A Novel MoCo-Based Self-Supervised Learning Framework for Solar Panel Defect Detection
Defect detection in solar panels remains constrained by the limitations of manual labeling and the inefficiency of traditional inspection methods, which often struggle with large, high-resolution imagery. This study presents a novel self-supervised learning approach using the Momentum Contrast (MoCo...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10840178/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1825207032246960128 |
|---|---|
| author | Jun Huang Shamsul Arrieya Ariffin Yongqiang Chen Jinghui Lin Wanting Xu |
| author_facet | Jun Huang Shamsul Arrieya Ariffin Yongqiang Chen Jinghui Lin Wanting Xu |
| author_sort | Jun Huang |
| collection | DOAJ |
| description | Defect detection in solar panels remains constrained by the limitations of manual labeling and the inefficiency of traditional inspection methods, which often struggle with large, high-resolution imagery. This study presents a novel self-supervised learning approach using the Momentum Contrast (MoCo) framework to address these challenges without relying on annotated datasets. Leveraging MoCo’s robust feature extraction and K-Nearest Neighbors (KNN) clustering, our method achieves accurate defect identification, bypassing the dependency on labeled data. Evaluated on the ELPV dataset, our approach attained a notable 96.95% accuracy, demonstrating significant improvement over existing unsupervised methods like KDAD, SAOE, DRA, and BGAD-FAS, and even outperforming some supervised models such as Adapted VGG19, Adapted VGG16, and ShuffleNet. Additionally, our model achieved 99.44% accuracy on the EL dataset, underscoring its adaptability and robustness across different environments. This framework offers a scalable, automated solution for image-level defect detection, poised to enhance efficiency and reduce manual intervention in industrial applications. |
| format | Article |
| id | doaj-art-933301c1215c4cfba92419aa8f7c1d06 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-933301c1215c4cfba92419aa8f7c1d062025-02-07T00:01:26ZengIEEEIEEE Access2169-35362025-01-0113229772298810.1109/ACCESS.2025.352970110840178A Novel MoCo-Based Self-Supervised Learning Framework for Solar Panel Defect DetectionJun Huang0https://orcid.org/0009-0006-8549-4098Shamsul Arrieya Ariffin1https://orcid.org/0000-0001-6266-6797Yongqiang Chen2Jinghui Lin3Wanting Xu4https://orcid.org/0000-0001-7236-176XFaculty of Information Technology, City University Malaysia, Petaling Jaya, Kuala Lumpur, MalaysiaFaculty of Information Technology, City University Malaysia, Petaling Jaya, Kuala Lumpur, MalaysiaFaculty of Intelligent Manufacturing, Wuhu Institute of Technology, Wuhu, Anhui, ChinaFaculty of Intelligent Manufacturing, Wuhu Institute of Technology, Wuhu, Anhui, ChinaFaculty of Information Technology, City University Malaysia, Petaling Jaya, Kuala Lumpur, MalaysiaDefect detection in solar panels remains constrained by the limitations of manual labeling and the inefficiency of traditional inspection methods, which often struggle with large, high-resolution imagery. This study presents a novel self-supervised learning approach using the Momentum Contrast (MoCo) framework to address these challenges without relying on annotated datasets. Leveraging MoCo’s robust feature extraction and K-Nearest Neighbors (KNN) clustering, our method achieves accurate defect identification, bypassing the dependency on labeled data. Evaluated on the ELPV dataset, our approach attained a notable 96.95% accuracy, demonstrating significant improvement over existing unsupervised methods like KDAD, SAOE, DRA, and BGAD-FAS, and even outperforming some supervised models such as Adapted VGG19, Adapted VGG16, and ShuffleNet. Additionally, our model achieved 99.44% accuracy on the EL dataset, underscoring its adaptability and robustness across different environments. This framework offers a scalable, automated solution for image-level defect detection, poised to enhance efficiency and reduce manual intervention in industrial applications.https://ieeexplore.ieee.org/document/10840178/Solar panel defect detectionself-supervised learningMoCo |
| spellingShingle | Jun Huang Shamsul Arrieya Ariffin Yongqiang Chen Jinghui Lin Wanting Xu A Novel MoCo-Based Self-Supervised Learning Framework for Solar Panel Defect Detection IEEE Access Solar panel defect detection self-supervised learning MoCo |
| title | A Novel MoCo-Based Self-Supervised Learning Framework for Solar Panel Defect Detection |
| title_full | A Novel MoCo-Based Self-Supervised Learning Framework for Solar Panel Defect Detection |
| title_fullStr | A Novel MoCo-Based Self-Supervised Learning Framework for Solar Panel Defect Detection |
| title_full_unstemmed | A Novel MoCo-Based Self-Supervised Learning Framework for Solar Panel Defect Detection |
| title_short | A Novel MoCo-Based Self-Supervised Learning Framework for Solar Panel Defect Detection |
| title_sort | novel moco based self supervised learning framework for solar panel defect detection |
| topic | Solar panel defect detection self-supervised learning MoCo |
| url | https://ieeexplore.ieee.org/document/10840178/ |
| work_keys_str_mv | AT junhuang anovelmocobasedselfsupervisedlearningframeworkforsolarpaneldefectdetection AT shamsularrieyaariffin anovelmocobasedselfsupervisedlearningframeworkforsolarpaneldefectdetection AT yongqiangchen anovelmocobasedselfsupervisedlearningframeworkforsolarpaneldefectdetection AT jinghuilin anovelmocobasedselfsupervisedlearningframeworkforsolarpaneldefectdetection AT wantingxu anovelmocobasedselfsupervisedlearningframeworkforsolarpaneldefectdetection AT junhuang novelmocobasedselfsupervisedlearningframeworkforsolarpaneldefectdetection AT shamsularrieyaariffin novelmocobasedselfsupervisedlearningframeworkforsolarpaneldefectdetection AT yongqiangchen novelmocobasedselfsupervisedlearningframeworkforsolarpaneldefectdetection AT jinghuilin novelmocobasedselfsupervisedlearningframeworkforsolarpaneldefectdetection AT wantingxu novelmocobasedselfsupervisedlearningframeworkforsolarpaneldefectdetection |