Journals Articles Conferences News About CLP
Submit a paper Email Alert
Search
Search
Articles
Journals
News
Advanced Search
Top Searches
2D MaterialsTransformation opticsQuantum PhotonicsSmall lasersSemiconductor UV PhotonicsSupersymmetric microring laser arrays

Laser & Optoelectronics Progress, Volume. 62, Issue 12, 1215009(2025)

Machine-Vision-Based Microtension Force Detection of Carbon Fiber Tows

Yue Ji1,2、*, Hai Zhang1,2, Jinyi Li1,2, Limei Song1,2, and Jiuzhi Dong3
Author Affiliations
  • 1School of Control Science and Engineering, Tiangong University, Tianjin 300387, China
  • 2Tianjin Key Laboratory of Intelligent Control of Electrical Equipment, Tiangong University, Tianjin 300387, China
  • 3School of Mechanical Engineering, Tiangong University, Tianjin 300387, China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (19)
    Equations (14)
    References (27)
    Tools
    Paper Information
    Topics
    Figures & Tables(19)
    Schematic diagram of vibrating string model

    Fig. 1. Schematic diagram of vibrating string model

    Download full size
    Schematic diagram of sagging string model

    Fig. 2. Schematic diagram of sagging string model

    Download full size
    Carbon fiber image classification solving process

    Fig. 3. Carbon fiber image classification solving process

    Download full size
    Optical inspection of carbon fiber tows

    Fig. 4. Optical inspection of carbon fiber tows

    Download full size
    Carbon fiber tows preprocessing images. (a)(b) Splicing; (c)(d) threshold segmentation; (e)(f) boundary extraction

    Fig. 5. Carbon fiber tows preprocessing images. (a)(b) Splicing; (c)(d) threshold segmentation; (e)(f) boundary extraction

    Download full size
    S-G smoothing filtering

    Fig. 6. S-G smoothing filtering

    Download full size
    Schematic diagram of filament bundle strip pooling module

    Fig. 7. Schematic diagram of filament bundle strip pooling module

    Download full size
    Improved ResNet18 network structure

    Fig. 8. Improved ResNet18 network structure

    Download full size
    Loss value curves for improved ResNet18 model

    Fig. 9. Loss value curves for improved ResNet18 model

    Download full size
    Carbon fiber tows tension experimental platform

    Fig. 10. Carbon fiber tows tension experimental platform

    Download full size
    Carbon fiber tows vibration frequency and tension fitting curve

    Fig. 11. Carbon fiber tows vibration frequency and tension fitting curve

    Download full size
    Force analysis diagram of three-roller tension sensor

    Fig. 12. Force analysis diagram of three-roller tension sensor

    Download full size
    Carbon fiber tows sag height and tension fitting curve

    Fig. 13. Carbon fiber tows sag height and tension fitting curve

    Download full size
    Carbon fiber tows microtension comparison experiment results

    Fig. 14. Carbon fiber tows microtension comparison experiment results

    Download full size
    Sensor measurements and machine vision measurements at preset value of 2 cN

    Fig. 15. Sensor measurements and machine vision measurements at preset value of 2 cN

    Download full size

    • Table 1. Performance comparison of different image classification models

      View table

      Table 1. Performance comparison of different image classification models

      ModelTraining set accuracy /%Validation set accuracy /%Precision /%Recall /%Recognition time /ms
      SVM85.284.584.584.51.5
      Traditional CNN89.889.589.589.67.1
      ResNet1890.388.291.285.23.3
      Improved ResNet1899.198.799.198.93.5

    • Table 2. Image classification result

      View table

      Table 2. Image classification result

      Image classificationVibration imageSagging image
      3 cN and above image983
      3 cN following image297
      Total100100

    • Table 3. Comparison of relative errors of dual model and sensor measurements

      View table

      Table 3. Comparison of relative errors of dual model and sensor measurements

      Sensor measured value /cNVibration model relative error /%Sagging model relative error /%
      1.84-56.47-9.24
      2.00-45.735.50
      2.28-33.78-7.02
      2.35-48.12-2.98
      2.40-26.75-4.58
      2.54-52.42-3.94
      2.61-29.651.92
      2.77-36.41-3.61
      2.89-31.22-4.84
      2.99-27.38-3.68

    • Table 4. Relative error of tension for different wire materials

      View table

      Table 4. Relative error of tension for different wire materials

      Line density /g/kmSensor measured value /cNMachine vision measured value /cNRelative error /%
      2251.591.52-4.40
      1.731.60-7.51
      1.881.79-4.79
      2.031.98-2.46
      2.252.21-1.78
      2.382.432.10
      2.472.511.62
      2.662.62-1.50
      2.842.78-2.11
      3.062.96-3.27
      5001.841.67-9.24
      2.002.115.50
      2.282.12-7.02
      2.352.28-2.98
      2.402.29-4.58
      2.542.44-3.94
      2.612.661.92
      2.772.67-3.61
      2.892.75-4.84
      2.992.88-3.68
      10001.561.54-1.28
      1.751.69-3.43
      1.931.88-2.59
      2.082.05-1.44
      2.252.09-7.11
      2.282.21-3.07
      2.582.37-8.14
      2.672.55-4.49
      2.762.863.62
      2.982.89-3.02
    Tools

    Get Citation

    Copy Citation Text

    Yue Ji, Hai Zhang, Jinyi Li, Limei Song, Jiuzhi Dong. Machine-Vision-Based Microtension Force Detection of Carbon Fiber Tows[J]. Laser & Optoelectronics Progress, 2025, 62(12): 1215009

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Set citation alerts for article
    Save article for my favorites
    Paper Information

    Category: Machine Vision

    Received: Dec. 3, 2024

    Accepted: Jan. 6, 2025

    Published Online: Jun. 12, 2025

    The Author Email: Yue Ji (jiyue@tiangong.edu.cn)

    DOI:10.3788/LOP242369

    CSTR:32186.14.LOP242369

    Topics

    laser devices and laser physics
    Lasers and Laser Optics
    Laser physics
    laser manufacturing
    Instrumentation, Measurement and Metrology