An automatic and unsupervised image mask acquisition method based on generative adversarial networks
This paper proposes an unsupervised method for automatically labelling and obtaining image masks in defect detection. Since it is very labour intensive to acquire the image masks needed for deep learning (e.g. in semantic segmentation tasks) via manual labelling, we propose a method that utilize a g...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2024-12-01
|
| Series: | Systems Science & Control Engineering |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/21642583.2023.2300835 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850063731634995200 |
|---|---|
| author | Hao Wu Yulong Liu Jiankang Yang |
| author_facet | Hao Wu Yulong Liu Jiankang Yang |
| author_sort | Hao Wu |
| collection | DOAJ |
| description | This paper proposes an unsupervised method for automatically labelling and obtaining image masks in defect detection. Since it is very labour intensive to acquire the image masks needed for deep learning (e.g. in semantic segmentation tasks) via manual labelling, we propose a method that utilize a generative adversarial network to obtain image masks automatically. Using this method, it is only necessary to input a considerable defect-free image to train. Then the proposed method can generate defect-free image samples like the input defect images, and the defect's location can be determined by comparing the input sample image containing defects with the generated sample image, thereby obtaining the input image mask. Our proposed method has been validated through experimental results, demonstrating its effectiveness. In addition to automatically and obtaining the required masks, our method achieved greater detection accuracy using the deep learning model Mask R-CNN compared with the manual labelling supervised method and a semi-supervised method. |
| format | Article |
| id | doaj-art-7e1bd8aa12ce42a7bcd20353e3e0f822 |
| institution | DOAJ |
| issn | 2164-2583 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | Systems Science & Control Engineering |
| spelling | doaj-art-7e1bd8aa12ce42a7bcd20353e3e0f8222025-08-20T02:49:30ZengTaylor & Francis GroupSystems Science & Control Engineering2164-25832024-12-0112110.1080/21642583.2023.2300835An automatic and unsupervised image mask acquisition method based on generative adversarial networksHao Wu0Yulong Liu1Jiankang Yang2School of Mechanical Engineering, Anhui University of Technology, Maanshan, People’s Republic of ChinaSchool of Mechanical Engineering, Anhui University of Technology, Maanshan, People’s Republic of ChinaSchool of Mechanical Engineering, Anhui University of Technology, Maanshan, People’s Republic of ChinaThis paper proposes an unsupervised method for automatically labelling and obtaining image masks in defect detection. Since it is very labour intensive to acquire the image masks needed for deep learning (e.g. in semantic segmentation tasks) via manual labelling, we propose a method that utilize a generative adversarial network to obtain image masks automatically. Using this method, it is only necessary to input a considerable defect-free image to train. Then the proposed method can generate defect-free image samples like the input defect images, and the defect's location can be determined by comparing the input sample image containing defects with the generated sample image, thereby obtaining the input image mask. Our proposed method has been validated through experimental results, demonstrating its effectiveness. In addition to automatically and obtaining the required masks, our method achieved greater detection accuracy using the deep learning model Mask R-CNN compared with the manual labelling supervised method and a semi-supervised method.https://www.tandfonline.com/doi/10.1080/21642583.2023.2300835Deep learningdefect detectionMask R-CNN |
| spellingShingle | Hao Wu Yulong Liu Jiankang Yang An automatic and unsupervised image mask acquisition method based on generative adversarial networks Systems Science & Control Engineering Deep learning defect detection Mask R-CNN |
| title | An automatic and unsupervised image mask acquisition method based on generative adversarial networks |
| title_full | An automatic and unsupervised image mask acquisition method based on generative adversarial networks |
| title_fullStr | An automatic and unsupervised image mask acquisition method based on generative adversarial networks |
| title_full_unstemmed | An automatic and unsupervised image mask acquisition method based on generative adversarial networks |
| title_short | An automatic and unsupervised image mask acquisition method based on generative adversarial networks |
| title_sort | automatic and unsupervised image mask acquisition method based on generative adversarial networks |
| topic | Deep learning defect detection Mask R-CNN |
| url | https://www.tandfonline.com/doi/10.1080/21642583.2023.2300835 |
| work_keys_str_mv | AT haowu anautomaticandunsupervisedimagemaskacquisitionmethodbasedongenerativeadversarialnetworks AT yulongliu anautomaticandunsupervisedimagemaskacquisitionmethodbasedongenerativeadversarialnetworks AT jiankangyang anautomaticandunsupervisedimagemaskacquisitionmethodbasedongenerativeadversarialnetworks AT haowu automaticandunsupervisedimagemaskacquisitionmethodbasedongenerativeadversarialnetworks AT yulongliu automaticandunsupervisedimagemaskacquisitionmethodbasedongenerativeadversarialnetworks AT jiankangyang automaticandunsupervisedimagemaskacquisitionmethodbasedongenerativeadversarialnetworks |