2017 Fiscal Year Final Research Report
Development of surface crack and internal damage identification system based on non-contact displacement measurements
| Project/Area Number |
15K06197
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Structural engineering/Earthquake engineering/Maintenance management engineering
|
|---|
| Research Institution | Ritsumeikan University |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
日下 貴之 立命館大学, 理工学部, 教授 (10309099)
|
|---|
| Project Period (FY) |
2015-04-01 – 2018-03-31
|
| Keywords | 構造ヘルスモニタリング / き裂検出 / デジタル画像相関法 / 深層学習 |
|---|
| Outline of Final Research Achievements |
This study attempted to develop the crack detection system using non-contact displacement measurements which can identify crack and the damage inside structure quantitatively without requiring baseline information. We succeeded in detecting the invisible crack by evaluating the maximum principal strain and the principal direction from the displacement fields and their time-history. It was demonstrated that the proposed system makes it possible to quantify the crack.
As another attempt in this study, we developed a screening system of cracks, which can be easily installed into UAV or inspection robots, using general object detection technology based on deep learning. We succeeded in detecting the crack instantaneously by using deep convolution neural network while shooting a video. It was demonstrated that the proposed system has potentialities to inspect the space where an inspector is difficult to approach and locate the crack to be quantified from the whole structure on the spot.
|
| Free Research Field |
応用情報学
|
