YOLOv8 and ResNet-50 Based Real-Time Fabric Defect Detection and Quality Grading System
Abstract
Automated fabric defect inspection is essential to address the low speed and inefficiency of manual inspection, which often requires machine stoppages to record defects. In contrast, this study presents a continuous, real-time defect detection and grading system based on the 4-point standard, integrating computer vision and deep learning techniques. A dataset of 1,456 annotated fabric images was developed from captured production videos. The proposed method employs ResNet-50 as the backbone for feature extraction and YOLOv8 for object detection, followed by defect quantification to determine fabric grades. Experimental evaluation shows that the model achieves a precision of 0.789, recall of 0.554, mAP50 of 0.675, mAP50-95 of 0.461, and an F1-score of 0.61. The class-wise mAP50 scores are 0.744 for line defects, 0.839 for stain defects, and 0.442 for hole defects. While the integration of convolutional neural networks with automated grading logic enhances detection accuracy and reliability in textile quality control, the relatively low performance on hole defects indicates the impact of dataset imbalance, as this category contained significantly fewer samples. Addressing this limitation through targeted data augmentation or additional data collection is recommended to improve generalization and achieve balanced detection performance across defect categories.
Full Text:
PDFReferences
M. Ateeq, A. P. P. Abdul Majeed, H. Hafizh, M. A. Mohd Razman, I. Mohd Khairuddin, and N. H. Noordin, “A Feature-Based Transfer Learning Method for Surface Defect Detection in Smart Manufacturing,” in Intelligent Manufacturing and Mechatronics, W. H. Mohd. Isa, I. Mohd. Khairuddin, Mohd. A. Mohd. Razman, S. ’Atifah Saruchi, S.-H. Teh, and P. Liu, Eds., Singapore: Springer Nature Singapore, 2024, pp. 455–461.
K. Gopalakrishnan and P. T. Vanathi, “Fabric Defect Detection Using Deep Learning Techniques,” in Ubiquitous Intelligent Systems, P. Karuppusamy, F. P. García Márquez, and T. N. Nguyen, Eds., Singapore: Springer Nature Singapore, 2022, pp. 101–113.
F. K. Konstantinidis, I. Kansizoglou, K. A. Tsintotas, S. G. Mouroutsos, and A. Gasteratos, “The role of machine vision in industry 4.0: A textile manufacturing perspective,” in IST 2021 - IEEE International Conference on Imaging Systems and Techniques, Proceedings, 2021. doi: 10.1109/IST50367.2021.9651459.
Akansha Priya Gandham, “ailed,” Oct. 2024. Accessed: Dec. 17, 2024. [Online]. Available: https://www.alliedmarketresearch.com/press-release/digital-clothing-market.html
D. J. Hemanth, V. D. A. Kumar, S. Malathi, O. Castillo, and B. Patrut, “Emerging Trends in Computing and Expert Technology,” 2019.
M. P. Sikka, A. Sarkar, and S. Garg, “Artificial intelligence (AI) in textile industry operational modernization,” 2024. doi: 10.1108/RJTA-04-2021-0046.
Y. A. Karayiannis et al., “Defect detection and classification on web textile fabric using multiresolution decomposition and neural networks,” in Proceedings of the IEEE International Conference on Electronics, Circuits, and Systems, 1999. doi: 10.1109/ICECS.1999.813221.
E. Shady, Y. Gowayed, M. Abouiiana, S. Youssef, and C. Pastore, “Detection and Classification of Defects in Knitted Fabric Structures,” Textile Research Journal, vol. 76, no. 4, 2006, doi: 10.1177/0040517506053906.
C. H. Chan, “Fabric defect detection by Fourier analysis,” IEEE Trans Ind Appl, vol. 36, no. 5, 2000, doi: 10.1109/28.871274.
S. Das, M. K. Selvaraj, S. Shanmugavelu, and S. Jayaram, “Deep learning neural networks for knitted fabric defect identification and classification,” Man-Made Textiles in India, vol. 50, no. 12, 2022.
L. Shi, J. Song, Y. Gao, G. Cheng, and B. Hao, “YOLO-GFD: A Fast and Accurate Fabric Defect Detection Model,” in 2023 4th International Conference on Big Data, Artificial Intelligence and Internet of Things Engineering, ICBAIE 2023, 2023. doi: 10.1109/ICBAIE59714.2023.10281292.
M. Li, D. Pi, and S. Qin, “An efficient single shot detector with weight-based feature fusion for small object detection,” Sci Rep, vol. 13, no. 1, p. 9883, 2023, doi: 10.1038/s41598-023-36972-x.
S. N. Appe, G. Arulselvi, and G. N. Balaji, “CAM-YOLO: tomato detection and classification based on improved YOLOv5 using combining attention mechanism,” PeerJ Comput Sci, vol. 9, 2023, doi: 10.7717/peerj-cs.1463.
Y. Liu, P. Sun, N. Wergeles, and Y. Shang, “A survey and performance evaluation of deep learning methods for small object detection,” Expert Syst Appl, vol. 172, 2021, doi: 10.1016/j.eswa.2021.114602.
N. D. Nguyen, T. Do, T. D. Ngo, and D. D. Le, “An Evaluation of Deep Learning Methods for Small Object Detection,” Journal of Electrical and Computer Engineering, 2020, doi: 10.1155/2020/3189691.
B. Liu et al., “PRC-Light YOLO: An Efficient Lightweight Model for Fabric Defect Detection,” Applied Sciences (Switzerland), vol. 14, no. 2, 2024, doi: 10.3390/app14020938.
X. Jun, J. Wang, J. Zhou, S. Meng, R. Pan, and W. Gao, “Fabric defect detection based on a deep convolutional neural network using a two-stage strategy,” Textile Research Journal, vol. 91, no. 1–2, 2021, doi: 10.1177/0040517520935984.
P. Bory, “Deep new: The shifting narratives of artificial intelligence from Deep Blue to AlphaGo,” Convergence, vol. 25, no. 4, 2019, doi: 10.1177/1354856519829679.
A. Zahra, M. Amin, F. E. A. El-Samie, and M. Emam, “Efficient utilization of deep learning for the detection of fabric defects,” Neural Comput Appl, vol. 36, no. 11, pp. 6037–6050, 2024, doi: 10.1007/s00521-023-09137-0.
G. Revathy and R. Kalaivani, “Fabric defect detection and classification via deep learning-based improved Mask RCNN,” Signal Image Video Process, vol. 18, no. 3, pp. 2183–2193, 2024, doi: 10.1007/s11760-023-02884-6.
A. International, “Standard Test Methods for Visually Inspecting and Grading Fabrics.” [Online]. Available: https://www.astm.org/d5430-93r00.html
J. Jing, D. Zhuo, H. Zhang, Y. Liang, and M. Zheng, “Fabric defect detection using the improved YOLOv3 model,” J Eng Fiber Fabr, vol. 15, 2020, doi: 10.1177/1558925020908268.
X. Yue, Q. Wang, L. He, Y. Li, and D. Tang, “Research on Tiny Target Detection Technology of Fabric Defects Based on Improved YOLO,” Applied Sciences (Switzerland), vol. 12, no. 13, 2022, doi: 10.3390/app12136823.
X. Luo, Q. Ni, R. Tao, and Y. Shi, “A Lightweight Detector Based on Attention Mechanism for Fabric Defect Detection,” IEEE Access, vol. 11, 2023, doi: 10.1109/ACCESS.2023.3264262.
A. Das Gupta et al., “An approach to automatic fault detection in four-point system for knitted fabric with our benchmark dataset Isl-Knit,” vol. 10, Mar. 2024, doi: 10.1016/j.heliyon.2024.e35931.
Y. Matsuzaka and R. Yashiro, “AI-Based Computer Vision Techniques and Expert Systems,” 2023. doi: 10.3390/ai4010013.
N. T. Allo, Z. Zainuddin, and others, “A Novel Approach of Hybrid Bounding Box Regression Mechanism to Improve Convergency Rate and Accuracy.,” International Journal of Intelligent Engineering & Systems, vol. 17, no. 2, 2024.
W. Xiong, Y. Wu, J. Cao, and R. Li, “IB-CBB: an improved spatial index considering intersection based on clipped bounding boxes,” Ann GIS, vol. 30, no. 2, pp. 233–250, 2024.
M. Sobirov, N. Sharibaev, A. Kayumov, and K. Musayev, “Method of assessment of structural properties of knitted fabrics based on image analysis,” in E3S Web of Conferences, 2024, p. 3020.
B. Koonce, “ResNet 50,” in Convolutional Neural Networks with Swift for Tensorflow: Image Recognition and Dataset Categorization, Berkeley, CA: Apress, 2021, pp. 63–72. doi: 10.1007/978-1-4842-6168-2_6.
DOI: https://doi.org/10.31449/inf.v49i19.10031
This work is licensed under a Creative Commons Attribution 3.0 License.