The Development of a Lane Identification and Assessment Framework for Maintenance Using AI Technology

This study proposes a vision-based framework to support AVs in maintaining stable lane-keeping by assessing the condition of lane markings. Unlike existing infrastructure standards focused on human visibility, this study addresses the need for criteria suited to sensor-based AV environments. Using r...

Full description

Saved in:
Bibliographic Details
Main Authors: Hohyuk Na, Do Gyeong Kim, Ji Min Kang, Chungwon Lee
Format: Article
Language:English
Published: MDPI AG 2025-07-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/13/7410
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849319856658186240
author Hohyuk Na
Do Gyeong Kim
Ji Min Kang
Chungwon Lee
author_facet Hohyuk Na
Do Gyeong Kim
Ji Min Kang
Chungwon Lee
author_sort Hohyuk Na
collection DOAJ
description This study proposes a vision-based framework to support AVs in maintaining stable lane-keeping by assessing the condition of lane markings. Unlike existing infrastructure standards focused on human visibility, this study addresses the need for criteria suited to sensor-based AV environments. Using real driving data from urban expressways in Seoul, a YOLOv5-based lane detection algorithm was developed and enhanced through multi-label annotation and data augmentation. The model achieved a mean average precision (mAP) of 97.4% and demonstrated strong generalization on external datasets such as KITTI and TuSimple. For lane condition assessment, a pixel occupancy–based method was applied, combined with Canny edge detection and morphological operations. A threshold of 80-pixel occupancy was used to classify lanes as intact or worn. The proposed framework reliably detected lane degradation under various road and lighting conditions. These results suggest that quantitative, image-based indicators can complement traditional standards and guide AV-oriented infrastructure policy. Limitations include a lack of adverse weather data and dataset-specific threshold sensitivity.
format Article
id doaj-art-6e5a2856bbf84f75aa65a98d78e73fbd
institution Kabale University
issn 2076-3417
language English
publishDate 2025-07-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-6e5a2856bbf84f75aa65a98d78e73fbd2025-08-20T03:50:17ZengMDPI AGApplied Sciences2076-34172025-07-011513741010.3390/app15137410The Development of a Lane Identification and Assessment Framework for Maintenance Using AI TechnologyHohyuk Na0Do Gyeong Kim1Ji Min Kang2Chungwon Lee3Department of Transportation Engineering, University of Seoul, Seoul 02504, Republic of KoreaDepartment of Transportation Engineering, University of Seoul, Seoul 02504, Republic of KoreaDepartment of Transportation Engineering, University of Seoul, Seoul 02504, Republic of KoreaDepartment of Civil and Environmental Engineering, Seoul National University, Seoul 08826, Republic of KoreaThis study proposes a vision-based framework to support AVs in maintaining stable lane-keeping by assessing the condition of lane markings. Unlike existing infrastructure standards focused on human visibility, this study addresses the need for criteria suited to sensor-based AV environments. Using real driving data from urban expressways in Seoul, a YOLOv5-based lane detection algorithm was developed and enhanced through multi-label annotation and data augmentation. The model achieved a mean average precision (mAP) of 97.4% and demonstrated strong generalization on external datasets such as KITTI and TuSimple. For lane condition assessment, a pixel occupancy–based method was applied, combined with Canny edge detection and morphological operations. A threshold of 80-pixel occupancy was used to classify lanes as intact or worn. The proposed framework reliably detected lane degradation under various road and lighting conditions. These results suggest that quantitative, image-based indicators can complement traditional standards and guide AV-oriented infrastructure policy. Limitations include a lack of adverse weather data and dataset-specific threshold sensitivity.https://www.mdpi.com/2076-3417/15/13/7410maintenance standardslane identification and assessmentlane degradation evaluationpixel occupancyframework
spellingShingle Hohyuk Na
Do Gyeong Kim
Ji Min Kang
Chungwon Lee
The Development of a Lane Identification and Assessment Framework for Maintenance Using AI Technology
Applied Sciences
maintenance standards
lane identification and assessment
lane degradation evaluation
pixel occupancy
framework
title The Development of a Lane Identification and Assessment Framework for Maintenance Using AI Technology
title_full The Development of a Lane Identification and Assessment Framework for Maintenance Using AI Technology
title_fullStr The Development of a Lane Identification and Assessment Framework for Maintenance Using AI Technology
title_full_unstemmed The Development of a Lane Identification and Assessment Framework for Maintenance Using AI Technology
title_short The Development of a Lane Identification and Assessment Framework for Maintenance Using AI Technology
title_sort development of a lane identification and assessment framework for maintenance using ai technology
topic maintenance standards
lane identification and assessment
lane degradation evaluation
pixel occupancy
framework
url https://www.mdpi.com/2076-3417/15/13/7410
work_keys_str_mv AT hohyukna thedevelopmentofalaneidentificationandassessmentframeworkformaintenanceusingaitechnology
AT dogyeongkim thedevelopmentofalaneidentificationandassessmentframeworkformaintenanceusingaitechnology
AT jiminkang thedevelopmentofalaneidentificationandassessmentframeworkformaintenanceusingaitechnology
AT chungwonlee thedevelopmentofalaneidentificationandassessmentframeworkformaintenanceusingaitechnology
AT hohyukna developmentofalaneidentificationandassessmentframeworkformaintenanceusingaitechnology
AT dogyeongkim developmentofalaneidentificationandassessmentframeworkformaintenanceusingaitechnology
AT jiminkang developmentofalaneidentificationandassessmentframeworkformaintenanceusingaitechnology
AT chungwonlee developmentofalaneidentificationandassessmentframeworkformaintenanceusingaitechnology