Distress Image Retrieval for Infrastructure Maintenance via Self-Trained Deep Metric Learning Using Experts’ Knowledge
Distress image retrieval for infrastructure maintenance via self-trained deep metric learning using experts’ knowledge is proposed in this paper. Since engineers take multiple images of a single distress part for inspection of road structures, it is necessary to construct a similar distre...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9406804/ |
| _version_ | 1819089698338897920 |
|---|---|
| author | Naoki Ogawa Keisuke Maeda Takahiro Ogawa Miki Haseyama |
| author_facet | Naoki Ogawa Keisuke Maeda Takahiro Ogawa Miki Haseyama |
| author_sort | Naoki Ogawa |
| collection | DOAJ |
| description | Distress image retrieval for infrastructure maintenance via self-trained deep metric learning using experts’ knowledge is proposed in this paper. Since engineers take multiple images of a single distress part for inspection of road structures, it is necessary to construct a similar distress image retrieval method considering the input of multiple images to support determination of the level of deterioration. Thus, the construction of an image retrieval method while selecting an effective input from multiple images is described in this paper. The proposed method performs deep metric learning by using a small number of effective images labeled by experts’ knowledge with information about their effectiveness and a large number of unlabeled images via a self-training approach. Specifically, an end-to-end learning approach that performs retraining of the model by assigning pseudo-labels to these unlabeled images according to the output confidence of the model is achieved. Thus, the proposed method can select an effective image from multiple images that are input at the retrieval as a query image. This is the main contribution of this paper. As a result, the proposed method realizes highly accurate retrieval of similar distress images considering the actual situation of inspection in which multiple images of a distress part are input. |
| first_indexed | 2024-12-21T22:12:04Z |
| format | Article |
| id | doaj.art-94594a4c72c04dfcb4b7eee0421753bf |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-21T22:12:04Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-94594a4c72c04dfcb4b7eee0421753bf2022-12-21T18:48:33ZengIEEEIEEE Access2169-35362021-01-019652346524510.1109/ACCESS.2021.30740199406804Distress Image Retrieval for Infrastructure Maintenance via Self-Trained Deep Metric Learning Using Experts’ KnowledgeNaoki Ogawa0https://orcid.org/0000-0002-3884-7325Keisuke Maeda1https://orcid.org/0000-0001-8039-3462Takahiro Ogawa2https://orcid.org/0000-0001-5332-8112Miki Haseyama3https://orcid.org/0000-0003-1496-1761Graduate School of Information Science and Technology, Hokkaido University, Sapporo, JapanOffice of Institutional Research, Hokkaido University, Sapporo, JapanFaculty of Information Science and Technology, Hokkaido University, Sapporo, JapanFaculty of Information Science and Technology, Hokkaido University, Sapporo, JapanDistress image retrieval for infrastructure maintenance via self-trained deep metric learning using experts’ knowledge is proposed in this paper. Since engineers take multiple images of a single distress part for inspection of road structures, it is necessary to construct a similar distress image retrieval method considering the input of multiple images to support determination of the level of deterioration. Thus, the construction of an image retrieval method while selecting an effective input from multiple images is described in this paper. The proposed method performs deep metric learning by using a small number of effective images labeled by experts’ knowledge with information about their effectiveness and a large number of unlabeled images via a self-training approach. Specifically, an end-to-end learning approach that performs retraining of the model by assigning pseudo-labels to these unlabeled images according to the output confidence of the model is achieved. Thus, the proposed method can select an effective image from multiple images that are input at the retrieval as a query image. This is the main contribution of this paper. As a result, the proposed method realizes highly accurate retrieval of similar distress images considering the actual situation of inspection in which multiple images of a distress part are input.https://ieeexplore.ieee.org/document/9406804/Distress image retrievalself-trained approachpseudo-labeldeep metric learning |
| spellingShingle | Naoki Ogawa Keisuke Maeda Takahiro Ogawa Miki Haseyama Distress Image Retrieval for Infrastructure Maintenance via Self-Trained Deep Metric Learning Using Experts’ Knowledge IEEE Access Distress image retrieval self-trained approach pseudo-label deep metric learning |
| title | Distress Image Retrieval for Infrastructure Maintenance via Self-Trained Deep Metric Learning Using Experts’ Knowledge |
| title_full | Distress Image Retrieval for Infrastructure Maintenance via Self-Trained Deep Metric Learning Using Experts’ Knowledge |
| title_fullStr | Distress Image Retrieval for Infrastructure Maintenance via Self-Trained Deep Metric Learning Using Experts’ Knowledge |
| title_full_unstemmed | Distress Image Retrieval for Infrastructure Maintenance via Self-Trained Deep Metric Learning Using Experts’ Knowledge |
| title_short | Distress Image Retrieval for Infrastructure Maintenance via Self-Trained Deep Metric Learning Using Experts’ Knowledge |
| title_sort | distress image retrieval for infrastructure maintenance via self trained deep metric learning using experts x2019 knowledge |
| topic | Distress image retrieval self-trained approach pseudo-label deep metric learning |
| url | https://ieeexplore.ieee.org/document/9406804/ |
| work_keys_str_mv | AT naokiogawa distressimageretrievalforinfrastructuremaintenanceviaselftraineddeepmetriclearningusingexpertsx2019knowledge AT keisukemaeda distressimageretrievalforinfrastructuremaintenanceviaselftraineddeepmetriclearningusingexpertsx2019knowledge AT takahiroogawa distressimageretrievalforinfrastructuremaintenanceviaselftraineddeepmetriclearningusingexpertsx2019knowledge AT mikihaseyama distressimageretrievalforinfrastructuremaintenanceviaselftraineddeepmetriclearningusingexpertsx2019knowledge |