skip to main content
10.1145/3573428.3573774acmotherconferencesArticle/Chapter ViewAbstractPublication PageseitceConference Proceedingsconference-collections
research-article

An Improved Faster RCNN based on Swin Transformer for Surface Defect Detection of Metal Workpieces

Published: 15 March 2023 Publication History

Abstract

Surface defects are an inevitable problem in the production process of metal workpieces. Computer vision based surface defect detection methods outperform manual inspection methods and are gradually being applied in industrial production. In this paper, we proposed an improved Faster RCNN model for metal workpiece surface defects to tackle the problems of large size variation and many morphological changes in surface defects. Swin Transformer is used as the backbone to enhance the feature extraction ability, and FPN is introduced to carry out multi-scale feature fusion on the feature maps from its four stages. Focal loss is adopted as the classification loss to further improve the detection performance. Experimental results on the NEU-DET dataset show that the proposed model improves the mAP by 4.24% over the Faster RCNN with ResNet50 as its backbone, and the training process converges faster.

References

[1]
Tang Bo, Kong Jianyi, Wu Shiqian. Review of surface defect detection based on machine vision [J]. Journal of Image and Graphics, 2017, 22(12): 1640–1663.
[2]
Wang Qian, Kim Min-Koo, Sohn Hoon, Surface flatness and distortion inspection of precast concrete elements using laser scanning technology[J]. Smart Structures and Systems, 2016, 18(3): 601–623.
[3]
Zhou X, Wang Y, Zhu Q, A Surface Defect Detection Framework for Glass Bottle Bottom Using Visual Attention Model and Wavelet Transform[J]. IEEE Transactions on Industrial Informatics, 2020, 16(4): 2189–2201.
[4]
Li Xudong, Ye Mao, Li Tao. Review of object detection based on convolutional neural networks[J]. Application Research of Computers, 2017, 34(10): 2881-2886+2891.
[5]
Ren S, He K, Girshick R, Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137–1149.
[6]
Redmon J, Farhadi A. YOLOv3: An Incremental Improvement[J]. arXiv, 2018.
[7]
Han K, Wang Y, Chen H, A Survey on Vision Transformer[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022: 1–1.
[8]
Liu Z, Lin Y, Cao Y, Swin Transformer: Hierarchical Vision Transformer using Shifted Windows[J]. arXiv, 2021.
[9]
Lin T-Y, Dollar P, Girshick R, Feature Pyramid Networks for Object Detection[A]. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)[C]. Honolulu, HI: IEEE, 2017: 936–944.
[10]
Lin T-Y, Goyal P, Girshick R, Focal Loss for Dense Object Detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(2): 318–327.
[11]
He Y, Song K, Meng Q, An End-to-End Steel Surface Defect Detection Approach via Fusing Multiple Hierarchical Features[J]. IEEE Transactions on Instrumentation and Measurement, 2020, 69(4): 1493–1504.
[12]
Wang Yongsheng, Ji Siyu, Du Binbin. Marine impact water column signal detection algorithm based on improved Faster R-CNN[J]. Journal of Ordnance Equipment Engineering, 2022, 43(06): 182–189.

Cited By

View all
  • (2024)A shunted-swin transformer for surface defect detection in roller bearingsMeasurement10.1016/j.measurement.2024.115283238(115283)Online publication date: Oct-2024
  • (2022)A New Convolutional Neural Network for Identification of Damaged Electronic Components2022 6th International Conference on Universal Village (UV)10.1109/UV56588.2022.10185470(1-6)Online publication date: 22-Oct-2022

Index Terms

  1. An Improved Faster RCNN based on Swin Transformer for Surface Defect Detection of Metal Workpieces

    Recommendations

    Comments

    Information & Contributors

    Information

    Published In

    cover image ACM Other conferences
    EITCE '22: Proceedings of the 2022 6th International Conference on Electronic Information Technology and Computer Engineering
    October 2022
    1999 pages
    ISBN:9781450397148
    DOI:10.1145/3573428
    Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

    Publisher

    Association for Computing Machinery

    New York, NY, United States

    Publication History

    Published: 15 March 2023

    Permissions

    Request permissions for this article.

    Check for updates

    Author Tags

    1. Deep learning
    2. Faster RCNN
    3. Surface defect detection
    4. Swin Transformer

    Qualifiers

    • Research-article
    • Research
    • Refereed limited

    Conference

    EITCE 2022

    Acceptance Rates

    Overall Acceptance Rate 508 of 972 submissions, 52%

    Contributors

    Other Metrics

    Bibliometrics & Citations

    Bibliometrics

    Article Metrics

    • Downloads (Last 12 months)41
    • Downloads (Last 6 weeks)6
    Reflects downloads up to 13 Feb 2025

    Other Metrics

    Citations

    Cited By

    View all
    • (2024)A shunted-swin transformer for surface defect detection in roller bearingsMeasurement10.1016/j.measurement.2024.115283238(115283)Online publication date: Oct-2024
    • (2022)A New Convolutional Neural Network for Identification of Damaged Electronic Components2022 6th International Conference on Universal Village (UV)10.1109/UV56588.2022.10185470(1-6)Online publication date: 22-Oct-2022

    View Options

    Login options

    View options

    PDF

    View or Download as a PDF file.

    PDF

    eReader

    View online with eReader.

    eReader

    HTML Format

    View this article in HTML Format.

    HTML Format

    Figures

    Tables

    Media

    Share

    Share

    Share this Publication link

    Share on social media