anomaly detection of bridge by use of image devices

2022 Fiscal Year Final Research Report

• Project/Area Number: 21K20448
• Research Category: Grant-in-Aid for Research Activity Start-up
• Allocation Type: Multi-year Fund
• Review Section: 0303:Civil engineering, social systems engineering, safety engineering, disaster prevention engineering, and related fields
• Research Institution: Kyoto University
• Principal Investigator: kawabe daigo 京都大学, 地球環境学堂, 研究員 (30907522)
• Project Period (FY): 2021-08-30 – 2023-03-31
• Keywords: 構造ヘルスモニタリング / 維持管理 / 画像処理 / 異常検知 / 振動 / 柱状構造物

Outline of Final Research Achievements

This study aims to detect an anomalous of structure by use of changing in modal parameters obtained from video footage. Extracting dynamic motion of the target structure within the angle of view of the video and converting it to the modal information were considered. This study achieved to obtain the motion of the structure accurately as displacement waves by use of KLT (Kanade-Lucas Tomasi) Tracker that is one of the Optical Flow method. Furthermore, the results of verifying the change in the estimated natural frequency before and after the introduction of damage using this waveform showed that the change was equivalent to the change in the natural frequency measured by high performance accelerometers.

Free Research Field

維持管理工学

Academic Significance and Societal Importance of the Research Achievements

動画情報から構造物の異常を検出する手法は研究が進められてきたが，対象物にマーキングを必要としない手間を要さない手法は限られる．また構造物の損傷による動的応答の変化は比較的小さいため，それらを動画情報から検知することは容易ではなく，成果の学術的意義は大きい．また，少子高齢化と社会インフラの劣化が社会課題となる中，人的労力を要さず，かつ経済性に富む異常検知手法を構築することは社会的意義が大きい．