浏览全部资源
扫码关注微信
Published Online： 23 December 2024 ，
移动端阅览
叶标华,康丹青,谢晓华等.基于视觉的非完全标注表面缺陷检测综述[J].中国图象图形学报,
Ye Biaohua,Kang Danqing,Xie Xiaohua,et al.Review of Surface Defect Inspection with Incomplete Annotations[J].Journal of Image and Graphics,
在现代制造业中，基于机器视觉的表面缺陷检测是保证产品质量的关键，在工业智能化发展中发挥着重要作用。然而，获取缺陷数据的标注需要花费大量人力和时间成本。随着深度学习、 大数据和传感器等技术的发展，如何在非完全标注的情况下实现准确、快速和鲁棒的缺陷识别成为当前的研究热点。该文对非完全标注场景下的表面缺陷检测技术的研究进展进行了全面的梳理回顾。首先简要介绍了缺陷检测领域的研究背景、基础概念的定义、常用数据集和相关技术。在此基础上，从标签策略以及任务策略两个角度详细介绍了多种非完全标注场景下的缺陷检测技术。在标签策略中，本文比较了基于无监督、半监督、弱监督学习下的不同缺陷检测算法的研究思路。在任务策略中，本文总结了领域自适应、小样本以及大模型的表面缺陷检测算法的最新进展。接着，本文在多个数据集上横向对比了不同标签策略以及任务策略中前沿算法的性能。最后，对该任务中的弱小目标检测、伪标签质量评估以及大模型的知识迁移等问题进行总结和展望。总体而言，非完全标注的表面缺陷检测是一个充满挑战且技术性极强的问题。同时，如何进一步推动表面缺陷检测技术进一步利用非完全标注的数据，并切实在工业制造场景中落地应用还需要更深入的研究。
Surface defect inspection is an industrial automation technology primarily used to identify and evaluate defects and anomalies in products or materials. Its purpose is to ensure product quality， reduce production costs， and improve production efficiency by obtaining relevant information such as the coordinates， categories， sizes， and contours of defects. In industrial contexts， defects are generally understood as the presence of missing parts， flaws， damages， errors， and anomalies compared to normal samples. Depending on the granularity of dataset annotations， defect information annotation can be categorized into three types： image-level annotation， instance-level annotation， and pixel-level annotation， corresponding to three computer vision tasks： classification， detection， and segmentation. In recent years， due to the rapid development of technologies such as machine vision， big data， and sensors， replacing human labor with machines has become possible. Using machine vision technology to replace manual inspection for surface defect inspection in industrial products has increasingly become a trend and mainstream practice. However， in industrial settings， detailed annotation of defect data requires significant time and effort， and there is an abundance of historical unannotated data in these settings. The research problem of surface defect inspection with incomplete annotations investigates how to utilize the rich unannotated data or leverage a small amount of annotated data to improve the effectiveness of surface defect inspection. Automated visual inspection systems are a crucial step in industrial manufacturing processes， replacing manual quality inspection and enabling smart manufacturing. Researchers have conducted studies on surface defect inspection technologies based on deep learning， involving the latest methods， applications， and challenges. However， there is still a lack of literature reviews on surface defect inspection methods with incomplete annotations in the field of industrial product surface defect inspection. This paper systematically summarizes the research background， definitions of basic concepts， commonly used datasets， and related technologies in the field of surface defect inspection with incomplete annotations. According to different industrial contexts， we have collected and organized commonly used datasets in fields such as metal surfaces， fabrics， building materials， and 3C electronic products. The paper meticulously categorizes various defect inspection techniques from the perspectives of label strategies and task strategies， pointing out the characteristics， advantages， and disadvantages of each method. Based on different label types， surface defect inspection tasks with incomplete annotations can be divided into unsupervised learning， semi-supervised learning， and weakly supervised learning methods. Unsupervised learning methods mainly include strategies such as clustering， positive sample modeling， and template matching. Semi-supervised learning methods primarily include techniques such as image reconstruction， pseudo-label generation， and data augmentation. Weakly supervised learning methods mainly involve the use of class activation maps and similar techniques. Based on different task strategies， surface defect inspection tasks with incomplete annotations can be categorized into domain adaptation-based methods， few-shot learning-based methods， and large model-based methods. Domain adaptation methods mainly include approaches such as distribution alignment， learning domain-invariant features， and dynamically adjusting hyperparameters. Few-shot learning methods primarily include techniques such as meta-learning， metric learning， and graph neural networks. Large model-based methods mainly involve the use of large models like SAM， CLIP， LLM， and VLM. Subsequently， this paper compares the performance of cutting-edge unsupervised， semi-supervised， and weakly supervised algorithms on multiple datasets. Finally， the paper discusses and forecasts future research trends in surface defect inspection with incomplete annotations， including the detection of weak small targets， efficient utilization of unannotated defect images， and the application of language models and vision-language models in defect inspection. Surface defect inspection tasks with incomplete annotations are commonly encountered in the industry， holding significant research and application value， deserving more attention and promotion from both the industrial and academic communities.
缺陷检测非完全标注无监督学习弱监督学习半监督学习域适应小样本
surface defect inspectionincomplete annotationsunsupervised learningweak-supervised learningsemi-supervised learningdomain adaptationfew-shot
Ahmadi M， Lonbar A G， Sharifi A， Beris A T， Nouri M and Javidi A S. 2023. Application of segment anything model for civil infrastructure defect assessment ［EB/OL］. ［2024-03-17］. http:// arxiv. org/ abs/ 2304. 12600. pdfhttp://arxiv.org/abs/2304.12600.pdf
Ai J S， Chen S H， Deng P， Bai L B， Tian L L and Zhang J. 2020. CycleGANs for semi-supervised defects segmentation// Proceedings of 2020 International Conference on Sensing， Measurement and Data Analytics in the era of Artificial Intelligence. Xi’an， China： IEEE： 611–616 ［DOI： 10. 1109/ ICSMD50554. 2020. 9261-743http://dx.doi.org/10.1109/ICSMD50554.2020.9261-743］
Bao Y Q， Song K C， Liu J， Wang Y Y， Yan Y H， Yu H and Li X J. 2021. Triplet-graph reasoning network for few-shot metal generic surface defect segmentation. IEEE Transactions on Instrumentation and Measurement， 70：1–11 ［DOI： 10. 1109/ TIM. 2021. 3083561http://dx.doi.org/10.1109/TIM.2021.3083561］
Bergmann P， Fauser M， Sattlegger D and Steger C. 2019a. MVTec ad - a comprehensive real-world dataset for unsupervised anomaly detection// Proceedings of 2019 IEEE/ CVF Conference on Computer Vision and Pattern Recognition. Long Beach， CA， USA： IEEE： 9584–9592 ［DOI： 10. 1109/ CVPR. 2019. 00982http://dx.doi.org/10.1109/CVPR.2019.00982］
Bergmann P， Löwe S， Fauser M， Sattlegger D and Steger C. 2019b. Improving unsupervised defect segmentation by applying structural similarity to autoencoders：// Proceedings of the 14th International Joint Conference on Computer Vision， Imaging and Computer Graphics Theory and Applications. Prague， Czech Republic： SCITEPRESS - Science and Technology Publications：372–380 ［DOI： 10. 5220/ 0007364503720380http://dx.doi.org/10.5220/0007364503720380］
Carrera D， Manganini F， Boracchi G and Lanzarone E. 2017. Defect detection in sem images of nanofibrous materials. IEEE Transactions on Industrial Informatics， 13（2）：551–561 ［DOI： 10. 1109/ TII. 2016. 2641472http://dx.doi.org/10.1109/TII.2016.2641472］
Chen L Y， Shu-Min L K， Jiang X H， Wang S J， Huang A A， Chen J E， Liang H C and Hsu C L. 2021. Semi-supervised framework for wafer defect pattern recognition with enhanced labeling// Proceedings of 2021 IEEE International Test Conference. Anaheim， CA， USA： IEEE： 208–212 ［DOI： 10. 1109/ ITC50571. 2021. 00029http://dx.doi.org/10.1109/ITC50571.2021.00029］
Chen Z W， Wong W K， Zhong Z F， Liao J P and Qu Y. 2023. Effective transfer of pretrained large visual model for fabric defect segmentation via specifc knowledge injection ［EB/OL］. ［2024-03-17］. http://arxiv. org/ pdf/ 2306. 16186. pdfhttp://arxiv.org/pdf/2306.16186.pdf
Chen R Y， Yu G T， Qin Z， Song K K， Tu J F， Jiang X L， Liang D and Peng C B. 2024. Patch matching for few-shot industrial defect detection. IEEE Transactions on Instrumentation and Measurement， 73：1–11［DOI： 10. 1109/ TIM. 2024. 3413170http://dx.doi.org/10.1109/TIM.2024.3413170］
Cheng D P， Sha W X， Xu Z， Li S D， Yin Z G， Lang Y L， Tang S and Cao Y C. 2023. AtomGAN： unsupervised deep learning for fast and accurate defect detection of 2d materials at the atomic scale. Science China Information Sciences， 66（6）：1–13 ［DOI： 10. 1007/ s11432-022-3757-xhttp://dx.doi.org/10.1007/s11432-022-3757-x］
Cheng J H， Guo B， Liu J Q， Liu S C， Wu G Z， Sun Y Q and Yu Z W. 2021. TL-sdd： a transfer learning-based method for surface defect detection with few samples// Proceedings of 2021 7th International Conference on Big Data Computing and Communications. Deqing， China： IEEE： 136–143［DOI： 10. 1109/ BigCom-53800. 2021. 00023http://dx.doi.org/10.1109/BigCom-53800.2021.00023］
DeVitis G A， Foglia P and Prete C A. 2020. Row‐level algorithm to improve real‐time performance of glass tube defect detection in the production phase. IET Image Processing， 14（12）： 2911–2921 ［DOI： 10. 1049/ iet-ipr. 2019. 1506http://dx.doi.org/10.1049/iet-ipr.2019.1506］
Deitsch S， Christlein V， Berger S， Buerhop-Lutz C， Maier A， Gallwitz F and Riess C. 2019. Automatic classification of defective photovoltaic module cells in electroluminescence images. Solar Energy， 185： 455–468 ［DOI： 10. 1016/ j. solener. 2019. 02. 067http://dx.doi.org/10.1016/j.solener.2019.02.067］
Deng Y B and Song Y H. 2023. Few-shot steel plate surface defect detection with multi-relation aggregation and adaptive support learning. ISIJ International， 63（10）： 1727–1737 ［DOI： 10. 2355/ isijinternational. ISIJINT-2023-118http://dx.doi.org/10.2355/isijinternational.ISIJINT-2023-118］
Di H， Ke X， Peng Z and Dong Z. 2019. Surface defect classification of steels with a new semi-supervised learning method. Optics and Lasers in Engineering， 117： 40–48 ［DOI： 10. 1016/ j. optlaseng. 2019. 01. 011http://dx.doi.org/10.1016/j.optlaseng.2019.01.011］
Ding R W， Dai L H， Li G P and Liu H. 2019. TDD‐net： a tiny defect detection network for printed circuit boards. CAAI Transactions on Intelligence Technology，4（2）： 110–116［DOI： 10. 1049/ trit. 2019. 0019http://dx.doi.org/10.1049/trit.2019.0019］
Duan G F， Song Y G， Liu Z Y， Ling S Q and Tan J R. 2023. Cross-domain few-shot defect recognition for metal surfaces. Measurement Science and Technology， 34（1）： 15202–15212 ［DOI： 10. 1088/ 1361-6501/ ac90dehttp://dx.doi.org/10.1088/1361-6501/ac90de］
Era I Z， Ahmed I， Liu Z C and Das S. 2023. An unsupervised approach towards promptable defect segmentation in laser-based additive manufacturing by segment anything ［EB/OL］. ［2024-03-17］. http: //arxiv. org/ pdf/ 2312. 04063. pdfhttp://arxiv.org/pdf/2312.04063.pdf
Fridman Y， Rusanovsky M T and Oren G. 2021. ChangeChip： a reference-based unsupervised change detection for pcb defect detection// Proceedings of 2021 IEEE Physical Assurance and Inspection of Electronics. Washington， DC， USA： IEEE： 1–8 ［DOI： 10. 1109/ PAINE54418. 2021. 9707699http://dx.doi.org/10.1109/PAINE54418.2021.9707699］
Gan J R， Li Q Y， Wang J Z and Yu H M. 2017. A hierarchical extractor-based visual rail surface inspection system. IEEE Sensors Journal， 17–34（23）： 7935–7944 ［DOI： 10. 1109/ JSEN. 2017. 2761858http://dx.doi.org/10.1109/JSEN.2017.2761858］
Gao Y P， Gao L， Li X Y and Yan X G. 2020. A semi-supervised convolutional neural network-based method for steel surface defect recognition. Robotics and Computer-Integrated Manufacturing， 61： 101825–101833 ［DOI： 10. 1016/ j. rcim. 2019. 101825http://dx.doi.org/10.1016/j.rcim.2019.101825］
Gu Z P， Zhu B K， Zhu G B， Chen Y Y， Tang M and Wang J Q. 2024. AnomalyGPT： detecting industrial anomalies using large vision-language models. Proceedings of The AAAI Conference on Artificial Intelligence， 38（3）： 1932–1940 ［DOI： 10. 1609/ aaai. v38i3. 27963http://dx.doi.org/10.1609/aaai.v38i3.27963］
Guo W and Jiang P Y. 2024. Weakly supervised anomaly detection with privacy preservation under a bi-level federated learning framework. Expert Systems With Applications， 254： 124450–124462 ［DOI： 10. 1016/ j. eswa. 2024. 124450http://dx.doi.org/10.1016/j.eswa.2024.124450］
Guo Y J， Zhong L W， Qiu Y， Wang H W， Gao F Q， Wen Z H and Zhan C J. 2022. Using isu-gan for unsupervised small sample defect detection. Scientific Reports， 12（1）： 11604–11617 ［DOI： 10. 1038/ s41- 598-022-15855-7http://dx.doi.org/10.1038/s41-598-022-15855-7］
Hamdi A A， Sayed M S， Fouad M M and Hadhoud M M. 2018. Unsupervised patterned fabric defect detection using texture filtering and k-means clustering// Proceedings of 2018 International Conference on Innovative Trends in Computer Engineering. Aswan： IEEE： 130–144 ［DOI： 10. 1109/ ITCE. 2018. 8316611http://dx.doi.org/10.1109/ITCE.2018.8316611］
He K K， Liu X T， Liu J and Wu P. 2021. A multitask learning-based neural network for defect detection on textured surfaces under weak supervision. IEEE Transactions on Instrumentation and Measurement， 70： 1–14 ［DOI： 10. 1109/ TIM. 2021. 3112784http://dx.doi.org/10.1109/TIM.2021.3112784］
He T J， Li H E， Qian Z D， Niu C Y and Huang R H. 2024. Research on weakly supervised pavement crack segmentation based on defect location by generative adversarial network and target re‐optimization. Construction and Building Materials， 411： 134668–134680 ［DOI： 10. 1016/ j. conbuildmat. 2023. 134668http://dx.doi.org/10.1016/j.conbuildmat.2023.134668］
He Y， Song K C， Dong H W and Yan Y H. 2019. Semi-supervised defect classification of steel surface based on multi-training and generative adversarial network. Optics and Lasers in Engineering， 122： 294–302 ［DOI： 10. 1016/ j. optlaseng. 2019. 06. 020http://dx.doi.org/10.1016/j.optlaseng.2019.06.020］
He Y， Song K C， Meng Q G and Yan Y H. 2020. An end-to-end steel surface defect detection approach via fusing multiple hierarchical features. IEEE Transactions on Instrumentation and Measurement， 69（4）： 1493–1504 ［DOI： 10. 1109/ tim. 2019. 2915404http://dx.doi.org/10.1109/tim.2019.2915404］
Hu B Z， Gao B， Tan C， Wu T L and Li S Z. 2023. Segment anything in defect detection ［EB/OL］. ［2024-03-17］. http: //arxiv. org/ pdf/ 231 1. 10245. pdfhttp://arxiv.org/pdf/2311.10245.pdf
Hu G H， Huang J F， Wang Q H， Li J R， Xu Z J and Huang X B. 2020. Unsupervised fabric defect detection based on a deep convolutional generative adversarial network. Textile Research Journal， 90（3–4）： 247–270 ［DOI： 10. 1177/ 0040517519862880http://dx.doi.org/10.1177/0040517519862880］
Hu Y D， Liu R N， Li X L， Chen D Y and Hu Q H. 2022. Task-sequencing meta learning for intelligent few-shot fault diagnosis with limited data. IEEE Transactions on Industrial Informatics， 18–36（6）： 3894–3904 ［DOI： 10. 1109/ TII. 2021. 3112504http://dx.doi.org/10.1109/TII.2021.3112504］
Hu T， Zhang J N， Yi R， Du Y Z， Chen X， Liu L， Wang Y B and Wang C J. 2024. Anomalydiffusion： few-shot anomaly image generation with diffusion model// Proceedings of the AAAI conference on artificial intelligence. Vancouver， Canada： AAAI Press： 8526–8534［DOI： 10. 1609/ aaai. v38i8. 28696http://dx.doi.org/10.1609/aaai.v38i8.28696］
Huang C， Chen M H and Wang L. 2024. Semi-supervised surface defect detection of wind turbine blades with yolov4. Global Energy Interconnection， 7（3）： 284–292 ［DOI： 10. 1016/ j. gloei. 2024. 06. 010http://dx.doi.org/10.1016/j.gloei.2024.06.010］
Huang Y B， Qiu C Y and Yuan K. 2020. Surface defect saliency of magnetic tile. The Visual Computer， 36（1）： 85–96 ［DOI： 10. 1007/ s00371-018-1588-5http://dx.doi.org/10.1007/s00371-018-1588-5］
Huang Y Y， Liu B G， Dong H C， Lu B H， Jiang Y H and Liu G D. 2023. A novel weakly supervised learning method for industrial surface defect detection// Proceedings of 2023 6th International Conference on Artificial Intelligence and Big Data. Chengdu， China： IEEE： 741–745 ［DOI： 10. 1109/ ICAIBD57115. 2023. 10206091http://dx.doi.org/10.1109/ICAIBD57115.2023.10206091］
Jain S， Seth G， Paruthi A， Soni U and Kumar G. 2022. Synthetic data augmentation for surface defect detection and classification using deep learning. Journal of Intelligent Manufacturing， 33（4）： 1007–1020 ［DOI： 10. 1007/ s10845-020-01710-xhttp://dx.doi.org/10.1007/s10845-020-01710-x］
Jamali A， Laflamme C， Huber R and Hammad A M. 2023. PAVEMENT defect classification and localization using weakly supervised deep learning// Proceedings of the Creative Construction Conference 2023. Keszthely， Hungary： Budapest University of Technology and Economics： 506–514 ［DOI： 10. 3311/ CCC2023- 067http://dx.doi.org/10.3311/CCC2023-067］
Jeong J， Zou Y， Kim T， Zhang D Q， Ravichandran A and Dabeer O. 2023. WinCLIP： zero-/few-shot anomaly classification and segmentation// Proceedings of 2023 IEEE/ CVF Conference on Computer Vision and Pattern Recognition. Vancouver， BC， Canada： IEEE： 19606–19616［DOI： 10. 1109/ CVPR52729. 2023. 01878http://dx.doi.org/10.1109/CVPR52729.2023.01878］
Ji W， Li J J， Bi Q， Liu T W， Li W B and Cheng L. 2023. Segment anything is not always perfect： an investigation of sam on different real-world applications ［EB/OL］. ［2024-03-17］. http: //arxiv. org/ pdf/ 2304. 05750. pdfhttp://arxiv.org/pdf/2304.05750.pdf
Jiang L L， Wang Y X， Tang Z H， Miao Y L and Chen S Y. 2021. Casting defect detection in x-ray images using convolutional neural networks and attention-guided data augmentation. Measurement， 170： 108736–108744 ［DOI： 10. 1016/ j. measurement. 2020. 10- 8736http://dx.doi.org/10.1016/j.measurement.2020.10-8736］
Jiang X H， Feng J， Yan F， Lu Y， Fa Q H， Zhang W J and Xu M L. 2024. Foreground–background separation transformer for weakly supervised surface defect detection. Journal of Intelligent Manufacturing， 1： 1–16 ［DOI： 10. 1007/ s10845-024-02446-8http://dx.doi.org/10.1007/s10845-024-02446-8］
Jing F W， Zhang M Y， Li J， Xu G Z and Wang J. 2022. Coil shape defects prediction algorithm for hot strip rolling based on siamese semi-supervised dae-cnn model. Assembly Automation， 42–54（6）： 773–781 ［DOI： 10. 1108/ AA-07-2022-0179http://dx.doi.org/10.1108/AA-07-2022-0179］
Kasahara J Y L， Fujii H， Yamashita A and Asama H. 2021. Weakly supervised acoustic defect detection in concrete structures using clustering-based augmentation. IEEE/ ASME Transactions on Mechatronics， 26（6）： 2826–2834 ［DOI： 10. 1109/ TMECH. 2021. 307- 7496http://dx.doi.org/10.1109/TMECH.2021.307-7496］
Konig J， Jenkins M D， Mannion M K， Barrie P and Morison G. 2022. Weakly-supervised surface crack segmentation by generating pseudo-labels using localization with a classifier and thresholding. IEEE Transactions on Intelligent Transportation Systems， 23（12）： 24083–24094 ［DOI： 10. 1109/ TITS. 2022. 3204853http://dx.doi.org/10.1109/TITS.2022.3204853］
Koulali I and Eskil M T. 2021. Unsupervised textile defect detection using convolutional neural networks. Applied Soft Computing， 113： 107913–107953 ［DOI： 10. 1016/ j. asoc. 2021. 107913http://dx.doi.org/10.1016/j.asoc.2021.107913］
Kuric I， Klarák J， Sága M L， Císar M， Hajdučík A and Wiecek D. 2021. Analysis of the possibilities of tire-defect inspection based on unsupervised learning and deep learning. Sensors， 21（21）： 7073–7097 ［DOI： 10. 3390/ s21217073http://dx.doi.org/10.3390/s21217073］
Laurençon H， Tronchon L， Cord M and Sanh V. 2024. What matters when building vision-language models？ ［EB/OL］. ［2024-07-11］. https: //arxiv. org/ pdf/ 2405. 02246. pdfhttps://arxiv.org/pdf/2405.02246.pdf
Lee H and Kim H. 2020. Semi-supervised multi-label learning for classification of wafer bin maps with mixed-type defect patterns. IEEE Transactions on Semiconductor Manufacturing， 33（4）： 653–662 ［DOI： 10. 1109/ TSM. 2020. 3027431http://dx.doi.org/10.1109/TSM.2020.3027431］
Lee X Y， Vidyaratne L， Alam M， Farahat A， Ghosh D， Diaz T G and Gupta C. 2023. XDNet： a few-shot meta-learning approach for cross-domain visual inspection// Proceedings of 2023 IEEE/ CVF Conference on Computer Vision and Pattern Recognition Workshops. Vancouver， BC， Canada： IEEE： 4375–4384 ［DOI： 10. 1109/ CVPRW 59228. 2023. 00460http://dx.doi.org/10.1109/CVPRW59228.2023.00460］
Li C， Yan H X， Qian X， Zhu S D， Zhu P， Liao C W， Tian H Y， Li X， Wang X H and Li H X. 2023a. A domain adaptation yolov5 model for industrial defect inspection. Measurement， 213： 112725–112734 ［DOI： 10. 1016/ j. measurement. 2023. 112725http://dx.doi.org/10.1016/j.measurement.2023.112725］
Li K M， Jiang X H， Chen L Y， Wang S J， Huang A A， Chen J E， Liang H C and Hsu C L. 2022a. Wafer defect pattern labeling and recognition using semi-supervised learning. IEEE Transactions on Semiconductor Manufacturing， 35–49（2）： 291–299 ［DOI： 10. 1109 / TSM. 2022. 3159246http://dx.doi.org/10.1109/TSM.2022.3159246］
Li L F， Ma W F， Li L and Lu C. （2018）. Research on detection algorithm for bridge cracks based on deep learning. Acta Automatica Sinica， 45（9）： 1727-1742
李良福， 马卫飞， 李丽， 陆铖. 2019. 基于深度学习的桥梁裂缝检测算法研究. 自动化学报， 45（9）： 1727-1742 ［DOI： 10. 16383/ j. aas. 2018. c170052http://dx.doi.org/10.16383/j.aas.2018.c170052］
Li S B， Yang J， Wang Z， Zhu S D and Yang G C. （2020）. Review of development and application of defect detection technology. Acta Automatica Sinica， 46（11）： 2319-2336
李少波， 杨静， 王铮， 朱书德， 杨观赐. 2020. 缺陷检测技术的发展与应用研究综述. 自动化学报， 46（11）： 2319-2336 ［DOI： 10. 16383/ j. aas. c180538http://dx.doi.org/10.16383/j.aas.c180538］
Li Y M， Wu X J， Li P and Liu Y H. 2023b. Ferrite beads surface defect detection based on spatial attention under weakly supervised learning. IEEE Transactions on Instrumentation and Measurement， 72： 1–12 ［DOI： 10. 1109/ TIM. 2023. 3246499http://dx.doi.org/10.1109/TIM.2023.3246499］
Li Y T， Bao T F， Xu B， Shu X S， Zhou Y H， Du Y， Wang R J and Zhang K. 2022b. A deep residual neural network framework with transfer learning for concrete dams patch-level crack classification and weakly-supervised localization. Measurement， 188： 110641 –110654 ［DOI： 10. 1016/ j. measurement. 2021. 110641http://dx.doi.org/10.1016/j.measurement.2021.110641］
Lin D Y， Cao Y P， Zhu W B and Li Y Q. 2021. Few-shot defect segmentation leveraging abundant defect-free training samples through normal background regularization and crop-and-paste operation// Proceedings of 2021 IEEE International Conference on Multimedia and Expo. Shenzhen， China： IEEE： 1–6［DOI： 10. 1109/ ICME51207. 2021. 9428468http://dx.doi.org/10.1109/ICME51207.2021.9428468］
Liu T and Ye W. 2022. A semi-supervised learning method for surface defect classification of magnetic tiles. Machine Vision and Applications， 33（2）： 35–49 ［DOI： 10. 1007/ s00138-022-01286-xhttp://dx.doi.org/10.1007/s00138-022-01286-x］
Liu J， Song K C， Feng M Z， Yan Y H， Tu Z B and Zhu L. 2021. Semi-supervised anomaly detection with dual prototypes autoencoder for industrial surface inspection. Optics and Lasers in Engineering， 136： 106324–106333 ［DOI： 10. 1016/ j. optlaseng. 2020. 106- 324http://dx.doi.org/10.1016/j.optlaseng.2020.106-324］
Liu J H， Guo F， Zhang Y， Hou B K and Zhou H M. 2022. Defect classification on limited labeled samples with multiscale feature fusion and semi-supervised learning. Applied Intelligence， 52（7）： 8243– 8258 ［DOI： 10. 1007/ s10489-021-02917-yhttp://dx.doi.org/10.1007/s10489-021-02917-y］
Liu J Q， Xie G Y， Wang J B， Li S N， Wang C J， Zheng F and Jin Y C. 2024a. Deep industrial image anomaly detection： a survey. Machine Intelligence Research， 21（1）： 104–135 ［DOI： 10. 1007/ s11633- 023-1459-zhttp://dx.doi.org/10.1007/s11633-023-1459-z］
Liu L， Zhang J H， Fu X D， Liu L J and Huang Q S. 2019. Unsupervised segmentation and elm for fabric defect image classification. Multimedia Tools and Applications， 78（9）： 12421–12449 ［DOI： 10. 1007/ s11042-018-6786-7http://dx.doi.org/10.1007/s11042-018-6786-7］
Liu W Q， Liu Z G， Wang H and Han Z W. 2020. An automated defect detection approach for catenary rod-insulator textured surfaces using unsupervised learning. IEEE Transactions on Instrumentation and Measurement， 69（10）： 8411–8423 ［DOI： 10. 1109/ tim. 2020. 2987503http://dx.doi.org/10.1109/tim.2020.2987503］
Liu Z Y， Song Y G， Tang R N， Duan G F and Tan J R. 2023. Few-shot defect recognition of metal surfaces via attention-embedding and self-supervised learning. Journal of Intelligent Manufacturing， 34（8）： 3507–3521 ［DOI： 10. 1007/ s10845-022-02022-yhttp://dx.doi.org/10.1007/s10845-022-02022-y］
Luo W， Niu T Z， Yao H M， Tang L X， Yu W Y and Li B. 2023. Unsupervised defect segmentation via forgetting-inputting-based feature fusion and multiple hierarchical feature difference. IEEE Sensors Journal， 23（13）： 14448–14459 ［DOI： 10. 1109/ JSEN. 2023. 3276762http://dx.doi.org/10.1109/JSEN.2023.3276762］
Lv C K， Zhang Z T， Shen F and Zhang F. 2023. Unsupervised automatic defect inspection based on image matching and local one-class classification// Proceedings of 2023 IEEE/ CVF Conference on Computer Vision and Pattern Recognition Workshops. Vancouver， BC， Canada： IEEE： 4435–4444 ［DOI： 10. 1109/ CVPR- W59228. 2023. 00466http://dx.doi.org/10.1109/CVPR-W59228.2023.00466］
Ma S， Song K C， Niu M H， Tian H K， Wang Y Y and Yan Y H. 2023. Shape-consistent one-shot unsupervised domain adaptation for rail surface defect segmentation. IEEE Transactions on Industrial Informatics， 19–38（9）： 9667–9679［DOI： 10. 1109/ TII. 2022. 3233654http://dx.doi.org/10.1109/TII.2022.3233654］
Ma S， Song K C， Niu M H， Tian H K and Yan Y H. 2024a. Cross-scale fusion and domain adversarial network for generalizable rail surface defect segmentation on unseen datasets. Journal of Intelligent Manufacturing， 35–49（1）： 367–386 ［DOI： 10. 1007/ s10845-022-02051-7http://dx.doi.org/10.1007/s10845-022-02051-7］
Ma S， Song K C， Niu M H， Tian H K， Wang Y Y and Yan Y H. 2024b. Feature-based domain disentanglement and randomization： a generalized framework for rail surface defect segmentation in unseen scenarios. Advanced Engineering Informatics， 59： 102274–102288 ［DOI： 10. 1016/ j. aei. 2023. 102274http://dx.doi.org/10.1016/j.aei.2023.102274］
Maeda H， Sekimoto Y， Seto T， Kashiyama T and Omata H. 2018. Road damage detection and classification using deep neural networks with smartphone images. Computer-Aided Civil and Infrastructure Engineering， 33（12）： 1127–1141 ［DOI： 10. 1111/ mice. 12387http://dx.doi.org/10.1111/mice.12387］
Manivannan S. 2023. Automatic quality inspection in additive manufacturing using semi-supervised deep learning. Journal of Intelligent Manufacturing， 34（7）： 3091–3108 ［DOI： 10. 1007/ s10845-022- 02000-4http://dx.doi.org/10.1007/s10845-022-02000-4］
Manivannan S. 2024. Semi-supervised imbalanced classification of wafer bin map defects using a dual-head cnn. Expert Systems With Applications， 238： 122301–122310 ［DOI： 10. 1016/ j. eswa. 2023. 122- 301http://dx.doi.org/10.1016/j.eswa.2023.122-301］
Marino S， Beauseroy P and Smolarz A. 2019. Weakly-supervised learning approach for potato defects segmentation. Engineering Applications of Artificial Intelligence， 85： 337–346 ［DOI： 10. 1016/ j. engappai. 2019. 06. 024http://dx.doi.org/10.1016/j.engappai.2019.06.024］
Marino S， Beauseroy P and Smolarz A. 2020. Unsupervised adversarial deep domain adaptation method for potato defects classification. Computers and Electronics in Agriculture， 174： 105501–105511 ［DOI： 10. 1016/ j. compag. 2020. 105501http://dx.doi.org/10.1016/j.compag.2020.105501］
Mei S， Yang H and Yin Z P. 2018. An unsupervised-learning-based approach for automated defect inspection on textured surfaces. IEEE Transactions on Instrumentation and Measurement， 67（6）： 1266–1277 ［DOI： 10. 1109/ TIM. 2018. 2795178http://dx.doi.org/10.1109/TIM.2018.2795178］
Mery D， Riffo V， Zscherpel U， Mondragón G， Lillo I， Zuccar I， Lobel H and Carrasco M. 2015. GDXray： the database of x-ray images for nondestructive testing. Journal of Nondestructive Evaluation， 34（4）： 42–54 ［DOI： 10. 1007/ s10921-015-0315-7http://dx.doi.org/10.1007/s10921-015-0315-7］
Min Y Z and Li Y X. 2022. Self-supervised railway surface defect detection with defect removal variational autoencoders. Energies， 15（10）： 3592–3607 ［DOI： 10. 3390/ en15103592http://dx.doi.org/10.3390/en15103592］
Min Y Z， Wang Z W， Liu Y and Wang Z. 2023. FS-rsdd： few-shot rail surface defect detection with prototype learning. Sensors， 23（18–36）： 7894–7913 ［DOI： 10. 3390/ s23187894http://dx.doi.org/10.3390/s23187894］
Minhas M S and Zelek J. 2019. Anonet： weakly supervised anomaly detection in textured surfaces ［EB/OL］. . https: //arxiv. org/ pdf/ 19- 11. 10608. pdfhttps://arxiv.org/pdf/19-11.10608.pdf
Minhas M S and Zelek J. 2020. Semi-supervised anomaly detection using autoencoders ［EB/OL］. ［2024-07-11］. http: //arxiv. org/ pdf/ 2 001. 03674. pdfhttp://arxiv.org/pdf/2001.03674.pdf
Mujeeb A， Dai W T， Erdt M and Sourin A. 2018. Unsupervised surface defect detection using deep autoencoders and data augmentation// Proceedings of 2018 International Conference on Cyberworlds. Singapore： IEEE： 391–398 ［DOI： 10. 1109/ cw. 2018. 00076http://dx.doi.org/10.1109/cw.2018.00076］
Nie M， Jiang W， Yang W K， Wang S L， Wen X Q and Ni T M. 2023. Enhancing defect diagnosis and localization in wafer map testing through weakly supervised learning// Proceedings of 2023 IEEE 32nd Asian Test Symposium. Beijing， China： IEEE： 1–6 ［DOI： 10. 1109/ ATS59501. 2023. 10317989http://dx.doi.org/10.1109/ATS59501.2023.10317989］
Niu M H， Song K C， Huang L M， Wang Q， Yan Y H and Meng Q G. 2020. Unsupervised saliency detection of rail surface defects using stereoscopic images. IEEE Transactions on Industrial Informatics， 17–34（3）： 2271–2281 ［DOI： 10. 1109/ tii. 2020. 3004397http://dx.doi.org/10.1109/tii.2020.3004397］
Niu S L， Li B， Wang X G， He S P and Peng Y R. 2022. Defect attention template generation cyclegan for weakly supervised surface defect segmentation. Pattern Recognition， 123： 108396–108410 ［DOI： 10. 1016/ j. patcog. 2021. 108396http://dx.doi.org/10.1016/j.patcog.2021.108396］
Ozdemir R and Koc M. 2024. on the enhancement of semi-supervised deep learning-based railway defect detection using pseudo-labels. Expert Systems With Applications， 251： 124105–124114 ［DOI： 10. 1016/ j. eswa. 2024. 124105http://dx.doi.org/10.1016/j.eswa.2024.124105］
Pang S C， Zhao W S， Wang S D， Zhang L and Wang S. 2024. Permute-maml： exploring industrial surface defect detection algorithms for few-shot learning. Complex and Intelligent Systems， 10（1）： 1473–1482 ［DOI： 10. 1007/ s40747-023-01219-9http://dx.doi.org/10.1007/s40747-023-01219-9］
Park J E and Kim Y K. 2023. Semi-supervised learning for steel surface inspection using magnetic flux leakage signal. Journal of Intelligent Manufacturing， 1： 1–11 ［DOI： 10. 1007/ s10845-023-02286-yhttp://dx.doi.org/10.1007/s10845-023-02286-y］
Peng C Q， Zhao L， Wang S， Abbas Z S， Liang F and Islam M S. 2024. LightFlow： lightweight unsupervised defect detection based on 2-d flow. IEEE Transactions on Instrumentation and Measurement， 73： 1–12 ［DOI： 10. 1109/ TIM. 2024. 3415769http://dx.doi.org/10.1109/TIM.2024.3415769］
Phalawan L， Phisanbut N and Piamsa-nga P. 2023. A weakly-supervised glass bottle defect detection system based on multi-view analysis// Proceedings of 2023 20th International Joint Conference on Computer Science and Software Engineering. Phitsanulok， Thailand： IEEE： 79–83 ［DOI： 10. 1109/ JCSSE58229. 2023. 10201963http://dx.doi.org/10.1109/JCSSE58229.2023.10201963］
Qi H C， Kong X W， Shen Z A， Liu Z T and Gu J Y. 2023. Progressively learning dynamic level set for weakly supervised industrial defect segmentation. IEEE Transactions on Instrumentation and Measurement， 72： 1–14 ［DOI： 10. 1109/ TIM. 2023. 3316271http://dx.doi.org/10.1109/TIM.2023.3316271］
Qi H C， Kong X W， Liu Z T， Gu J Y and Cheng L. 2024a. SAIT： harnessing sparse annotations and intrinsic tasks for semisupervised aeroengine defect segmentation. IEEE Transactions on Industrial Informatics， 1–10 ［DOI： 10. 1109/ TII. 2024. 3396550http://dx.doi.org/10.1109/TII.2024.3396550］
Qi H C， Cheng L， Kong X W， Zhang J Q and Gu J Y. 2024b. WDLS： deep level set learning for weakly supervised aeroengine defect segmentation. IEEE Transactions on Industrial Informatics， 20（1）： 303–313 ［DOI： 10. 1109/ TII. 2023. 3261889http://dx.doi.org/10.1109/TII.2023.3261889］
Rasheed A， Zafar B， Rasheed A， Ali N， Sajid M， Dar S H， Habib U， Shehryar T and Mahmood M T. 2020. Fabric defect detection using computer vision techniques： a comprehensive review. Mathematical Problems in Engineering， 2020： 1–24 ［DOI： 10. 1155/ 2020/ 8189 403http://dx.doi.org/10.1155/2020/8189403］
Rongsheng L， Ang W， Tengda Z and Yonghong W. 2018. Review on automated optical inspection and its applications in defect detection. Acta Optica Sinica， 38（8）： 0815002 ［DOI： 10. 3788/ AOS 201838. 0815002http://dx.doi.org/10.3788/AOS201838.0815002］
Rudolph M， Wandt B and Rosenhahn B. 2021. Same same but differnet： semi-supervised defect detection with normalizing flows// Proceedings of 2021 IEEE Winter Conference on Applications of Computer Vision. Waikoloa， HI， USA： IEEE： 1906–1915 ［DOI： 10. 1109/ WACV48630. 2021. 00195http://dx.doi.org/10.1109/WACV48630.2021.00195］
Sae-ang B I， Kumwilaisak W and Kaewtrakulpong P. 2022. Semi-supervised learning for defect segmentation with autoencoder auxiliary module. Sensors， 22（8）： 2915–2930 ［DOI： 10. 3390/ s2208 2915http://dx.doi.org/10.3390/s22082915］
Shao L H， Zhang E H， Ma Q R and Li M. 2022. Pixel-wise semisupervised fabric defect detection method combined with multitask mean teacher. IEEE Transactions on Instrumentation and Measurement， 71： 1–11 ［DOI： 10. 1109/ TIM. 2022. 3162286http://dx.doi.org/10.1109/TIM.2022.3162286］
Sheynin S， Benaim S and Wolf L. 2021. A hierarchical transformation-discriminating generative model for few shot anomaly detection// Proceedings of IEEE International Conference on Computer Vision. Montreal， QC， Canada： IEEE： 8475–8484［DOI： 10. 1109/ ICCV48922. 2021. 00838http://dx.doi.org/10.1109/ICCV48922.2021.00838］
Shi S J， Tan J H and Chen H Y. 2023a. Dynamic domain adaptation object detection for photovoltaic cell defects// Proceedings of 2023 42nd Chinese Control Conference. Tianjin， China： IEEE： 6903–6907 ［DOI： 10. 23919/ CCC58697. 2023. 10241054http://dx.doi.org/10.23919/CCC58697.2023.10241054］
Shi Y， Cui L M， Qi Z Q， Meng F and Chen Z S. 2016. Automatic road crack detection using random structured forests. IEEE Transactions on Intelligent Transportation Systems， 17–34（12）： 3434–3445 ［DOI： 10. 1109/ TITS. 2016. 2552248http://dx.doi.org/10.1109/TITS.2016.2552248］
Shi Y D， Wang H M， Jing C and Zhang X Z. 2023b. A few-shot defect detection method for transmission lines based on meta-attention and feature reconstruction. Applied Sciences， 13（10）： 5896–5910 ［DOI： 10. 3390/ app13105896http://dx.doi.org/10.3390/app13105896］
Silvestre-Blanes J， Albero-Albero T， Miralles I， Pérez-Llorens R and Moreno J. 2019. A public fabric database for defect detection methods and results. Autex Research Journal， 19–38（4）： 363–374 ［DOI： 10. 2478/ aut-2019-0035http://dx.doi.org/10.2478/aut-2019-0035］
Sime D M， Wang G T， Zeng Z and Peng B. 2023a. Uncertainty-aware and dynamically-mixed pseudo-labels for semi-supervised defect segmentation. Computers in Industry， 152： 103995–104007 ［DOI： 10. 1016/ j. compind. 2023. 103995http://dx.doi.org/10.1016/j.compind.2023.103995］
Sime D M， Wang G T， Zeng Z， Wang W and Peng B. 2023b. Semisupervised defect segmentation with pairwise similarity map consistency and ensemble-based cross pseudolabels. IEEE Transactions on Industrial Informatics， 19–38（9）： 9535–9545 ［DOI： 10. 1109/ tii. 2022. 3230785http://dx.doi.org/10.1109/tii.2022.3230785］
Simmler N， Sager P， andermatt P， Chavarriaga R， Schilling F P， Rosenthal M and Stadelmann T. 2021. A survey of un-， weakly-， and semi-supervised learning methods for noisy， missing and partial labels in industrial vision applications// Proceedings of 2021 8th Swiss Conference on Data Science. Lucerne， Switzerland： IEEE： 26–31 ［DOI： 10. 1109/ SDS51136. 2021. 00012http://dx.doi.org/10.1109/SDS51136.2021.00012］
Song K C and Yan Y H. 2013. A noise robust method based on completed local binary patterns for hot-rolled steel strip surface defects. Applied Surface Science， 285： 858–864 ［DOI： 10. 1016/ j. apsusc. 2013. 09. 002http://dx.doi.org/10.1016/j.apsusc.2013.09.002］
Song Y G， Liu Z Y， Wang J H， Tang R N， Duan G F and Tan J R. 2021. Multiscale adversarial and weighted gradient domain adaptive network for data scarcity surface defect detection. IEEE Transactions on Instrumentation and Measurement， 70： 1–10 ［DOI： 10. 1109/ tim. 2021. 3096284http://dx.doi.org/10.1109/tim.2021.3096284］
Song Y G， Liu Z Y， Ling S Q， Tang R N， Duan G F and Tan J R. 2022. Coarse-to-fine few-shot defect recognition with dynamic weighting and joint metric. IEEE Transactions on Instrumentation and Measurement， 71： 1–10 ［DOI： 10. 1109/ TIM. 2022. 3193204http://dx.doi.org/10.1109/TIM.2022.3193204］
Sresakoolchai J and Kaewunruen S. 2022. Railway defect detection based on track geometry using supervised and unsupervised machine learning. Structural Health Monitoring， 21（4）： 1757–1767 ［DOI： 10. 1177/ 14759217211044492http://dx.doi.org/10.1177/14759217211044492］
Sun C， Gao L， Li X Y and Gao Y P. 2022. A new knowledge distillation network for incremental few-shot surface defect detection［EB/OL］. ［2022-8-22］. https: //arxiv. org/ pdf/ 2209. 00519. pdfhttps://arxiv.org/pdf/2209.00519.pdf
Szarski M and Chauhan S. 2022. An unsupervised defect detection model for a dry carbon fiber textile. Journal of Intelligent Manufacturing， 33（7）： 2075–2092 ［DOI： 10. 1007/ s10845-022-01964-7http://dx.doi.org/10.1007/s10845-022-01964-7］
Tabernik D， Šela S M， Skvarč J R and Skočaj D. 2020. Segmentation-based deep-learning approach for surface-defect detection. Journal of Intelligent Manufacturing， 31（3）： 759–776 ［DOI： 10. 1007/ s10845-019-01476-xhttp://dx.doi.org/10.1007/s10845-019-01476-x］
Taherkhani K， Eischer C and Toyserkani E. 2022. An unsupervised machine learning algorithm for in-situ defect-detection in laser powder-bed fusion. Journal of Manufacturing Processes， 81： 476–489 ［DOI： 10. 1016/ j. jmapro. 2022. 06. 074http://dx.doi.org/10.1016/j.jmapro.2022.06.074］
Tang B， Kong J Y， Wu S Q．2017. Review of surface defect detection based on machine vision．Journal of Image and Graphics，22（ 12） ： 1640-1663．
汤勃， 孔建益， 伍世虔．2017. 机器视觉表面缺陷检测综述． 中国图象图形学报， 22（ 12） ： 1640-1663．［DOI： 10. 11834 / jig. 160623http://dx.doi.org/10.11834/jig.160623］
Tang S L， He F， Huang X L and Yang J. 2019. online pcb defect detector on a new pcb defect dataset ［EB/OL］. ［2024-07-08］. http: //arxiv. org/ pdf/ 1902. 06197. pdfhttp://arxiv.org/pdf/1902.06197.pdf
Tao X， Zhang D P， Wang Z H， Liu X L， Zhang H Y and Xu D. 2020. Detection of power line insulator defects using aerial images analyzed with convolutional neural networks. IEEE Transactions on Systems， Man， and Cybernetics： Systems， 50（4）： 1486–1498 ［DOI： 10. 1109/ TSMC. 2018. 2871750http://dx.doi.org/10.1109/TSMC.2018.2871750］
Tao X， Hou W and Xu De. 2021. A survey of surface defect detection methods based on deep learning. Acta Automatica Sinica， 47（5）： 1017-1034
陶显， 侯伟， 徐德. 基于深度学习的表面缺陷检测方法综述. 2021. 自动化学报， 47（5）： 1017-1034 ［DOI： 10. 16383/ j. aas. c190811http://dx.doi.org/10.16383/j.aas.c190811］
Volkau I， Mujeeb A， Wenting D， Marius E and Alexei S. 2019. Detection defect in printed circuit boards using unsupervised feature extraction upon transfer learning// Proceedings of 2019 International Conference on Cyberworlds. Kyoto， Japan： IEEE： 101–108 ［DOI： 10. 1109/ CW. 2019. 00025http://dx.doi.org/10.1109/CW.2019.00025］
Wan Y S， Gao L， Li X Y and Gao Y P. 2022. Semi-supervised defect detection method with data-expanding strategy for pcb quality inspection. Sensors， 22（20）： 7971–7988 ［DOI： 10. 3390/ s22207971http://dx.doi.org/10.3390/s22207971］
Wang D， Zhang M， Xu Y C， Lu W N， Yang J and Zhang T. 2021a. Metric-based meta-learning model for few-shot fault diagnosis under multiple limited data conditions. Mechanical Systems and Signal Processing， 155： 107510–107542 ［DOI： 10. 1016/ j. ymssp. 2020. 107510http://dx.doi.org/10.1016/j.ymssp.2020.107510］
Wang D T， Yu W J， Lian P Y and Zhang M J. 2022a. Textile defect detection algorithm based on unsupervised learning// Proceedings of 2022 7th International Conference on Image， Vision and Computing. Xi’an， China： IEEE： 81–86 ［DOI： 10. 1109/ ICIVC55077. 2022. 9887216http://dx.doi.org/10.1109/ICIVC55077.2022.9887216］
Wang H， Qin J Y， Bastola A， Chen X W， Suchanek J， Gong Z H and Razi A. 2024a. VisionGPT： llm-assisted real-time anomaly detection for safe visual navigation ［EB/OL］. ［2024-07-16］. http: //arxiv. org/ pdf/ 2403. 12415. pdfhttp://arxiv.org/pdf/2403.12415.pdf
Wang H H， Li Z L and Wang H Q. 2022e. Few-shot steel surface defect detection. IEEE Transactions on Instrumentation and Measurement， 71： 1–12［DOI： 10. 1109/ TIM. 2021. 3128208http://dx.doi.org/10.1109/TIM.2021.3128208］
Wang H X， Li Y F， Dang L M， Lee S J and Moon H. 2021b. Pixel-level tunnel crack segmentation using a weakly supervised annotation approach. Computers in Industry， 133： 103545–103556 ［DOI： 10. 1016/ j. compind. 2021. 103545http://dx.doi.org/10.1016/j.compind.2021.103545］
Wang J C， Yi G D， Zhang S Y and Wang Y. 2020a. An unsupervised generative adversarial network-based method for defect inspection of texture surfaces. Applied Sciences， 11（1）： 283–298 ［DOI： 10. 3390/ app11010283http://dx.doi.org/10.3390/app11010283］
Wang L， Jia K Y， Li M and Yin J. 2023a. Industrial product surface defects detection based on weakly supervised learning// Proceedings of 2023 42nd Chinese Control Conference. Tianjin， China： IEEE： 6987–6992 ［DOI： 10. 23919/ CCC58697. 2023. 10240953http://dx.doi.org/10.23919/CCC58697.2023.10240953］
Wang R X， Cheung C F and Wang C J. 2023b. Unsupervised defect segmentation in selective laser melting. IEEE Transactions on Instrumentation and Measurement， 72： 1–10 ［DOI： 10. 1109/ TIM. 2023. 3291002http://dx.doi.org/10.1109/TIM.2023.3291002］
Wang S H， Wang D Z， Kong D S， Wang J X， Li W H and Zhou S. 2020b. Few-shot rolling bearing fault diagnosis with metric-based meta learning. Sensors， 20（22）： 6437–6452 ［DOI： 10. 3390/ s2022- 6437http://dx.doi.org/10.3390/s2022-6437］
Wang W J， Su C， Han G H and Hu S P. 2024b. Semi-supervised learning for concrete defect segmentation from images. Structural Health Monitoring， 1： 1–20 ［DOI： 10. 1177/ 14759217- 231217097http://dx.doi.org/10.1177/14759217-231217097］
Wang Y L， Zhang Y L， Zheng L， Yin L D， Chen J S and Lu J A. 2021c. Unsupervised learning with generative adversarial network for automatic tire defect detection from x-ray images. Sensors， 21（20）： 6773–6792 ［DOI： 10. 3390/ s21206773http://dx.doi.org/10.3390/s21206773］
Wang Y X， Yan J， Yang Z， Qi Z K， Wang J H and Geng Y S. 2022b. Gas-insulated switchgear insulation defect diagnosis via a novel domain adaptive graph convolutional network. IEEE Transactions on Instrumentation and Measurement， 71： 1–10 ［DOI： 10. 1109/ tim. 2022. 3160543http://dx.doi.org/10.1109/tim.2022.3160543］
Wang Y X， Yan J， Yang Z， Qi Z K， Wang J H and Geng Y S. 2022c. A novel domain adversarial graph convolutional network for insulation defect diagnosis in gas-insulated substations. IEEE Transactions on Power Delivery， 38（1）： 442–452 ［DOI： 10. 1109/ tpwrd. 2022. 3190938http://dx.doi.org/10.1109/tpwrd.2022.3190938］
Wang Y Y， Song K C， Niu M H， Bao Y Q， Dong H W and Yan Y H. 2022d. Unsupervised defect detection with patch-aware mutual reasoning network in image data. Automation in Construction， 142： 104472–104488 ［DOI： 10. 1016/ j. autcon. 2022. 104472http://dx.doi.org/10.1016/j.autcon.2022.104472］
Wang Z L， Leng Z F and Zhang Z X. 2024c. A weakly-supervised transformer-based hybrid network with multi-attention for pavement crack detection. Construction and Building Materials， 411： 134134–134145 ［DOI： 10. 1016/ j. conbuildmat. 2023. 134134http://dx.doi.org/10.1016/j.conbuildmat.2023.134134］
Wu H Y， Li B， Tian L F， Sun Z Z， Dong C and Liao W Z. 2023. CoRe： contrastive and restorative self-supervised learning for surface defect inspection. IEEE Transactions on Instrumentation and Measurement， 72： 1–12 ［DOI： 10. 1109/ TIM. 2023. 3291776http://dx.doi.org/10.1109/TIM.2023.3291776］
Wu K， Tan J and Liu C B. 2022a. Cross-domain few-shot learning approach for lithium-ion battery surface defects classification using an improved siamese network. IEEE Sensors Journal， 22（12）： 11847–11856 ［DOI： 10. 1109/ JSEN. 2022. 3161331http://dx.doi.org/10.1109/JSEN.2022.3161331］
Wu X J， Wang T， Li Y M， Li P and Liu Y H. 2022b. A cam-based weakly supervised method for surface defect inspection. IEEE Transactions on Instrumentation and Measurement， 71： 1–10 ［DOI： 10. 1109/ TIM. 2022. 3168895http://dx.doi.org/10.1109/TIM.2022.3168895］
Wu Y， Xing M W， Zhang Y C， Xie Y， Wu Z Z and Qu Y Y. 2024. Perturbed progressive learning for semisupervised defect segmentation. IEEE Transactions on Neural Networks and Learning Systems， 35–49（5）： 6118–6132 ［DOI： 10. 1109/ TNNLS. 2023. 3324188http://dx.doi.org/10.1109/TNNLS.2023.3324188］
Xiao W W， Song K C， Liu J and Yan Y H. 2022. Graph embedding and optimal transport for Explainable Deep Few-sho of metal surface defect. IEEE Transactions on Instrumentation and Measurement， 71： 1–10 ［DOI： 10. 1109/ TIM. 2022. 3169547http://dx.doi.org/10.1109/TIM.2022.3169547］
Xie H S and Lin S F. 2021. A weakly supervised defect detection based on dual path networks and gma-cam// Proceedings of Image and Graphics. Haikou， China： Springer International Publishing： 467–478 ［DOI： 10. 1007/ 978-3-030-87355-4_39http://dx.doi.org/10.1007/978-3-030-87355-4_39］
Xie X， Li Z X， Huang Y H and Wu D Q. 2023. A weakly supervised anomaly detection method based on deep anomaly scoring network. Signal， Image and Video Processing， 17–34（8）： 3903–3911 ［DOI： 10. 1007/ s11760-023-02619-7http://dx.doi.org/10.1007/s11760-023-02619-7］
Xu J Q， Zuo Z L， Wu D C， Li B， Li X N and Kong D Y. 2021a. Bearing defect detection with unsupervised neural networks. Shock and Vibration， 2021： 1–11 ［DOI： 10. 1155/ 2021/ 9544809http://dx.doi.org/10.1155/2021/9544809］
Xu R G， Hao R Y and Huang B Q. 2022. Efficient surface defect detection using self-supervised learning strategy and segmentation network. Advanced Engineering Informatics， 52： 101566–101576 ［DOI： 10. 1016/ j. aei. 2022. 101566http://dx.doi.org/10.1016/j.aei.2022.101566］
Xu Z， Li L， Yan M， Liu J， Luo X P， Grundy J， Zhang Y F and Zhang X H. 2021b. A comprehensive comparative study of clustering-based unsupervised defect prediction models. Journal of Systems and Software， 172： 110862–110884 ［DOI： 10. 1016/ j. jss. 2020. 110862http://dx.doi.org/10.1016/j.jss.2020.110862］
Yan Y H， Ma S， Song K C， Wang Y Y， Tian H K and Guo J B. 2024. Uncertainty inspired domain adaptation network for rail surface defect segmentation. Engineering Applications of Artificial Intelligence， 135： 108860–108870 ［DOI： 10. 1016/ j. engappai. 2024. 10886 0http://dx.doi.org/10.1016/j.engappai.2024.108860］
Yang J and Qian H B. 2016. Defect prediction on unlabeled datasets by using unsupervised clustering// Proceedings of 2016 IEEE 18th International Conference on High Performance Computing and Communications. Sydney， Australia： IEEE： 465–472 ［DOI： 10. 1109/ HPCC-SmartCity-DSS. 2016. 0073http://dx.doi.org/10.1109/HPCC-SmartCity-DSS.2016.0073］
Yang J， Li S B， Wang Z， Dong H， Wang J and Tang S H. 2020. Using deep learning to detect defects in manufacturing： a comprehensive survey and current challenges. Materials， 13（24）： 5755–5778 ［DOI： 10. 3390/ ma13245755http://dx.doi.org/10.3390/ma13245755］
Yang M H， Wu P and Feng H. 2023. MemSeg： a semi-supervised method for image surface defect detection using differences and commonalities. Engineering Applications of Artificial Intelligence， 119： 105835–105845 ［DOI： 10. 1016/ j. engappai. 2023. 105835http://dx.doi.org/10.1016/j.engappai.2023.105835］
Yang P X， Zhu L， Zhang Y J， Ma C X， Liu L M， Yu X and Hu W H. 2024. on the relative value of clustering techniques for unsupervised effort-aware defect prediction. Expert Systems With Applications， 245： 123041–123065 ［DOI： 10. 1016/ j. eswa. 2023. 123041http://dx.doi.org/10.1016/j.eswa.2023.123041］
Yao K， Ortiz A and Bonnin-Pascual F. 2021. A weakly-supervised semantic segmentation approach based on the centroid loss： application to quality control and inspection. IEEE Access ： Practical Innovations， Open Solutions， 9： 69010–69026 ［DOI： 10. 1109/ access. 2021. 3077847http://dx.doi.org/10.1109/access.2021.3077847］
Ye B H， Lai J H， Xie X H and Zhu J Y. 2024. Prototype-guided domain adaptive one-stage object detector for defect detection. Advanced Engineering Informatics， 62： 102618–102631 ［DOI： 10. 1016/ j. aei. 2024. 102618http://dx.doi.org/10.1016/j.aei.2024.102618］
Yi X， Song Y H and Tang X. 2019. Weak supervised surface defect detection method based on selective search and cam// Proceedings of 2019 Chinese Automation Congress. Hangzhou， China： IEEE： 4386–4391 ［DOI： 10. 1109/ CAC48633. 2019. 8996269http://dx.doi.org/10.1109/CAC48633.2019.8996269］
Yong G， Jeon K， Gil D and Lee G. 2023. Prompt engineering for zero‐shot and few‐shot defect detection and classification using a visual‐language pretrained model. Computer-Aided Civil and Infrastructure Engineering， 38（11）： 1536–1554 ［DOI： 10. 1111/ mice. 12954http://dx.doi.org/10.1111/mice.12954］
Yu R Y and Guo B Y. 2024. Dynamic reasoning network for image-level supervised segmentation on metal surface defect. IEEE Transactions on Instrumentation and Measurement， 73： 1–10 ［DOI： 10. 1109/ TIM. 2024. 3382737http://dx.doi.org/10.1109/TIM.2024.3382737］
Yu J， Kim D Y， Lee Y and Jeon M. 2020. Unsupervised pixel-level road defect detection via adversarial image-to-frequency transform// Proceedings of 2020 IEEE Intelligent Vehicles Symposium. Las Vegas， NV， USA： IEEE： 1708–1713 ［DOI： 10. 1109/ IV47402. 2020. 9304843http://dx.doi.org/10.1109/IV47402.2020.9304843］
Yu J Y， Liu K P， Qin L， Zhao F， Wang Q L， Liu H F， Li B Q， Wang J and Li K X. 2022. DMnet： a new few-shot framework for wind turbine surface defect detection. Machines， 10（6）： 487–503 ［DOI： 10. 3390/ machines10060487http://dx.doi.org/10.3390/machines10060487］
Yu Z A， Zhang Y M， Zhang S Y， Yi G D and Qiu L M. 2024. Semi-supervised auxiliary learning for surface defect detection and segmentation of injection-molded products from small image datasets. The International Journal of Advanced Manufacturing Technology， 131（9–10）： 5243–5264 ［DOI： 10. 1007/ s00170-024- 13275-7http://dx.doi.org/10.1007/s00170-024-13275-7］
Zakharov A， Pugin E and Zakharova M. 2023. Surface defect detection based on weakly supervised learning and graph convolutional network// Proceedings of 2023 International Ural Conference on Electrical Power Engineering. Magnitogorsk， Russian Federation： IEEE： 578–582 ［DOI： 10. 1109/ UralCon59258. 2023. 102910 95http://dx.doi.org/10.1109/UralCon59258.2023.10291095］
Zhang C and Zhang X. 2023a. Multi-target domain-based hierarchical dynamic instance segmentation method for steel defects detection. Neural Computing and Applications， 35–49（10）： 7389–7406 ［DOI： 10. 1007/ s00521-022-07990-zhttp://dx.doi.org/10.1007/s00521-022-07990-z］
Zhang Z X and Zhang L M. 2023b. Unsupervised pixel-level detection of rail surface defects using multistep domain adaptation. IEEE Transactions on Systems， Man， and Cybernetics： Systems， 53（9）： 5784–5795 ［DOI： 10. 1109/ TSMC. 2023. 3274878http://dx.doi.org/10.1109/TSMC.2023.3274878］
Zhang G W， Pan Y and Zhang L M. 2021a. Semi-supervised learning with gan for automatic defect detection from images. Automation in Construction， 128： 103764–103776 ［DOI： 10. 1016/ j. autcon. 2021. 103764http://dx.doi.org/10.1016/j.autcon.2021.103764］
Zhang H， Zhang Z Q， Chen Y R， Wu T Y， Zhong H and Wang Y N. 2022. Application advance and prospect of nondestructive testing technology for industrial casting defects. Acta Automatica Sinica， 48（4）： 935-956
张辉， 张邹铨， 陈煜嵘， 吴天月， 钟杭， 王耀南. 2022. 工业铸件缺陷无损检测技术的应用进展与展望. 自动化学报， 48（4）： 935-956 ［DOI： 10. 16383/ j. aas. c210161http://dx.doi.org/10.16383/j.aas.c210161］
Zhang J B， Su H， Zou W， Gong X Y， Zhang Z T and Shen F. 2021b. CADN： a weakly supervised learning-based category-aware object detection network for surface defect detection. Pattern Recognition， 109： 107571–107581 ［DOI： 10. 1016/ j. patcog. 2020. 107571http://dx.doi.org/10.1016/j.patcog.2020.107571］
Zhang L， Yang F， Daniel Zhang Y M and Zhu Y J. 2016. Road crack detection using deep convolutional neural network// Proceedings of 2016 IEEE International Conference on Image Processing. Phoenix， AZ， USA： IEEE： 3708–3712 ［DOI： 10. 1109/ ICIP. 2016. 7533052http://dx.doi.org/10.1109/ICIP.2016.7533052］
Zhang N X， Zhong Y Z and Dian S Y. 2023a. Rethinking unsupervised texture defect detection using pca. Optics and Lasers in Engineering， 163： 107470–107481 ［DOI： 10. 1016/ j. optlaseng. 2022. 10 7470http://dx.doi.org/10.1016/j.optlaseng.2022.107470］
Zhang P C， Zheng P X， Guo X and Chen E Q. 2024a. Few-shot defect classification via feature aggregation based on graph neural network. Journal of Visual Communication and Image Representation， 101： 104172–104181
Zhang R， Zhao N， Fu L H， Bai X L and Cai J H. 2023b. Recognizing defects in stainless steel welds based on multi-domain feature expression and self-optimization. Journal of Intelligent Manufacturing， 34（3）： 1293–1309 ［DOI： 10. 1007/ s10845-021-01849-1http://dx.doi.org/10.1007/s10845-021-01849-1］
Zhang S， Ye F， Wang B N and Habetler T G. 2019a. Semi-supervised learning of bearing anomaly detection via deep variational autoencoders ［EB/OL］. ［2024-07-11］. http: //arxiv. org / pdf/ 1912. 01096. pdfhttp://arxiv.org/pdf/1912.01096.pdf
Zhang S， Ye F， Wang B N and Habetler T G. 2020. Few-shot bearing anomaly detection via model-agnostic meta-learning// Proceedings of 2020 23rd International Conference on Electrical Machines and Systems. Hamamatsu， Japan： IEEE： 1341–1346 ［DOI： 10. 23919/ ICEMS50442. 2020. 9291099http://dx.doi.org/10.23919/ICEMS50442.2020.9291099］
Zhang S Y， Su L， Gu J F， Li K， Wu W T and Pecht M. 2024b. Category-level selective dual-adversarial network using significance-augmented unsupervised domain adaptation for surface defect detection. Expert Systems With Applications， 238： 121879– 121893 ［DOI： 10. 1016/ j. eswa. 2023. 121879http://dx.doi.org/10.1016/j.eswa.2023.121879］
Zhang W Z， Li W W and Jia X Y. 2019b. Effort-aware tri-training for semi-supervised just-in-time defect prediction// Proceedings of Advances in Knowledge Discovery and Data Mining. Taipei， Taiwan： Springer International Publishing： 293–304 ［DOI： 10. 1007/ 978-3-030-16145-3_23http://dx.doi.org/10.1007/978-3-030-16145-3_23］
Zhang X Y， Zhang J P， Chen J S， Guo R X and Wu J. 2024c. A semi-supervised aircraft fuselage defect detection network with dynamic attention and class-aware adaptive pseudo-label assignment. IEEE Transactions on Artificial Intelligence， 1–14 ［DOI： 10. 1109/ TAI. 2024. 3372474http://dx.doi.org/10.1109/TAI.2024.3372474］
Zhang Y L， Wang Y L， Jiang Z Q， Zheng L， Chen J S and Lu J A. 2022. Tire defect detection by dual-domain adaptation-based transfer learning strategy. IEEE Sensors Journal， 22（19–38）： 18804–18814 ［DOI： 10. 1109/ jsen. 2022. 3201201http://dx.doi.org/10.1109/jsen.2022.3201201］
Zhang Y L， Wang Y L， Jiang Z Q， Zheng L， Chen J S and Lu J A. 2023c. Domain adaptation via transferable swin transformer for tire defect detection. Engineering Applications of Artificial Intelligence， 122： 106109–106120 ［DOI： 10. 1016/ j. engappai. 2023. 1061 09http://dx.doi.org/10.1016/j.engappai.2023.106109］
Zhang Y L， Wang Y L， Jiang Z Q， Zheng L， Chen J S and Lu J A. 2023d. A novel class-level weighted partial domain adaptation network for defect detection. Applied Intelligence， 53（20）： 23083–23096 ［DOI： 10. 1007/ s10489-023-04733-yhttp://dx.doi.org/10.1007/s10489-023-04733-y］
Zhang Z， Shang Z Q， Wang X and Ma J. 2024d. Combined anomaly aware weakly supervised lightweight model for surface defect inspection. IEEE Transactions on Industrial Informatics， 20（4）： 6652–6663 ［DOI： 10. 1109/ TII. 2023. 3348835http://dx.doi.org/10.1109/TII.2023.3348835］
Zhao S X， Zhong R Y， Wang J L， Xu C Q and Zhang J. 2023. Unsupervised fabric defects detection based on spatial domain saliency and features clustering. Computers and Industrial Engineering， 185： 109681–109693 ［DOI： 10. 1016/ j. cie. 2023. 109681http://dx.doi.org/10.1016/j.cie.2023.109681］
Zhao J Q， Qian X L， Zhang Y Z， Shan D X， Liu X Z， Coleman S and Kerr D. 2024. A knowledge distillation-based multi-scale relation-prototypical network for cross-domain few-shot defect classification. Journal of Intelligent Manufacturing， 35（2）： 841–857［DOI： 10. 1007/ s10845-023-02080-whttp://dx.doi.org/10.1007/s10845-023-02080-w］
Zhou Q H， Mei J， Zhang Q， Wang S Z and Chen G. 2021. Semi-supervised fabric defect detection based on image reconstruction and density estimation. Textile Research Journal， 91（9–10）： 962–972 ［DOI： 10. 1177/ 0040517520966733http://dx.doi.org/10.1177/0040517520966733］
Zhou Q H， Pang G S， Tian Y， He S B and Chen J M. 2024. AnomalyCLIP： object-agnostic prompt learning for zero-shot anomaly detection ［EB/OL］. ［2024-07-11］. http: //arxiv. org/ pdf/ 2310. 18961http://arxiv.org/pdf/2310.18961
Zhou C F， Liu M Q， Zhang S L， Wei P and Chen B D. 2023. Few-shot classification of screen defects with class-agnostic mask and context-based classifier. IEEE Transactions on Instrumentation and Measurement， 72： 1–16［DOI： 10. 1109/ TIM. 2023. 3280532http://dx.doi.org/10.1109/TIM.2023.3280532］
Zhu J Q， Cai S F， Deng F and Wu J R. 2024. Do llms understand visual anomalies？uncovering llm capabilities in zero-shot anomaly detection ［EB/OL］. ［2024-07-11］. http: //arxiv. org/ pdf/ 2404. 09654http://arxiv.org/pdf/2404.09654
相关文章
相关作者
相关机构
京公网安备11010802024621