Road Event Detection and Situation Classification Framework

The limitations of the current freeway traffic information collection system are being evaluated, and a more optimal system that can meet the needs of the future traffic environment is being sought. To address these system and service needs, we propose the development of an on-board based incident d...

Full description

Saved in:
Bibliographic Details
Published inJournal of the Eastern Asia Society for Transportation Studies Vol. 15; pp. 2305 - 2314
Main Authors LEE, Soomok, LEE, Sanghyun, KIM, Jinyoung, NOH, Jongmin, LEE, Hwanpil, YUN, Ilsoo
Format Journal Article
LanguageEnglish
Published Eastern Asia Society for Transportation Studies 2024
Subjects
Algorithm
Object Detection
Road Event
Scene Classification
YOLO
Online AccessGet full text

Cover

More Information
Summary:The limitations of the current freeway traffic information collection system are being evaluated, and a more optimal system that can meet the needs of the future traffic environment is being sought. To address these system and service needs, we propose the development of an on-board based incident detection system using artificial intelligence, to improve the efficiency of traffic management on highways and respond to emergencies more quickly. The developed system uses in-vehicle cameras mounted on expressway service patrol vehicles to detect incidents, and includes a detection algorithm and a dataset specifically designed for this purpose. We demonstrate the feasibility of our algorithm using a dataset obtained from Korean expressways. Our main contributions include the development of a comprehensive system for detecting emergencies on highways, the creation of a dataset and detection algorithm specifically designed for this task, and the successful demonstration of the algorithm using real-world data.
ISSN:1881-1124
DOI:10.11175/easts.15.2305