MRA-YOLOv8: A Network Enhancing Feature Extraction Ability for Photovoltaic Cell Defects

To address the challenges posed by complex backgrounds and the low occurrence in photovoltaic cell images captured by industrial sensors, we propose a novel defect detection method: MRA-YOLOv8. First, a multi-branch coordinate attention network (MBCANet) is introduced into the backbone. The coordina...

Full description

Saved in:
Bibliographic Details
Main Authors: Nannan Wang, Siqi Huang, Xiangpeng Liu, Zhining Wang, Yi Liu, Zhe Gao
Format: Article
Language:English
Published: MDPI AG 2025-03-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/5/1542
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850053056315523072
author Nannan Wang
Siqi Huang
Xiangpeng Liu
Zhining Wang
Yi Liu
Zhe Gao
author_facet Nannan Wang
Siqi Huang
Xiangpeng Liu
Zhining Wang
Yi Liu
Zhe Gao
author_sort Nannan Wang
collection DOAJ
description To address the challenges posed by complex backgrounds and the low occurrence in photovoltaic cell images captured by industrial sensors, we propose a novel defect detection method: MRA-YOLOv8. First, a multi-branch coordinate attention network (MBCANet) is introduced into the backbone. The coordinate attention network (CANet) is incorporated to mitigate the noise impact of background information on the detection task, and multiple branches are employed to enhance the model’s feature extraction capability. Second, we integrate a multi-path feature extraction module, ResBlock, into the neck. This module provides finer-grained multi-scale features, improving feature extraction from complex backgrounds and enhancing the model’s robustness. Finally, we implement alpha-minimum point distance-based IoU (AMPDIoU) to the head. This loss function enhances the accuracy and robustness of small object detection by integrating minimum point distance-based IoU (MPDIoU) and Alpha-IoU methods. The results demonstrate that MRA-YOLOv8 outperforms other mainstream methods in detection performance. On the photovoltaic electroluminescence anomaly detection (PVEL-AD) dataset, the proposed method achieves a <i>mAP</i><sub>50</sub> of 91.7%, representing an improvement of 3.1% over YOLOv8 and 16.1% over detection transformer (DETR). On the SPDI dataset, our method achieves a <i>mAP</i><sub>50</sub> of 69.3%, showing a 2.1% improvement over YOLOv8 and a 6.6% improvement over DETR. The proposed MRA-YOLOv8 also exhibits great deployment potential. It can be effectively integrated with drone-based inspection systems, allowing for efficient and accurate PV plant inspections. Moreover, to tackle the issue of data imbalance, we propose generating synthetic defect data via generative adversarial networks (GANs), which can supplement the limited defect samples and improve the model’s generalization ability.
format Article
id doaj-art-00735f5bf6794c22bbb023091e3fbf4e
institution DOAJ
issn 1424-8220
language English
publishDate 2025-03-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-00735f5bf6794c22bbb023091e3fbf4e2025-08-20T02:52:38ZengMDPI AGSensors1424-82202025-03-01255154210.3390/s25051542MRA-YOLOv8: A Network Enhancing Feature Extraction Ability for Photovoltaic Cell DefectsNannan Wang0Siqi Huang1Xiangpeng Liu2Zhining Wang3Yi Liu4Zhe Gao5College of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 201418, ChinaCollege of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 201418, ChinaCollege of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 201418, ChinaCollege of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 201418, ChinaCollege of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 201418, ChinaCollege of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 201418, ChinaTo address the challenges posed by complex backgrounds and the low occurrence in photovoltaic cell images captured by industrial sensors, we propose a novel defect detection method: MRA-YOLOv8. First, a multi-branch coordinate attention network (MBCANet) is introduced into the backbone. The coordinate attention network (CANet) is incorporated to mitigate the noise impact of background information on the detection task, and multiple branches are employed to enhance the model’s feature extraction capability. Second, we integrate a multi-path feature extraction module, ResBlock, into the neck. This module provides finer-grained multi-scale features, improving feature extraction from complex backgrounds and enhancing the model’s robustness. Finally, we implement alpha-minimum point distance-based IoU (AMPDIoU) to the head. This loss function enhances the accuracy and robustness of small object detection by integrating minimum point distance-based IoU (MPDIoU) and Alpha-IoU methods. The results demonstrate that MRA-YOLOv8 outperforms other mainstream methods in detection performance. On the photovoltaic electroluminescence anomaly detection (PVEL-AD) dataset, the proposed method achieves a <i>mAP</i><sub>50</sub> of 91.7%, representing an improvement of 3.1% over YOLOv8 and 16.1% over detection transformer (DETR). On the SPDI dataset, our method achieves a <i>mAP</i><sub>50</sub> of 69.3%, showing a 2.1% improvement over YOLOv8 and a 6.6% improvement over DETR. The proposed MRA-YOLOv8 also exhibits great deployment potential. It can be effectively integrated with drone-based inspection systems, allowing for efficient and accurate PV plant inspections. Moreover, to tackle the issue of data imbalance, we propose generating synthetic defect data via generative adversarial networks (GANs), which can supplement the limited defect samples and improve the model’s generalization ability.https://www.mdpi.com/1424-8220/25/5/1542photovoltaic cell defect detectionMBCANetResBlockAMPDIoUMRA-YOLOv8
spellingShingle Nannan Wang
Siqi Huang
Xiangpeng Liu
Zhining Wang
Yi Liu
Zhe Gao
MRA-YOLOv8: A Network Enhancing Feature Extraction Ability for Photovoltaic Cell Defects
Sensors
photovoltaic cell defect detection
MBCANet
ResBlock
AMPDIoU
MRA-YOLOv8
title MRA-YOLOv8: A Network Enhancing Feature Extraction Ability for Photovoltaic Cell Defects
title_full MRA-YOLOv8: A Network Enhancing Feature Extraction Ability for Photovoltaic Cell Defects
title_fullStr MRA-YOLOv8: A Network Enhancing Feature Extraction Ability for Photovoltaic Cell Defects
title_full_unstemmed MRA-YOLOv8: A Network Enhancing Feature Extraction Ability for Photovoltaic Cell Defects
title_short MRA-YOLOv8: A Network Enhancing Feature Extraction Ability for Photovoltaic Cell Defects
title_sort mra yolov8 a network enhancing feature extraction ability for photovoltaic cell defects
topic photovoltaic cell defect detection
MBCANet
ResBlock
AMPDIoU
MRA-YOLOv8
url https://www.mdpi.com/1424-8220/25/5/1542
work_keys_str_mv AT nannanwang mrayolov8anetworkenhancingfeatureextractionabilityforphotovoltaiccelldefects
AT siqihuang mrayolov8anetworkenhancingfeatureextractionabilityforphotovoltaiccelldefects
AT xiangpengliu mrayolov8anetworkenhancingfeatureextractionabilityforphotovoltaiccelldefects
AT zhiningwang mrayolov8anetworkenhancingfeatureextractionabilityforphotovoltaiccelldefects
AT yiliu mrayolov8anetworkenhancingfeatureextractionabilityforphotovoltaiccelldefects
AT zhegao mrayolov8anetworkenhancingfeatureextractionabilityforphotovoltaiccelldefects