摘要
光伏板是光伏发电系统的核心部件,其质量好坏直接影响发电效率及电路安全。为了精准检测出光伏板的缺陷,提出了1种融合注意力机制的YOLOv5改进算法,该算法将有效通道注意力(efficient channel attention,ECA)与YOLOv5模型主干网络中的C3模块相融合形成C3-ECA模块。同时将融合注意力机制YOLOv5改进算法与YOLOv3、YOLOX等多个模型做对比实验,结果表明融合注意力机制YOLOv5改进算法精确率为97.5%,比原版YOLOv5提高了1.1%。改进的算法在引入少量参数的情况下,提高了模型的检测精度,并能够对光伏板表面的多种缺陷进行有效识别,且精度高、耗时少。
As photovoltaic panel is the core component of photovoltaic power generation system,its quality directly affects the power generation efficiency and circuit safety.In order to accurately detect defects of photovoltaic panels,an improved algorithm of YOLOv5 fusing attention mechanism was proposed,which fused efficient channel attention(ECA)with C3 module in the backbone network of YOLOv5 model to form C3-ECA module.Meanwhile,a comparison experiment was conducted between the improved algorithm of YOLOv5 fusing attention mechanism and YOLOv3,YOLOX and other models.The results show that the accuracy of the improved algorithm of YOLOv5 fusing attention mechanism is 97.5%,which is 1.1%higher than that of the original YOLOv5.With the introduction of a few parameters,the improved algorithm improves the detection accuracy of the model and can effectively identify a variety of defects on the surface of the photovoltaic panel,with high precision and less time.
作者
赵晓雨
高林
杨校李
彭运猛
ZHAO Xiaoyu;GAO Lin;YANG Xiaoli;PENG Yunmeng(College of Intelligent Systems Science and Engineering,Hubei Minzu University,Enshi 445000,China)
出处
《湖北民族大学学报（自然科学版）》
CAS
2023年第1期65-70,共6页
Journal of Hubei Minzu University:Natural Science Edition
基金
国家自然科学基金项目(61562025,61962019)
湖北省高等学校省级教学研究项目(2017387)。
关键词
光伏板
YOLOv5
注意力机制
缺陷检测
电致发光图像
photovoltaic panels
YOLOv5
mechanism of attention
defect detection
electroluminescent image
