Real-Time Fabric Defect Detection Using a Lightweight Deformable YOLO Network
Article Sidebar
Main Article Content
Fabric defect detection is a critical task in textile manufacturing, where manual inspection remains inconsistent, labour-intensive, and unsuitable for high-speed production environments. Although deep learning–based detectors have shown strong potential, many existing models are too computationally demanding for practical deployment in real-time industrial inspection systems. This study proposes a lightweight deformable YOLO-based framework for accurate and efficient fabric defect detection. The model is built on YOLOv5s and enhanced through three efficiency-oriented architectural improvements: Bidirectional Feature Pyramid Network (BiFPN) for improved multi-scale feature fusion, Deformable Convolutional Networks (DCNv2) for stronger geometric adaptability, and Efficient Pyramid Split Attention (EPSA) for enhanced feature discrimination. The proposed model was trained and evaluated on the Alibaba Tianchi fabric defect dataset, comprising 5,913 images across 20 defect categories. Experimental evaluation was conducted using mean Average Precision (mAP), model size, and real-time suitability, supported by ablation and comparative analyses. Results show that the proposed method improved mAP from 41.9% for the baseline YOLOv5s to 48.2%, representing a gain of 6.3 percentage points. The findings indicate that targeted architectural optimisation can improve detection accuracy while preserving the lightweight characteristics required for industrial implementation. The proposed framework offers a practical solution for automated fabric inspection and provides a useful reference for efficiency-oriented defect detection in smart manufacturing environments.
Downloads
References
C. H. Chan and G. K. H. Pang, “Fabric defect detection by Fourier analysis,” IEEE Transactions on Industry Applications, vol. 36, no. 5, pp. 1267–1276, Sep.–Oct. 2000, doi: 10.1109/28.871275.
A. Kumar, “Computer-vision-based fabric defect detection: A survey,” IEEE Transactions on Industrial Electronics, vol. 55, no. 1, pp. 348–363, Jan. 2008, doi: 10.1109/TIE.2007.896476.
J. Ngan, G. K. H. Pang, and N. H. C. Yung, “Automated fabric defect detection—A review,” Image and Vision Computing, vol. 29, no. 7, pp. 442–458, Jun. 2011, doi: 10.1016/j.imavis.2011.02.002.
Y. LeCun, Y. Bengio, and G. Hinton, “Deep learning,” Nature, vol. 521, no. 7553, pp. 436–444, May 2015, doi: 10.1038/nature14539.
H. Hanbay, M. F. Talu, and Ö. F. Özgüven, “Fabric defect detection systems and approaches,” Expert Systems with Applications, vol. 44, pp. 68–84, Jan. 2016, doi: 10.1016/j.eswa.2015.08.028.
L. Zhang, Y. Li, and X. Liu, “Deep learning-based fabric defect detection: A review,” IEEE Access, vol. 8, pp. 213855–213871, 2020, doi: 10.1109/ACCESS.2020.3040873.
R. Rasheed, M. Z. Khan, and A. U. Rehman, “A comprehensive review of fabric defect detection techniques,” Measurement, vol. 152, Art. no. 107301, Feb. 2020, doi: 10.1016/j.measurement.2019.107301.
Y. Jin, X. Liu, and J. Chen, “Fabric defect detection using convolutional neural networks,” Neural Computing and Applications, vol. 33, no. 4, pp. 1197–1210, Feb. 2021, doi: 10.1007/s00521-020-05074-7.
D. Denil, B. Shakibi, L. Dinh, M. Ranzato, and N. de Freitas, “Predicting parameters in deep learning,” in Advances in Neural Information Processing Systems (NeurIPS), Lake Tahoe, NV, USA, 2013, pp. 2148–2156.
F. Li and H. Li, “Multi-scale defect detection in complex textile textures,” Pattern Recognition Letters, vol. 147, pp. 61–68, Jun. 2021, doi: 10.1016/j.patrec.2021.03.018.
This work is licensed under a Creative Commons Attribution 4.0 International License.
All articles published in our journal are licensed under CC-BY 4.0, which permits authors to retain copyright of their work. This license allows for unrestricted use, sharing, and reproduction of the articles, provided that proper credit is given to the original authors and the source.