面向光伏组件缺陷检测的多源图像融合方法研究

陈登峰; 刘茜茜; 刘世鹏; 肖海燕

doi:doi:10.19912/j.0254-0096.tynxb.2024-2110

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

面向光伏组件缺陷检测的多源图像融合方法研究

更新时间：2026-07-22

- 面向光伏组件缺陷检测的多源图像融合方法研究
- Vol. 47, Issue 4, Pages: 622-629(2026)
- 作者机构：
- 作者简介：
- 基金信息：
- DOI：doi:10.19912/j.0254-0096.tynxb.2024-2110
  CLC： TK51
- Online First：08 May 2026，
  
  Published：2026
- 稿件说明：
移动端阅览
陈登峰, 刘茜茜, 刘世鹏, et al. 面向光伏组件缺陷检测的多源图像融合方法研究[J]. 2026, 47(4): 622-629.
DOI：

陈登峰, 刘茜茜, 刘世鹏, et al. 面向光伏组件缺陷检测的多源图像融合方法研究[J]. 2026, 47(4): 622-629. DOI： doi:10.19912/j.0254-0096.tynxb.2024-2110.

摘要

针对光伏组件单一图像缺陷检测准确性和全面性不足的问题

提出一种面向光伏组件缺陷检测的多源图像融合方法。首先

在图像融合网络的编码器中嵌入交叉调制策略

以特征对齐代替图像配准

揭示光伏组件可见光与红外图像之间的特征对应关系;其次

采用“跳过”机制

实现注意力机制的选择性应用

加快图像处理速度并降低资源消耗;最后

利用融合后的图像在YOLO系列缺陷检测网络上实现检测。实验结果表明

所提多源图像融合方法有效克服了单一图像缺陷检测的局限

在多个缺陷检测网络上的精确率平均提升5.6%。

Abstract

To address the limitations of accuracy and comprehensiveness in single-image defect detection for photovoltaic modules

a multi-source image fusion method is proposed for detecting defects in photovoltaic modules. Firstly

the Cross-Modulation is embedded in the encoder of the image fusion network. Feature alignment is used instead of image registration to reveal the correspondence between the features of visible light and infrared images of photovoltaic modules

enabling the fusion of unregistered images. Secondly

the Skip Mechanism is adopted to selectively apply the attention mechanism

accelerating image processing and reducing resource consumption. Finally

the fused images are utilized for detection using the YOLO series defect detection networks. Experimental results indicate that the proposed multi-source image fusion method effectively overcomes the limitations of single-image defect detection

resulting in an average precision improvement of 5.6% across multiple networks.

关键词

Keywords

references

王道累, 姚从荣, 李超, 等. 光伏组件热斑效应智能化检测的综述及展望[J]. 太阳能学报, 2024, 45(5): 527-536.

CHIRIAC G, NITUCA C, CUCIUREANU D, et al.Analysis of hot-spots effects on the performances of a photovoltaic system used in Romania[C]//2017 International Conference on Electromechanical and Power Systems (SIELMEN). Iasi, Romania, 2017: 239-244.

TANG W Q, YANG Q, DAI Z, et al.Module defect detection and diagnosis for intelligent maintenance of solar photovoltaic plants: techniques, systems and perspectives[J]. Energy, 2024, 297: 131222.

APPIAH A Y, ZHANG X H, AYAWLI B B K, et al. Review and performance evaluation of photovoltaic array fault detection and diagnosis techniques[J]. International journal of photoenergy, 2019, 2019(1): 6953530.

KÖNTGES M, KURTZ S, PACKARD CE, et al. Review of failures of photovoltaic modules[R].IEA-PVPS T13-01:2014.

LI X X, LI W, YANG Q, et al.Edge-computing-enabled unmanned module defect detection and diagnosis system for large-scale photovoltaic plants[J]. IEEE internet of things journal, 2020, 7(10): 9651-9663.

RICO ESPINOSA A, BRESSAN M, GIRALDO L F.Failure signature classification in solar photovoltaic plants using RGB images and convolutional neural networks[J]. Renewable energy, 2020, 162: 249-256.

李琼, 吴文宝, 刘斌, 等. 基于迁移学习的光伏组件鸟粪覆盖检测[J]. 太阳能学报, 2022, 43(2): 233-237.

NAVEEN VENKATESH S, REBECCA JEYAVADHANAM B, MORADI SIZKOUHI A M, et al. Automatic detection of visual faults on photovoltaic modules using deep ensemble learning network[J]. Energy reports, 2022, 8: 14382-14395.

蒋琳, 苏建徽, 施永, 等. 基于红外热图像处理的光伏阵列热斑检测方法[J]. 太阳能学报, 2020, 41(8): 180-184.

王道累, 李超, 李明山, 等. 基于深度卷积神经网络的光伏组件热斑检测[J]. 太阳能学报, 2022, 43(1): 412-417.

孙海蓉, 李帆. 基于注意力机制的光伏热斑识别[J]. 太阳能学报, 2023, 44(2): 453-459.

蒋琳, 苏建徽, 李欣, 等. 基于可见光和红外热图像融合的光伏阵列热斑检测方法[J]. 太阳能学报, 2022, 43(1): 393-397.

HONG F, SONG J, MENG H, et al.A novel framework on intelligent detection for module defects of PV plant combining the visible and infrared images[J]. Solar energy, 2022, 236: 406-416.

范钧玮, 饶全瑞, 赵薇, 等. 改进的YOLOv5双影像光伏故障小目标检测[J]. 太阳能学报, 2024, 45(7): 510-516.

JEFFREY KUO C F, CHEN S H, HUANG C Y. Automatic detection, classification and localization of defects inlargephotovoltaic plants using unmanned aerial vehicles (UAV) based infrared (IR) and RGB imaging[J]. Energy conversion and management, 2023, 276: 116495.

孔松涛, 徐甄泽, 林星宇, 等. 基于改进YOLOv5算法的光伏组件红外热成像缺陷检测[J]. 红外技术, 2023, 45(9): 974-981.

LI H F, ZHAO J Z, LI J X, et al.Feature dynamic alignment and refinement for infrared-visible image fusion: translation robust fusion[J]. Information fusion, 2023, 95: 26-41.

VENKATARAMANAN S, GHODRATI A, ASANO Y M, et al. Skip-attention: improving vision transformers by paying less attention[J]. arXiv preprint arXiv:2301. 02240, 2023.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

PHOTOVOLTAIC MODULE DEFECT DETECTION METHOD BASED ON IMPROVED YOLOV5S NETWORK

BIRD DROPPINGS COVERAGE DETECTION OF PHOTOVOLTAIC MODULE BASED ON TRANSFER LEARNING

PHOTOVOLTAIC DEFECT DETECTION BASED ON MULTI-SCALE CODING COMPLEMENTARY ATTENTION NETWORK

基于改进YOLOv8的红外光伏组件缺陷检测方法

基于深度残差注意力网络的光伏组件分割

Related Author

Ren Xiwei

Yu Jie

Han Xin

Li Zhaoyun

Yang Menglu

He Lifeng

Li Qiong

Wu Wenbao

Related Institution

Shaanxi Joint Laboratory of Artificial Intelligence

Shaanxi University of Science and Technology, School of Electronic Information and Artificial Intelligence

School of Information Engineering, Nanchang Hangkong University

Jiangxi Electric Power Construction Co., Ltd. of China Power Construction Group

School of Artificial Intelligence, Hebei University of Technology

AI问答

⁰