A deep learning fusion approach to retrieve images of People's unsafe behavior from construction sites
Retrieving unsafe behaviours from an existing digital database can provide managers and the like with the necessary information to put in place strategies to improve safety in construction. Prevailing studies have focused on developing content-based image retrieval (CBIR) approaches (e.g., color-bas...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-12-01
|
| Series: | Developments in the Built Environment |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666165922000199 |
| _version_ | 1818482589322706944 |
|---|---|
| author | Weili Fang Peter E.D. Love Hanbin Luo Shuangjie Xu |
| author_facet | Weili Fang Peter E.D. Love Hanbin Luo Shuangjie Xu |
| author_sort | Weili Fang |
| collection | DOAJ |
| description | Retrieving unsafe behaviours from an existing digital database can provide managers and the like with the necessary information to put in place strategies to improve safety in construction. Prevailing studies have focused on developing content-based image retrieval (CBIR) approaches (e.g., color-based) to retrieve objects and materials obtained from construction sites. While CBIR approaches are effective in extracting low-level features from digital images they are unable to accurately retrieve unsafe behaviours those from existing databases. To address this limitation, we develop an improved CBIR approach to retrieve unsafe behaviour images more accurately and automatically, which combines features extracted from different models. We utilise a digital database developed by Huazhong University of Science and Technology to validate the feasibility of our proposed approach. Our research demonstrates that the fusion of ResNet-101 and VGG-19 can obtain higher levels of Top-K recall and outperform the one feature extraction method. |
| first_indexed | 2024-12-10T11:49:19Z |
| format | Article |
| id | doaj.art-3ca92c3685034224978b69b74188163a |
| institution | Directory Open Access Journal |
| issn | 2666-1659 |
| language | English |
| last_indexed | 2024-12-10T11:49:19Z |
| publishDate | 2022-12-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Developments in the Built Environment |
| spelling | doaj.art-3ca92c3685034224978b69b74188163a2022-12-22T01:49:58ZengElsevierDevelopments in the Built Environment2666-16592022-12-0112100085A deep learning fusion approach to retrieve images of People's unsafe behavior from construction sitesWeili Fang0Peter E.D. Love1Hanbin Luo2Shuangjie Xu3Department of Civil and Building Systems, Technische Universität Berlin, Gustav-Meyer-Allee 25, 13156, Berlin, GermanySchool of Civil and Mechanical Engineering, Curtin University, GPO U1987, Perth, WA, 6845, AustraliaSchool of Civil Engineering and Mechanics, Huazhong University of Science and Technology, Wuhan, Hubei, 430074, China; Corresponding author.Department of Computer Science and Engineering, The Hong Kong University of Science and Technology, Hong Kong, ChinaRetrieving unsafe behaviours from an existing digital database can provide managers and the like with the necessary information to put in place strategies to improve safety in construction. Prevailing studies have focused on developing content-based image retrieval (CBIR) approaches (e.g., color-based) to retrieve objects and materials obtained from construction sites. While CBIR approaches are effective in extracting low-level features from digital images they are unable to accurately retrieve unsafe behaviours those from existing databases. To address this limitation, we develop an improved CBIR approach to retrieve unsafe behaviour images more accurately and automatically, which combines features extracted from different models. We utilise a digital database developed by Huazhong University of Science and Technology to validate the feasibility of our proposed approach. Our research demonstrates that the fusion of ResNet-101 and VGG-19 can obtain higher levels of Top-K recall and outperform the one feature extraction method.http://www.sciencedirect.com/science/article/pii/S2666165922000199Digital imageImage retrievalDeep learningFeature fusionContent-based image retrieval |
| spellingShingle | Weili Fang Peter E.D. Love Hanbin Luo Shuangjie Xu A deep learning fusion approach to retrieve images of People's unsafe behavior from construction sites Developments in the Built Environment Digital image Image retrieval Deep learning Feature fusion Content-based image retrieval |
| title | A deep learning fusion approach to retrieve images of People's unsafe behavior from construction sites |
| title_full | A deep learning fusion approach to retrieve images of People's unsafe behavior from construction sites |
| title_fullStr | A deep learning fusion approach to retrieve images of People's unsafe behavior from construction sites |
| title_full_unstemmed | A deep learning fusion approach to retrieve images of People's unsafe behavior from construction sites |
| title_short | A deep learning fusion approach to retrieve images of People's unsafe behavior from construction sites |
| title_sort | deep learning fusion approach to retrieve images of people s unsafe behavior from construction sites |
| topic | Digital image Image retrieval Deep learning Feature fusion Content-based image retrieval |
| url | http://www.sciencedirect.com/science/article/pii/S2666165922000199 |
| work_keys_str_mv | AT weilifang adeeplearningfusionapproachtoretrieveimagesofpeoplesunsafebehaviorfromconstructionsites AT peteredlove adeeplearningfusionapproachtoretrieveimagesofpeoplesunsafebehaviorfromconstructionsites AT hanbinluo adeeplearningfusionapproachtoretrieveimagesofpeoplesunsafebehaviorfromconstructionsites AT shuangjiexu adeeplearningfusionapproachtoretrieveimagesofpeoplesunsafebehaviorfromconstructionsites AT weilifang deeplearningfusionapproachtoretrieveimagesofpeoplesunsafebehaviorfromconstructionsites AT peteredlove deeplearningfusionapproachtoretrieveimagesofpeoplesunsafebehaviorfromconstructionsites AT hanbinluo deeplearningfusionapproachtoretrieveimagesofpeoplesunsafebehaviorfromconstructionsites AT shuangjiexu deeplearningfusionapproachtoretrieveimagesofpeoplesunsafebehaviorfromconstructionsites |