A vision‐based weigh‐in‐motion approach for vehicle load tracking and identification DOI Creative Commons

Phat Tai Lam,

Jaehyuk Lee, Yun-Woo Lee

et al.

Computer-Aided Civil and Infrastructure Engineering, Journal Year: 2025, Volume and Issue: unknown

Published: March 16, 2025

Abstract With the rapid increase in number of vehicles, accurately identifying vehicle loads is crucial for maintaining and operating transportation infrastructure systems. Existing load identification methods typically rely on collecting data from weigh‐in‐motion (WIM) systems when vehicles pass over them. However, cumbersome installation, high costs, regular maintenance are main obstacles that prevent WIM being widely used practice. This study introduces visual (V‐WIM) framework, a vision‐based approach tracking moving loads. The V‐WIM framework consists two components, weight estimation location estimation. Vehicle estimated using tire deformation parameters extracted images through object detection optical character recognition techniques. A deep learning‐based YOLOv8 algorithm employed as detector, combined with ByteTrack location. its corresponding then integrated to enable simultaneous tracking. performance proposed was evaluated component validation tests one on‐site test, demonstrating capability overcome limitations existing methods.

Language: Английский

Automated digital transformation for pedestrian suspension bridges using hybrid semantic structure from motion DOI

Yeongseo Park,

Jaehyuk Lee, Kevin Han

et al.

Automation in Construction, Journal Year: 2025, Volume and Issue: 176, P. 106232 - 106232

Published: May 9, 2025

Language: Английский

Citations

0
A vision‐based weigh‐in‐motion approach for vehicle load tracking and identification DOI Creative Commons

Phat Tai Lam,

Jaehyuk Lee, Yun-Woo Lee

et al.

Computer-Aided Civil and Infrastructure Engineering, Journal Year: 2025, Volume and Issue: unknown

Published: March 16, 2025

Abstract With the rapid increase in number of vehicles, accurately identifying vehicle loads is crucial for maintaining and operating transportation infrastructure systems. Existing load identification methods typically rely on collecting data from weigh‐in‐motion (WIM) systems when vehicles pass over them. However, cumbersome installation, high costs, regular maintenance are main obstacles that prevent WIM being widely used practice. This study introduces visual (V‐WIM) framework, a vision‐based approach tracking moving loads. The V‐WIM framework consists two components, weight estimation location estimation. Vehicle estimated using tire deformation parameters extracted images through object detection optical character recognition techniques. A deep learning‐based YOLOv8 algorithm employed as detector, combined with ByteTrack location. its corresponding then integrated to enable simultaneous tracking. performance proposed was evaluated component validation tests one on‐site test, demonstrating capability overcome limitations existing methods.

Language: Английский

Citations

0