Abstract
Fabric defect detection can greatly enhance the quality of fabric production. However, the high cost of annotating defects and the computational complexity of networks are the main challenges in defect detection. To address these challenges, this paper proposes a semi-supervised lightweight fabric defect detection algorithm (SDA-Net). During the semi-supervised training process, the algorithm uses labeled defect samples and normal samples to learn latent features and detect defect positions accurately. First, to solve the issue of insufficient labeled defect samples due to high annotation costs, a data augmentation method called Sel-fill is proposed. The Sel-fill randomly samples image blocks of various sizes from a truncated normal distribution. These image blocks are then inserted into random positions within normal images, thereby generating labeled defect samples. Second, A lightweight neural network architecture is constructed using depth-wise separable convolution (DSConv). This architecture effectively reduces the number of parameters and computations while maintaining performance. Final, the max pooling coordinate attention mechanism (MpCA) effectively suppresses background noise during the multi-scale feature fusion process, resulting in improved detection precision. By using depth-wise separable convolution and MpCA attention, SDA-Net achieves an average detection precision of 62.6%, improved by 4.5% over the previous method. The number of trainable parameters is only 9.35 MB, reduced by 42.53%. Moreover, the computations are reduced by 68.84%.
Supported by the Basic Science (Natural Science) Research Projects of Universities in Jiangsu Province (Grant No.22KJB520011).
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Bao, X., Liang, J., Xia, Y., Hou, Z., Huan, Z.: Low-rank decomposition fabric defect detection based on prior and total variation regularization. Vis. Comput. 38(8), 2707–2721 (2022)
Cheng, L., Yi, J., Chen, A., Zhang, Y.: Fabric defect detection based on separate convolutional UNet. Multimedia Tools Appl. 82(2), 3101–3122 (2023)
Gu, M., Zhou, J., Pan, R., Gao, W.: Unsupervised defect segmentation on denim fabric via local patch prediction and residual fusion. Text. Res. J. 93(15–16), 3573–3587 (2023)
Hou, Q., Zhou, D., Feng, J.: Coordinate attention for efficient mobile network design. Comput. Vision Pattern Recogn. 13713–13722 (2021)
Hu, G., Huang, J., Wang, Q., Li, J., Xu, Z., Huang, X.: Unsupervised fabric defect detection based on a deep convolutional generative adversarial network. Text. Res. J. 90(3–4), 247–270 (2020)
Ji, X., Liang, J., Di, L., Xia, Y., Hou, Z., Huan, Z., Huan, Y.: Fabric defect fetection via weighted low-rank decomposition and Laplacian regularization. J. Eng. Fibers Fabr. 15(5), 1558925020957654 (2020)
Jing, J., Wang, Z., Rätsch, M., Zhang, H.: Mobile-Unet: an efficient convolutional neural network for fabric defect detection. Text. Res. J. 92(1–2), 30–42 (2022)
Khanzhina, N., Kashirin, M., Filchenkov, A.: Monte Carlo concrete DropPath for epistemic uncertainty estimation in brain tumor segmentation. In: Lecture Notes in Computer Science, pp. 64–74 (2021)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. Adv. Neural. Inf. Process. Syst. 25(2), 1097–1105 (2012)
Kumar, D.D., Fang, C., Zheng, Y., Gao, Y.: Semi-supervised transfer learning-based automatic weld defect detection and visual inspection. Eng. Struct. 292(10–1), 116580 (2023)
Lawhern, V.J., Solon, A.J., Waytowich, N.R., Gordon, S.M., Hung, C.P., Lance, B.J.: EEGNet: a compact convolutional neural network for EEG-based brain-computer interfaces. J. Neural Eng. 15(5), 056013 (2018)
Li, C.L., Sohn, K., Yoon, J., Pfister, T.: Cutpaste: Self-supervised learning for anomaly detection and localization. Comput. Vision Pattern Recogn. 9664–9674 (2021)
Li, C., Gao, G., Liu, Z., Huang, D., Xi, J.: Defect detection for patterned fabric images based on GHOG and low-rank decomposition. IEEE Access 7(99), 83962–83973 (2019)
Li, Z., Tian, X., Liu, X., Liu, Y., Shi, X.: A two-stage industrial defect detection framework based on improved-yolov5 and optimized-inception-resnetv2 models. Appl. Sci. 12(2), 834 (2022)
Liu, B., Wang, H., Cao, Z., Wang, Y., Tao, L., Yang, J., Zhang, K.: PRC-Light YOLO: an efficient lightweight model for fabric defect detection. Appl. Sci. 14(2), 938 (2024)
Peng, P., Wang, Y., Hao, C., Zhu, Z., Liu, T., Zhou, W.: Automatic fabric defect detection method using PRAN-net. Appl. Sci. 10(23), 8434 (2020)
Ren, M., Shen, R., Gong, Y.: A surface defect detection method via fusing multi-level features. J. Comput. Inf. Sci. Eng. 22(5), 051005 (2022)
Shi, W., Wang, W., Zhu, L., Wu, K., Wu, J.: Clustering-Based Cycle Gan for Fabric Defect Detection. Social Science Electronic Publishing (2022)
Tang, S., Jin, Z., Zhang, Y., Lu, J., Li, H., Yang, J.: A timestep-adaptive-diffusion-model-oriented unsupervised detection method for fabric surface defects. Processes 11(9), 2615 (2023)
Wang, Y., Luo, S., Wu, H.: Retracted: Defect detection of solar cell based on data augmentation. J. Phys. Conf. Ser. 1952, 022010 (2021)
Wei, C., Liang, J., Liu, H., Hou, Z., Huan, Z.: Multi-stage unsupervised fabric defect detection based on DCGAN. Vis. Comput. 39(12), 6655–6671 (2023)
Xiao, H., Zhao, C., Zhang, Z., et al.: A semi-supervised method for steel surface defect detection based on soft-teacher. J. Comput. Inf. Sci. Eng. 6(3), 11–19 (2023)
Yang, M., Wu, P., Feng, H.: Memseg: A semi-supervised method for image surface defect detection using differences and commonalities. Eng. Appl. Artif. Intell. 119, 105835 (2023)
Yao, H., Yu, W., Wang, X.: A feature memory rearrangement network for visual inspection of textured surface defects toward edge intelligent manufacturing. IEEE Trans. Autom. Sci. Eng. (2022)
Yi, C., Xu, B., Chen, J., Chen, Q., Zhang, L.: An improved yolox model for detecting strip surface defects. Steel Res. Int. 93(11), 2200505 (2022)
Zavrtanik, V., Kristan, M., Skočaj, D.: Draem-a discriminatively trained reconstruction embedding for surface anomaly detection. Comput. Vis. Pattern Recogn. 8330–8339 (2021)
Zhang, H., Tan, Q., Lu, S., Ge, Z., Gu, D.: Yarn-dyed fabric defect detection using u-shaped de-noising convolutional auto-encoder. In: 2020 IEEE 9th Data Driven Control and Learning Systems Conference, pp. 18–24 (2020)
Zhao, S., Yin, L., Zhang, J., Wang, J., Zhong, R.: Real-time fabric defect detection based on multi-scale convolutional neural network. IET Collab. Intell. Manuf. 2(4), 189–196 (2020)
Zhou, K., Deng, K., Chen, P., Hu, Y.: An improved lightweight network based on mobilenetv3 for palmprint recognition. In: Chinese Conference on Pattern Recognition and Computer Vision, pp. 749–761 (2022)
Zhou, Q., Mei, J., Zhang, Q., Wang, S., Chen, G.: Semi-supervised fabric defect detection based on image reconstruction and density estimation. Text. Res. J. 91(9–10), 962–972 (2021)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2025 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Dong, X., Liu, H., Luo, Y., Yan, Y., Liang, J. (2025). Semi-supervised Lightweight Fabric Defect Detection. In: Lin, Z., et al. Pattern Recognition and Computer Vision. PRCV 2024. Lecture Notes in Computer Science, vol 15034. Springer, Singapore. https://doi.org/10.1007/978-981-97-8505-6_8
Download citation
DOI: https://doi.org/10.1007/978-981-97-8505-6_8
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-97-8504-9
Online ISBN: 978-981-97-8505-6
eBook Packages: Computer ScienceComputer Science (R0)