A Few-Shot Defect Detection Method for Transmission Lines Based on Meta-Attention and Feature Reconstruction
In tasks of transmission line defect detection, traditional object detection algorithms are ineffective, with few training samples of defective components. Meta-learning uses multi-task learning as well as fine-tuning to learn common features in different tasks, which has the ability to adapt to new...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-05-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/13/10/5896 |
| _version_ | 1797601267422330880 |
|---|---|
| author | Yundong Shi Huimin Wang Chao Jing Xingzhong Zhang |
| author_facet | Yundong Shi Huimin Wang Chao Jing Xingzhong Zhang |
| author_sort | Yundong Shi |
| collection | DOAJ |
| description | In tasks of transmission line defect detection, traditional object detection algorithms are ineffective, with few training samples of defective components. Meta-learning uses multi-task learning as well as fine-tuning to learn common features in different tasks, which has the ability to adapt to new tasks quickly, shows good performance in few-shot object detection, and has good generalization in new tasks. For this reason, we proposed a few-shot defect detection method (Meta PowerNet) with a Meta-attention RPN and Feature Reconstruction Module for transmission lines based on meta-learning. First, in the stage of region proposal, a new region proposal network (Meta-Attention Region Proposal Network, MA-RPN) is designed to fuse the support set features and the query set features to filter the noise in anchor boxes. In addition, it has the ability to focus on the subtle texture features of smaller-sized objects by fusing low-level features from the query set. Second, in the meta-feature construction stage, we designed a meta-learner with the defect feature reconstruction module as the core to capture and focus on the defect-related feature channels. The experimental results show that under the condition, there are only 30 training objects for various types of component defects. The method achieves 72.5% detection accuracy for component defects, which is a significant improvement compared with other mainstream few-shot object detection. Meanwhile, the MA-RPN designed in this paper can be used in other meta-learning object detection models universally. |
| first_indexed | 2024-03-11T03:58:45Z |
| format | Article |
| id | doaj.art-844113a13dca43b081178e9ede78ccb6 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-11T03:58:45Z |
| publishDate | 2023-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-844113a13dca43b081178e9ede78ccb62023-11-18T00:17:21ZengMDPI AGApplied Sciences2076-34172023-05-011310589610.3390/app13105896A Few-Shot Defect Detection Method for Transmission Lines Based on Meta-Attention and Feature ReconstructionYundong Shi0Huimin Wang1Chao Jing2Xingzhong Zhang3College of Software, Taiyuan University of Technology, Taiyuan 030024, ChinaShanxi Energy Internet Research Institute, Taiyuan 030024, ChinaShanxi Energy Internet Research Institute, Taiyuan 030024, ChinaCollege of Software, Taiyuan University of Technology, Taiyuan 030024, ChinaIn tasks of transmission line defect detection, traditional object detection algorithms are ineffective, with few training samples of defective components. Meta-learning uses multi-task learning as well as fine-tuning to learn common features in different tasks, which has the ability to adapt to new tasks quickly, shows good performance in few-shot object detection, and has good generalization in new tasks. For this reason, we proposed a few-shot defect detection method (Meta PowerNet) with a Meta-attention RPN and Feature Reconstruction Module for transmission lines based on meta-learning. First, in the stage of region proposal, a new region proposal network (Meta-Attention Region Proposal Network, MA-RPN) is designed to fuse the support set features and the query set features to filter the noise in anchor boxes. In addition, it has the ability to focus on the subtle texture features of smaller-sized objects by fusing low-level features from the query set. Second, in the meta-feature construction stage, we designed a meta-learner with the defect feature reconstruction module as the core to capture and focus on the defect-related feature channels. The experimental results show that under the condition, there are only 30 training objects for various types of component defects. The method achieves 72.5% detection accuracy for component defects, which is a significant improvement compared with other mainstream few-shot object detection. Meanwhile, the MA-RPN designed in this paper can be used in other meta-learning object detection models universally.https://www.mdpi.com/2076-3417/13/10/5896few-shot learningmeta-attentiontransmission linesdefects detectionfeature regeneration |
| spellingShingle | Yundong Shi Huimin Wang Chao Jing Xingzhong Zhang A Few-Shot Defect Detection Method for Transmission Lines Based on Meta-Attention and Feature Reconstruction Applied Sciences few-shot learning meta-attention transmission lines defects detection feature regeneration |
| title | A Few-Shot Defect Detection Method for Transmission Lines Based on Meta-Attention and Feature Reconstruction |
| title_full | A Few-Shot Defect Detection Method for Transmission Lines Based on Meta-Attention and Feature Reconstruction |
| title_fullStr | A Few-Shot Defect Detection Method for Transmission Lines Based on Meta-Attention and Feature Reconstruction |
| title_full_unstemmed | A Few-Shot Defect Detection Method for Transmission Lines Based on Meta-Attention and Feature Reconstruction |
| title_short | A Few-Shot Defect Detection Method for Transmission Lines Based on Meta-Attention and Feature Reconstruction |
| title_sort | few shot defect detection method for transmission lines based on meta attention and feature reconstruction |
| topic | few-shot learning meta-attention transmission lines defects detection feature regeneration |
| url | https://www.mdpi.com/2076-3417/13/10/5896 |
| work_keys_str_mv | AT yundongshi afewshotdefectdetectionmethodfortransmissionlinesbasedonmetaattentionandfeaturereconstruction AT huiminwang afewshotdefectdetectionmethodfortransmissionlinesbasedonmetaattentionandfeaturereconstruction AT chaojing afewshotdefectdetectionmethodfortransmissionlinesbasedonmetaattentionandfeaturereconstruction AT xingzhongzhang afewshotdefectdetectionmethodfortransmissionlinesbasedonmetaattentionandfeaturereconstruction AT yundongshi fewshotdefectdetectionmethodfortransmissionlinesbasedonmetaattentionandfeaturereconstruction AT huiminwang fewshotdefectdetectionmethodfortransmissionlinesbasedonmetaattentionandfeaturereconstruction AT chaojing fewshotdefectdetectionmethodfortransmissionlinesbasedonmetaattentionandfeaturereconstruction AT xingzhongzhang fewshotdefectdetectionmethodfortransmissionlinesbasedonmetaattentionandfeaturereconstruction |