2022 Fiscal Year Research-status Report
Domain adaptation for acoustic inspection of concrete structures
| Project/Area Number |
21K17829
|
|---|
| Research Institution | The University of Tokyo |
|---|
Principal Investigator |
ルイ笠原 純ユネス 東京大学, 大学院工学系研究科(工学部), 特任講師 (20885412)
|
|---|
| Project Period (FY) |
2021-04-01 – 2024-03-31
|
| Keywords | Inspection / Domain adaptation / Clustering / Weak supervision |
|---|
| Outline of Annual Research Achievements |
This research aims at the automation of the hammering test for the inspection of concrete structures. Supervised learning approaches have the
issue that when the concrete structure differs between training and deployment of the system, the performance is degraded (domain gap).
This year, efforts have been shifted towards approaches based on Deep Learning for outlier detection, namely Autoencoder-type approaches, which have been gaining attention in the field. The potential of combining such methods with the weak supervision framework are being currently investigated.
|
| Current Status of Research Progress |
Current Status of Research Progress
3: Progress in research has been slightly delayed.
Reason
Day-to-day operations were still greatly affected by restrictions related to COVID-19. Additionally, difficulties were encountered when attempting to
purchase sensors due to the shortage of electronic components.
|
| Strategy for Future Research Activity |
Several pipelines for audio processing which Autoencoders have been established this year. In the next fiscal year, it is planned to expand those to include weak supervision and start tackling the domain adaptation issue.
|
| Causes of Carryover |
Delay in research activities due to the pandemic and personal reasons. Remaining budget planned to be used for speeding up efforts in the next fiscal year.
|
