Deep Learning-Driven Automated Inspection for Defect Detection in Leather and Footwear Manufacturing: A Comprehensive Review
محورهای موضوعی : Artificial Intelligence
1 - Department of Industrial Engineering College of Engineering, Chennai – 600025
کلید واژه: Deep learning, automated inspection, leather defect detection, footwear quality control, computer vision, Industry 4.0, species identification, color analysis, AI in manufacturing,
چکیده مقاله :
Automated inspection systems powered by deep learning are revolutionizing quality control in leather and footwear manufacturing by replacing subjective, time-consuming manual methods with objective, high-throughput solutions. This review presents a comprehensive analysis of deep learning-driven inspection approaches for defect detection in leather and footwear, covering both conventional image processing techniques and state-of-the-art architectures such as convolutional neural networks (CNNs), YOLO series models, and vision transformers. Key application areas include color prediction and sorting, leather species identification, and defect segmentation/classification, with emphasis on integration into real-time industrial workflows. The study examines how the adoption of AI-based inspection improves product quality compliance, reduces rejection rates, and enhances manufacturing competitiveness. It also highlights the transition toward Industry 4.0-aligned inspection systems and identifies current challenges such as dataset scarcity, small-defect detection limitations, and integration with high-speed production lines. Finally, the review proposes future research directions for developing adaptive, domain-specific deep learning models that support scalable, reliable, and sustainable leather and footwear production.
Automated inspection systems powered by deep learning are revolutionizing quality control in leather and footwear manufacturing by replacing subjective, time-consuming manual methods with objective, high-throughput solutions. This review presents a comprehensive analysis of deep learning-driven inspection approaches for defect detection in leather and footwear, covering both conventional image processing techniques and state-of-the-art architectures such as convolutional neural networks (CNNs), YOLO series models, and vision transformers. Key application areas include color prediction and sorting, leather species identification, and defect segmentation/classification, with emphasis on integration into real-time industrial workflows. The study examines how the adoption of AI-based inspection improves product quality compliance, reduces rejection rates, and enhances manufacturing competitiveness. It also highlights the transition toward Industry 4.0-aligned inspection systems and identifies current challenges such as dataset scarcity, small-defect detection limitations, and integration with high-speed production lines. Finally, the review proposes future research directions for developing adaptive, domain-specific deep learning models that support scalable, reliable, and sustainable leather and footwear production.
[1] N. Banduka, K. Tomić, J. Živadinović, and M. Mladineo, “Automated Dual-Side Leather Defect Detection and Classification Using YOLOv11: A Case Study in the Finished Leather Industry,” Processes, vol. 12, no. 12, p. 2892, 2024.
[2] C.-F. Lee, Y.-C. Chen, J.-J. Shen, and A. U. Rehman, “Lightweight Leather Surface Defect Inspection Model Design for Fast Classification and Segmentation,” Symmetry, vol. 17, no. 3, p. 358, 2025.
[3] X. Tao, D. Zhang, W. Ma, X. Liu, and D. Xu, “Automatic Metallic Surface Defect Detection and Recognition with Convolutional Neural Networks,” Applied Sciences, vol. 8, no. 9, p. 1575, 2018.
[4] Z. Peng, C. Zhang, and W. Wei, “Leather Defect Detection Based on Improved YOLOv8 Model,” Applied Sciences, vol. 14, no. 24, p. 11566, 2024.
[5] H. O. Ataç, A. Kayabaşı, and M. F. Aslan, “The Study on Multi-Defect Detection for Leather Using Object Detection Techniques,” Collagen and Leather, vol. 6, no. 37, 2024.
[6] Farahani, Gholamreza, Seyedamirhossein Mousavi, Ameneh Farahani, and Hamidreza Farahani. "Identification of grape leaf diseases using proposed enhanced VGG16." In 2022 27th international conference on automation and computing (ICAC), pp. 1-6. IEEE, 2022, doi: 10.1109/ICAC55051.2022.9911074.
[7] M. Jawahar, R. Venba, G. Jyothi, S. V. Kanth, M. J. Doss, and N. K. C. Babu, “Dry colour prediction of leather from its wet state,” Coloration Technology, vol. 129, no. 1, pp. 1–7, 2013, doi: 10.1111/cote.12033.
[8] M. Jawahar, K. C. Divya, and V. Thankaiselvan, “Sensor based color sorting system for leather shoe components,” in Proc. IEEE Int. Conf. Sensing, Signal Processing and Security (ICSSS), 2017, pp. 296–301, doi: 10.1109/ssps.2017.8071609.
[9] M. Jawahar, N. K. C. Babu, and M. K. Manobhai, “Artificial neural networks for colour prediction in leather dyeing on the basis of a tristimulus system,” Coloration Technology, vol. 131, no. 1, pp. 48–57, 2015, doi: 10.1111/cote.12123.
[10] M. Jawahar, K. Vani, and N. K. C. Babu, “Leather species identification based on surface morphological characteristics using image analysis technique,” J. Amer. Leather Chemists Assoc., vol. 111, pp. 308–319, 2016.
[11] A. Varghese, S. Jain, A. A. Prince, and M. Jawahar, “Digital microscopic image sensing and processing for leather species identification,” IEEE Sensors Journal, vol. XX, no. XX, pp. 1–14, 2020, doi: 10.1109/JSEN.2020.2991881.
[12] M. Jawahar, N. K. C. Babu, and K. Vani, “Leather texture classification using wavelet feature extraction technique,” in Proc. IEEE Int. Conf. Computational Intelligence and Computing Research (ICCIC), 2014, pp. 1–6, doi: 10.1109/iccic.2014.7238475.
[13] M. Jawahar, N. K. C. Babu, K. Vani, L. J. Anbarasi, and S. Geetha, “Vision based inspection system for leather surface defect detection using fast convergence particle swarm optimization ensemble classifier approach,” Multimedia Tools and Applications, vol. 80, pp. 14513–14541, 2021, doi: 10.1007/s11042-020-09727-3.
[14] Liong, H. Gan, Y. Chai, and Y. Cheong, “Two-Stage CNN-Based Leather Flaw Classification and Segmentation Using AlexNet and U-Net,” Journal of Imaging, vol. 5, no. 5, p. 45, 2019.
[15] S. Smith, T. Liu, and H. Chen, “A Vision Transformer-Based Approach for Multi-Class Leather Defect Detection,” Computers in Industry, vol. 148, p. 103901, 2023.
[16] Redmon, J.; Farhadi, A. YOLOv3: An Incremental Improvement. arXiv preprint arXiv:1804.02767, 2018.
[17] Chen, J.; Li, S.; Wang, Y. “Semi-Supervised Learning for Industrial Defect Segmentation.” IEEE Access 2020, 8, 145620–145632.
[18] Dosovitskiy, A.; Beyer, L.; Kolesnikov, A.; et al. “An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale.” arXiv preprint arXiv:2010.11929, 2020.
[19] T. Tao, L. Xu, and S. Gao, “CASAE: Cascaded autoencoder for defect segmentation in metal materials,” Appl. Sci., vol. 8, no. 9, p. 1575, Sep. 2018, doi: 10.3390/app8091575.
[20] X. Xu, G. Zhang, W. Zheng, A. Zhao, Y. Zhong, and H. Wang, “High-precision detection algorithm for metal workpiece defects based on deep learning,” Machines, vol. 11, no. 8, p. 834, Aug. 2023, doi: 10.3390/machines11080834.
[21] M. H. Zubayer, C. Zhang, and Y. Wang, “Deep learning-based automatic defect detection of additive manufactured stainless steel,” Metals, vol. 13, no. 12, p. 1987, Dec. 2023, doi: 10.3390/met13121987.
[22] D. Tabernik, S. Šela, J. Skvarč, and D. Skočaj, “Segmentation-based deep-learning approach for surface-defect detection,” J. Intell. Manuf., vol. 31, pp. 759–776, 2019, doi: 10.1007/s10845-019-01476-x.
[23] H. Li et al., “Automated detection of micro-scale porosity defects in reflective metal parts via deep learning and polarization imaging,” Nanomaterials, vol. 15, no. 11, p. 795, May 2025, doi: 10.3390/nano15110795.
[24] J. Yang, S. Li, Z. Wang, H. Dong, J. Wang, and S. Tang, “Using deep learning to detect defects in manufacturing: A comprehensive survey and current challenges,” Materials, vol. 13, no. 24, p. 5755, Dec. 2020, doi: 10.3390/ma13245755.
[25] Y. Ma, J. Yin, F. Huang, and Q. Li, “Surface defect inspection of industrial products with object detection deep networks: A systematic review,” Artif. Intell. Rev., vol. 57, p. 333, Oct. 2024, doi: 10.1007/s10462-024-10956-3.
[26] A. A. M. S. Ibrahim and J.-R. Tapamo, “A survey of vision-based methods for surface defects’ detection and classification in steel products,” Informatics, vol. 11, no. 2, p. 25, Apr. 2024, doi: 10.3390/informatics11020025.
