification of traffic accidents’ factors using TrafficRiskClassifier
title_full Classification of traffic accidents’ factors using TrafficRiskClassifier
title_fullStr Classification of traffic accidents’ factors using TrafficRiskClassifier
title_full_unstemmed Classification of traffic accidents’ factors using TrafficRiskClassifier
title_sort classification of traffic accidents’ factors using trafficriskclassifier
publisher KeAi Communications
publishDate 2024
url http://psasir.upm.edu.my/id/eprint/113360/1/113360.pdf
http://psasir.upm.edu.my/id/eprint/113360/
https://linkinghub.elsevier.com/retrieve/pii/S2046043024000492
_version_ 1817844627267387392
score 13.244413

Similar Items