Defect Detection of Eddy Current Thermal Imaging of Workpiece Based on Deep Learning and Domain Adaptation
-
摘要: 机械设备运行过程中，标记的故障样本量小，导致建立的模型故障诊断准确率低，为此本文提出一种结合深度学习与域自适应的工件涡流热成像的缺陷检测方法。首先将注意力机制引入深度残差网络ResNet50中，加强模型的特征提取能力；然后将源域和目标域数据送入改进的ResNet50网络中提取深度特征，并且在网络的全连接层中引入局部最大均值差异，用于缩小两域特征间的分布差异，以此实现相关子域的分布对齐；最后在网络的Softmax分类器中实现对工件金属材料的缺陷检测。在公开的磁瓦数据集和本文实验采集的金属板涡流红外图像数据集上进行实验，结果表明，本文方法对涡流红外图像的裂纹缺陷检测识别准确率较高，通过t分布随机邻居嵌入方法对分析结果可视化，验证了本文方法的优越性。Abstract: When operating mechanical equipment, the number of fault samples marked is small, which leads to low accuracy of the fault diagnosis of the established model. Therefore, this study proposes a defect detection method for eddy current thermal imaging of a workpiece that combines depth learning and domain adaptation. First, the attention mechanism is introduced into the deep residual network ResNet50 to enhance the feature extraction capability of the model. Then, the source and target domain data are sent into the improved ResNet50 network to extract the depth features. The local maximum mean difference is introduced into the full connection layer of the network to reduce the distribution difference between the two domain features to achieve the distribution alignment of related sub-domains. Finally, workpiece metal material defects were detected in the Softmax classifier of the network. The experiment was conducted on the open magnetic tile dataset and eddy current infrared image dataset of the metal plate collected during the experiment. The results show that the method proposed in this paper is highly accurate in detecting and recognizing crack defects in eddy current infrared images. The advantages of the method in this study were verified by visualizing the analysis results using the t-distribution random neighbor embedding method.
-
-
![]()
图 3 CBAM_ResNet50和子域自适应网络模型
Figure 3. CBAM_ResNet50 and subdomain adaptive network model
![]()
图 4 涡流加热设备（左）和缺陷金属板（右）
Figure 4. Eddy current heating equipment (left) and defective metal plate (right)
![]()
图 7 不同方法的精确度对比图
Figure 7. Accuracy comparison chart of different methods
Magnetic tile data set→sheet metal data set
表 1 添加CBAM的ResNet50网络结构
Table 1 ResNet50 network structure with CBAM added
Network layer Parameters Activation function Conv1 64×7×7 Relu CBAM 64×1×1
7×7Sigmoid Conv2_x $ \left. {\begin{array}{*{20}{c}} {64 \times 1 \times 1} \\ {64 \times 3 \times 3} \\ {256 \times 1 \times 1} \end{array}} \right\} \times 3 $ Relu Conv3_x $ \left. {\begin{array}{*{20}{c}} {128 \times 1 \times 1} \\ {128 \times 3 \times 3} \\ {512 \times 1 \times 1} \end{array}} \right\} \times 4 $ Relu Conv4_x $ \left. {\begin{array}{*{20}{c}} {256 \times 1 \times 1} \\ {256 \times 3 \times 3} \\ {1024 \times 1 \times 1} \end{array}} \right\} \times 6 $ Relu Conv5_x $ \left. {\begin{array}{*{20}{c}} {512 \times 1 \times 1} \\ {512 \times 3 \times 3} \\ {2048 \times 1 \times 1} \end{array}} \right\} \times 3 $ Relu CBAM 2048×1×1
7×7Sigmoid FC 2 Softmax 
下载: 导出CSV
表 2 不同模型的检测精度
Table 2 Detection accuracy of different models
% Methods Magnetic tile→sheet metal Sheet metal→magnetic tile Average accuracy ResNet50 63.93 59.18 61.56 DAN 78.19 73.53 75.86 ResNet50_LMMD 88.29 86.10 87.20 This paper 90.11 86.93 88.52
下载: 导出CSV
-
[1] 李宝芸, 范玉刚, 高阳. 基于OTSU和Canny算子的红外图像特征提取[J]. 陕西理工大学学报: 自然科学版, 2019, 35(6): 33-40. https://www.cnki.com.cn/Article/CJFDTOTAL-SXGX201906007.htm LI Baoyun, FAN Yugang, GAO Yang. Infrared image feature extraction based on OTSU and Canny operator[J]. Journal of Shaanxi University of Technology: Natural Science Edition, 2019, 35(6): 33-40. https://www.cnki.com.cn/Article/CJFDTOTAL-SXGX201906007.htm
[2] PENG Y, HUANG S, HE Y, et al. Eddy current pulsed thermography for noncontact nondestructive inspection of motor winding defects[J]. IEEE Sensors Journal, 2020, 20(5): 2625-2634. DOI: 10.1109/JSEN.2019.2952691
[3] YI Q, Malekmohammadi H, TIAN G Y, et al. Quantitative evaluation of crack depths on thin aluminum plate using eddy current pulse-compression thermography[J]. IEEE Transations on Industrial Informatics, 2021, 16(6): 3963-3973.
[4] 董绍江, 朱朋, 裴雪武, 等. 基于子领域自适应的变工况下滚动轴承故障诊断[J]. 吉林大学学报: 工学版, 2022, 52(2): 288-295. https://www.cnki.com.cn/Article/CJFDTOTAL-JLGY202202004.htm DONG Shaojiang, ZHU Peng, PEI Xuewu, et al. Fault diagnosis of rolling bearing under variable operating conditions based on subdomain adaptation[J]. Journal of Jilin University: Engineering and Technology Edition, 2022, 52(2): 288-295. https://www.cnki.com.cn/Article/CJFDTOTAL-JLGY202202004.htm
[5] 姜万录, 岳毅, 张淑清, 等. 基于特征迁移学习的变工况下轴向柱塞泵故障诊断[J]. 农业工程学报, 2022, 38(5): 45-55. https://www.cnki.com.cn/Article/CJFDTOTAL-NYGU202205005.htm JIANG Wanlu, Yue Yi, ZHANG Shuqing, et al. Axial piston pump fault diagnosis under variable working conditions based on feature transfer learning[J]. Transactions of the Chinese Society of Agricultural Engineering, 2022, 38(5): 45-55. https://www.cnki.com.cn/Article/CJFDTOTAL-NYGU202205005.htm
[6] LONG M S, CAO Y, WANG J, et al. Learning transferable features with deep adaptation networks[C]//International Conference on Machine Learning, 2015: 97-105.
[7] Ganin Y, Ustinova E, Ajakan H, et al. Domain-adversarial training of neural networks[J]. Journal of Machine Learning Research, 2016, 17(1): 2096-2030.
[8] 陈佛计, 朱枫, 吴清潇, 等. 生成对抗网络及其在图像生成中的应用研究综述[J]. 计算机学报, 2021, 44(2): 347-369. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJX202102007.htm CHEN Foji, ZHU Feng, WU Qingxiao, et al. A survey about image generation with generative adversarial nets[J]. Chinese Journal of Computers, 2021, 44(2): 347-369. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJX202102007.htm
[9] 王格格, 郭涛, 余游, 等. 基于生成对抗网络的无监督域适应分类模型[J]. 电子学报, 2020, 48(6): 1190-1197. https://www.cnki.com.cn/Article/CJFDTOTAL-DZXU202006021.htm WANG Gege, GUO Tao, YU You, et al. Unsupervised domain adaptation classification model based on generative adversarial network[J]. Acta Electronica Sinica, 2020, 48(6): 1190-1197. https://www.cnki.com.cn/Article/CJFDTOTAL-DZXU202006021.htm
[10] ZHU Y C, ZHUANG F Z, WANG J D, et al. Deep subdomain adaptation network for image classification[J]. IEEE Transactions on Neural Networks and Learning Systems, 2020(99): 1-10.
[11] Gretton A, Borgwardt K M, Rasch M J, et al. A kernel two-sample test[J]. Journal of Machine Learning Research, 2012, 13: 723-773.
[12] Woo S, Park J, Lee J Y, et al. CBAM: Convolutional block attention module[C]//Proceedings of the European Conference on Computer Vision, 2018: 3-19.
[13] 郝帅, 张旭, 马旭, 等. 基于CBAM-YOLOv5的煤矿输送带异物检测[J]. 煤炭学报, 2021, 1644: 1-11. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB202211026.htm HAO Shuai, ZHANG Xu, MA Xu, et al. Foreign object detection in coal mine conveyor belt based on CBAM-YOLOv5[J]. Journal of China Coal Society, 2021, 1644: 1-11. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB202211026.htm
[14] HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//IEEE Conference on Computer Vision & Pattern Recognition, 2016: 770-778.
[15] 吴静然, 刘建华, 崔冉. 子域适应无监督轴承故障诊断[J]. 振动与冲击, 2021, 40(15): 34-40. https://www.cnki.com.cn/Article/CJFDTOTAL-ZDCJ202115005.htm WU Jingran, LIU Jianhua, CUI Ran. Sub-domain adaptive unsupervised bearing fault diagnosis[J]. Journal of Vibration and Shock, 2021, 40(15): 34-40. https://www.cnki.com.cn/Article/CJFDTOTAL-ZDCJ202115005.htm
[16] HUANG Y B, QIU C Y, YUAN K. Surface defect saliency of magnetic tile[J]. The Visual Computer, 2020, 36(1): 85-96. DOI: 10.1007/s00371-018-1588-5
[17] Laurens V D M, Hinton G. Visualizing data using t-SNE[J]. Journal of Machine Learning Research, 2008, 9(2605): 2579-2605.
计量
- 文章访问数: 87
- HTML全文浏览量: 25
- PDF下载量: 27
