IDEAS home Printed from https://ideas.repec.org/a/spr/joinma/v35y2024i1d10.1007_s10845-022-02051-7.html
   My bibliography  Save this article

Cross-scale fusion and domain adversarial network for generalizable rail surface defect segmentation on unseen datasets

Author

Listed:
  • Shuai Ma

    (Northeastern University
    Northeastern University
    Key Laboratory of Data Analytics and Optimization for Smart Industry (Northeastern University))

  • Kechen Song

    (Northeastern University
    Northeastern University
    Key Laboratory of Data Analytics and Optimization for Smart Industry (Northeastern University))

  • Menghui Niu

    (Northeastern University
    Northeastern University
    Key Laboratory of Data Analytics and Optimization for Smart Industry (Northeastern University))

  • Hongkun Tian

    (Northeastern University
    Northeastern University
    Key Laboratory of Data Analytics and Optimization for Smart Industry (Northeastern University))

  • Yunhui Yan

    (Northeastern University
    Northeastern University
    Key Laboratory of Data Analytics and Optimization for Smart Industry (Northeastern University))

Abstract

Surface quality control is a crucial part of rail manufacturing. Deep neural networks have shown impressive accuracy in rail surface defect segmentation under the assumption that the test images have the same distribution as the training images. However, in practice detection, the rail images exhibit variations in appearance and scale for different rail types and production conditions. Directly deploying the deep neural network on unseen images shows a performance degradation due to the distribution discrepancies of training images. To this end, we propose a cross-scale fusion and domain adversarial network (CFDANet) to improve the generalization ability of deep neural networks on unseen datasets. To alleviate the domain shift caused by defect scale differences, we design a dual-encoder to extract multi-scale features from images of different resolutions. Then, those features are adaptively fused through a cross-scale fusion module. For the domain shift caused by inconsistent rail appearance, we introduce transferable-aware domain adversarial learning to extract domain invariant features from different datasets. Moreover, we further propose a transferable curriculum to suppress the negative impact of images with low transferability. Experimental results show that our CFDANet can accurately segment defects in unseen datasets and surpass other state-of-the-art domain generalization methods in all five target domain settings. The source code is released at https://github.com/dotaball/railseg_dg .

Suggested Citation

  • Shuai Ma & Kechen Song & Menghui Niu & Hongkun Tian & Yunhui Yan, 2024. "Cross-scale fusion and domain adversarial network for generalizable rail surface defect segmentation on unseen datasets," Journal of Intelligent Manufacturing, Springer, vol. 35(1), pages 367-386, January.
  • Handle: RePEc:spr:joinma:v:35:y:2024:i:1:d:10.1007_s10845-022-02051-7
    DOI: 10.1007/s10845-022-02051-7
    as

    Download full text from publisher

    File URL: http://link.springer.com/10.1007/s10845-022-02051-7
    File Function: Abstract
    Download Restriction: Access to the full text of the articles in this series is restricted.

    File URL: https://libkey.io/10.1007/s10845-022-02051-7?utm_source=ideas
    LibKey link: if access is restricted and if your library uses this service, LibKey will redirect you to where you can use your library subscription to access this item
    ---><---

    As the access to this document is restricted, you may want to search for a different version of it.

    References listed on IDEAS

    as
    1. Mohamed Ben Gharsallah & Ezzedine Ben Braiek, 2021. "Computer aided manufacturing method for surface silicon steel inspection based on an efficient anisotropic diffusion algorithm," Journal of Intelligent Manufacturing, Springer, vol. 32(4), pages 1025-1041, April.
    2. Ruiyang Hao & Bingyu Lu & Ying Cheng & Xiu Li & Biqing Huang, 2021. "A steel surface defect inspection approach towards smart industrial monitoring," Journal of Intelligent Manufacturing, Springer, vol. 32(7), pages 1833-1843, October.
    3. Olatomiwa Badmos & Andreas Kopp & Timo Bernthaler & Gerhard Schneider, 2020. "Image-based defect detection in lithium-ion battery electrode using convolutional neural networks," Journal of Intelligent Manufacturing, Springer, vol. 31(4), pages 885-897, April.
    Full references (including those not matched with items on IDEAS)

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. José M. Navarro-Jiménez & José V. Aguado & Grégoire Bazin & Vicente Albero & Domenico Borzacchiello, 2023. "Reconstruction of 3D surfaces from incomplete digitisations using statistical shape models for manufacturing processes," Journal of Intelligent Manufacturing, Springer, vol. 34(5), pages 2345-2358, June.
    2. Xinyu Suo & Jian Liu & Licheng Dong & Chen Shengfeng & Lu Enhui & Chen Ning, 2022. "A machine vision-based defect detection system for nuclear-fuel rod groove," Journal of Intelligent Manufacturing, Springer, vol. 33(6), pages 1649-1663, August.
    3. Li Wei & Mahmud Iwan Solihin & Sarah ‘Atifah Saruchi & Winda Astuti & Lim Wei Hong & Ang Chun Kit, 2024. "Surface Defects Detection of Cylindrical High-Precision Industrial Parts Based on Deep Learning Algorithms: A Review," SN Operations Research Forum, Springer, vol. 5(3), pages 1-71, September.
    4. Pan, Yue & Kong, Xiangdong & Yuan, Yuebo & Sun, Yukun & Han, Xuebing & Yang, Hongxin & Zhang, Jianbiao & Liu, Xiaoan & Gao, Panlong & Li, Yihui & Lu, Languang & Ouyang, Minggao, 2023. "Detecting the foreign matter defect in lithium-ion batteries based on battery pilot manufacturing line data analyses," Energy, Elsevier, vol. 262(PB).
    5. Seunghwan Jung & Minseok Kim & Eunkyeong Kim & Baekcheon Kim & Jinyong Kim & Kyeong-Hee Cho & Hyang-A Park & Sungshin Kim, 2024. "The Early Detection of Faults for Lithium-Ion Batteries in Energy Storage Systems Using Independent Component Analysis with Mahalanobis Distance," Energies, MDPI, vol. 17(2), pages 1-23, January.
    6. Zeqing Yang & Mingxuan Zhang & Yingshu Chen & Ning Hu & Lingxiao Gao & Libing Liu & Enxu Ping & Jung Il Song, 2024. "Surface defect detection method for air rudder based on positive samples," Journal of Intelligent Manufacturing, Springer, vol. 35(1), pages 95-113, January.
    7. Chia-Yu Hsu & Ju-Chien Chien, 2022. "Ensemble convolutional neural networks with weighted majority for wafer bin map pattern classification," Journal of Intelligent Manufacturing, Springer, vol. 33(3), pages 831-844, March.
    8. Yuanyuan Wang & Ling Ma & Lihua Jian & Huiqin Jiang, 2023. "Conductive particle detection via efficient encoder–decoder network," Journal of Intelligent Manufacturing, Springer, vol. 34(8), pages 3563-3577, December.
    9. Swarit Anand Singh & K. A. Desai, 2023. "Automated surface defect detection framework using machine vision and convolutional neural networks," Journal of Intelligent Manufacturing, Springer, vol. 34(4), pages 1995-2011, April.
    10. Zhuxi Ma & Yibo Li & Minghui Huang & Qianbin Huang & Jie Cheng & Si Tang, 2023. "Automated real-time detection of surface defects in manufacturing processes of aluminum alloy strip using a lightweight network architecture," Journal of Intelligent Manufacturing, Springer, vol. 34(5), pages 2431-2447, June.
    11. Ruiyang Hao & Bingyu Lu & Ying Cheng & Xiu Li & Biqing Huang, 2021. "A steel surface defect inspection approach towards smart industrial monitoring," Journal of Intelligent Manufacturing, Springer, vol. 32(7), pages 1833-1843, October.
    12. Minyoung Lee & Joohyoung Jeon & Hongchul Lee, 2022. "Explainable AI for domain experts: a post Hoc analysis of deep learning for defect classification of TFT–LCD panels," Journal of Intelligent Manufacturing, Springer, vol. 33(6), pages 1747-1759, August.
    13. Feiyang Li & Nian Cai & Xueliang Deng & Jiahao Li & Jianfa Lin & Han Wang, 2022. "Serial number inspection for ceramic membranes via an end-to-end photometric-induced convolutional neural network framework," Journal of Intelligent Manufacturing, Springer, vol. 33(5), pages 1373-1392, June.
    14. Simon Müller & Christina Sauter & Ramesh Shunmugasundaram & Nils Wenzler & Vincent Andrade & Francesco Carlo & Ender Konukoglu & Vanessa Wood, 2021. "Deep learning-based segmentation of lithium-ion battery microstructures enhanced by artificially generated electrodes," Nature Communications, Nature, vol. 12(1), pages 1-12, December.
    15. Omid Davtalab & Ali Kazemian & Xiao Yuan & Behrokh Khoshnevis, 2022. "Automated inspection in robotic additive manufacturing using deep learning for layer deformation detection," Journal of Intelligent Manufacturing, Springer, vol. 33(3), pages 771-784, March.
    16. Saksham Jain & Gautam Seth & Arpit Paruthi & Umang Soni & Girish Kumar, 2022. "Synthetic data augmentation for surface defect detection and classification using deep learning," Journal of Intelligent Manufacturing, Springer, vol. 33(4), pages 1007-1020, April.
    17. Nhat-To Huynh & Duong-Dong Ho & Hong-Nguyen Nguyen, 2023. "An Approach for Designing an Optimal CNN Model Based on Auto-Tuning GA with 2D Chromosome for Defect Detection and Classification," Sustainability, MDPI, vol. 15(6), pages 1-14, March.

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:spr:joinma:v:35:y:2024:i:1:d:10.1007_s10845-022-02051-7. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Sonal Shukla or Springer Nature Abstracting and Indexing (email available below). General contact details of provider: http://www.springer.com .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.