Machine Learning for Fabric Type Recognition in Sewing Processes

Palvannazirova, Nasiba; Butovskiy, Aybek

doi:10.25673/123515

Proceedings of International Conference on Applied Innovation in IT · 2026/03/31 · Vol. 14 · Issue 1 · pp. 41–47

Machine Learning for Fabric Type Recognition in Sewing Processes

Nasiba Palvannazirova, Aybek Mavlyanov and Petr Butovskiy

📄 Download PDF DOI: 10.25673/123515

Abstract

This scientific article presents an artificial intelligence system for the automatic identification of types of fabric in the garment industry. The system differentiates between six categories of materials: cotton, polyester, silk, wool, denim and mixed fabrics. The development of the system was based on the pre-training neural network VGG16, which was modified to fit the data of textiles. On the other hand, actual production situations were taken into consideration, e.g. variable lighting in workshops, varied material textures. To train the system, a data set of 540 images was created, each of which were created under different illumination and in different angles to improve the reliability of the recognition. The training process consisted of three different stages. First of all, it was necessary to configure the classifier for the particular categories of fabric. Subsequently, further layers of neural network were unfrozen for subsequent further tuning. In the last stage, the entire network went through retraining. Consequently, the accuracy of the recognition achieved 95.8%. The researchers also tested light versions of the system, with different architectures available, on a Raspberry Pi 4 mini computer. This way we were able to determine the best balance between accuracy of recognition and speed of real-time operation. The results showed neural networks are capable of recognising fabrics at an almost human level. At the same time, the system can be optimised to operate on small devices directly in the sewing workshops. Thus work that combines scientific developments, practical applications into industrial use and opens the way to full automation of sewing production.

Keywords

Fabric Recognition VGG16 Transfer Learning Edge Computing Sewing Automation Raspberry Pi 4 Deep Learning.

References

V. Hariharan and K. Sivaraman, “Machine Learning Techniques for Quality Control in Textile Fabric Manufacturing,” Proc. IEEE ICSC-SmartSys, pp. 902-910, 2024, doi:10.1109/ICSC-SmartSys.2024.1035128.
K. Kailasam et al., “Fabric Defect Detection Using Deep Learning,” Proc. IEEE ICSADL, pp. 399-405, 2024, doi:10.1109/ICSADL.2024.1042173.
S. K. Subramanian et al., “AI in Textile Quality Control: An Overview of Industrial Applications,” IEEE Transactions on Industrial Informatics, vol. 20, no. 2, pp. 1123-1135, 2025, doi:10.1109/TII.2025.1047629.
J.-Y. Lee et al., “Study on Sensing and Monitoring of Sewing Machine for Textile Stream Smart Manufacturing Innovation,” Proc. IEEE M2VIP, pp. 1-6, 2025, doi:10.1109/M2VIP.2025.1053927.
J. H. Park et al., “Machine Learning and Cyber-Physical Systems in Smart Textile Manufacturing,” Proc. IEEE ICSS, pp. 120-128, 2024, doi:10.1109/ICSS.2024.1049802.
S. B. M. Shabi et al., “Artificial Intelligence, Intelligent Manufacturing, and Sustainable Futures in Next-Gen Textiles: A Comprehensive Approach,” Proc. IEEE IDCIoT, pp. 2144-2150, 2025, doi:10.1109/IDCIoT.2025.1065812.
J. L. Park et al., “Fixture-Free Automated Sewing System Using Dual-Arm Manipulator and High-Speed Fabric Edge Detection,” IEEE Transactions on Automation Science and Engineering, vol. 21, no. 4, pp. 234-241, 2023, doi:10.1109/TASE.2023.1018467.
L. Kumar et al., “A Fabric Defect Detection Method Based on Deep Learning,” IEEE Access, vol. 10, pp. 4284-4296, 2023, doi:10.1109/ACCESS.2023.3246752.
A. Ahmed et al., “Classification of Pineapple Fiber Woven Fabrics on Raspberry Pi Based on Convolutional Neural Network,” Proc. IEEE ECA, pp. 221-227, 2023, doi:10.1109/ECA.2023.1024671.
P. M. Butovskiy et al., “Textiles 5.0: Advanced Intelligent Systems for Automated Manufacturing,” Textiles, vol. 5, no. 15, pp. 1-12, 2025, doi:10.3390/textiles5010015.
R. O. Meza Torres et al., “Design of a Machine for the Production of Advanced Fabrics,” Proc. IEEE UEMCON, pp. 731-740, 2024, doi:10.1109/UEMCON.2024.1067289.
A. G. Song et al., “A Novel Texture Sensor for Fabric Texture Measurement and Classification,” IEEE Transactions on Instrumentation and Measurement, vol. 63, no. 7, pp. 1739-1748, 2024, doi:10.1109/TIM.2013.2293812.