首页|基于YOLOv4的风电机组叶片半监督表面缺陷检测

基于YOLOv4的风电机组叶片半监督表面缺陷检测

扫码查看
原文链接
  • 国家科技期刊平台
  • NETL
  • NSTL
  • 万方数据
及时检测出风力涡轮机叶片表面的缺陷可有效预防难以预测的事故。为此，本研究提出了一种基于YOLOv4的半监督目标检测网络。为了克服获取足够标签的困难，本研究设计了一种由生成式对抗网络(GAN)组成的半监督结构。在生成式对抗网络中，生成器由编码器-解码器网络实现，其中编码器的骨干是 YOLOv4，解码器由反卷积层组成。来自生成器的部分特征被传递给缺陷检测网络。利用大量未标记图像可以显著提高缺陷检测模型的泛化和识别能力。通过在 YOLOv4 网络的三个部分中添加scSE注意力模块，增强特征图中的基本特征，可以提高网络的小范围目标检测能力。此外，对 YOLOv4 的损失函数进行了平衡改进，以克服缺陷物种的不平衡问题。单类别和多类别缺陷数据集上的实验结果表明，改进后的模型可以很好地利用未标注的缺陷数据集的特征。风力涡轮机叶片缺陷检测的准确性与传统的物体检测算法(包括faster R-CNN 和 DETR)相比也具有显著优势。
Semi-supervised surface defect detection of wind turbine blades with YOLOv4
Timely inspection of defects on the surfaces of wind turbine blades can effectively prevent unpredictable accidents.To this end,this study proposes a semi-supervised object-detection network based on You Only Looking Once version 4(YOLOv4).A semi-supervised structure comprising a generative adversarial network(GAN)was designed to overcome the difficulty in obtaining sufficient samples and sample labeling.In a GAN,the generator is realized by an encoder-decoder network,where the backbone of the encoder is YOLOv4 and the decoder comprises inverse convolutional layers.Partial features from the generator are passed to the defect detection network.Deploying several unlabeled images can significantly improve the generalization and recognition capabilities of defect-detection models.The small-scale object detection capacity of the network can be improved by enhancing essential features in the feature map by adding the concurrent spatial and channel squeeze and excitation(scSE)attention module to the three parts of the YOLOv4 network.A balancing improvement was made to the loss function of YOLOv4 to overcome the imbalance problem of the defective species.The results for both the single-and multi-category defect datasets show that the improved model can make good use of the features of the unlabeled images.The accuracy of wind turbine blade defect detection also has a significant advantage over classical object detection algorithms,including faster R-CNN and DETR.

Defect detectionGenerative adversarial networkscSE attentionSemi-supervisionWind turbine

黄超、陈明辉、王龙

展开 >

Department of Computer and Communication Engineering,University of Science and Technology Beijing,Beijing 100083,P.R.China

Shunde Innovation School,University of Science and Technology Beijing,Foshan 528399,P.R.China

缺陷检测 生成式对抗网络 scSE注意力机制 半监督 风力涡轮机

National Natural Science Foundation of ChinaNational Natural Science Foundation of ChinaScientific and Technological Innovation Foundation of FoshanExcellent Youth Team Project for the Central UniversitiesFundamental Research Funds for the Central UniversitiesFundamental Research Funds for the Central UniversitiesGuangdong Basic and Applied Basic Research FoundationBeijing Natural Science Foundation

6220204462372039BK22BF009FRF-EYIT-23-0106500103065000782022A15152400444232040

2024

10.1016/j.gloei.2024.06.010

全球能源互联网(英文)

全球能源互联网(英文)

CSTPCDEI
ISSN：2096-5117
年,卷(期)：2024.7(3)