Structural Damage Propagation Evaluation Technology Based on Image Analysis of Vehicle-mounted Camera and Ground Penetrating Radar and Vibration Monitoring

2023 Fiscal Year Final Research Report

• Project/Area Number: 21K04240
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 22020:Structure engineering and earthquake engineering-related
• Research Institution: Ritsumeikan University
• Principal Investigator: Nomura Yasutoshi 立命館大学, 理工学部, 教授 (20372667)
• Co-Investigator(Kenkyū-buntansha): 日下 貴之 立命館大学, 理工学部, 教授 (10309099)
• Project Period (FY): 2021-04-01 – 2024-03-31
• Keywords: 深層学習 / 物体検出 / 画像処理 / ひび割れ / 腐食 / インフラ維持管理 / 地中レーダ

Outline of Final Research Achievements

The purpose of this study is to develop a comprehensive system for diagnosing the structural condition of monorail and highway bridges through patrol vehicle-based inspection, patrol inspection, and vibration monitoring. Currently, when evaluating the progression of damage on roads, necessary images are manually searched from continuously captured images over time, and comparisons are made manually. This also applies to cases where a ground-penetrating radar device is mounted on inspection vehicles to examine the internal state of the road slabs. In this study, we aim to automatically evaluate surface corrosion, cracks, and the internal condition of hollow slab structures and their aging changes by applying the latest technologies such as deep learning and image processing to the footage from inspection vehicle cameras and ground-penetrating radar data.

Free Research Field

応用情報学

Academic Significance and Societal Importance of the Research Achievements

本研究では，点検車載カメラ映像および地中レーダデータに対して，深層学習に基づく物体検出技術，画像レジストレーションや加色混合法など画像処理技術を適用することで，モノレール橋梁の構造表面腐食・ひび割れおよびコンクリート中空床版内部のボイドを自動的に検出するだけでなく，これまで人手によるところの大きかった腐食・ひび割れの進展およびコンクリート中空床版のかぶり厚を定量的に評価することに成功した．本研究の成果は，構造物の維持管理における点検の精度と効率性を大幅に向上させ，労働者人口が減少する中で，インフラの健全性を持続的に保つために重要な社会的意義を有すると考える．