A Comprehensive Review of Surface Defect Detection in Manufacturing Using Convolutional Neural Networks
Keywords:
Surface defect detection, Convolutional Neural Networks, Deep learning, Industrial inspection, Machine vision.Abstract
Surface defect detection is a critical component of quality assurance in modern manufacturing, directly influencing product reliability and operational efficiency. With the rise of Industry 4.0, traditional inspection methods have been increasingly replaced by intelligent vision-based systems powered by Convolutional Neural Networks (CNNs). This paper presents a comprehensive review of CNN-based approaches for detecting surface defects across various manufacturing domains, including metal, textile, and electronics industries. It examines key architectures, detection strategies such as classification, object detection, and segmentation, and widely used industrial datasets. The review also highlights recent advancements, including transfer learning, attention mechanisms, and hybrid CNN-Transformer models, which enhance detection accuracy and robustness. Furthermore, it discusses major challenges such as data imbalance, small defect detection, and real-time deployment constraints, while outlining future research directions for scalable and explainable solutions.
References
Ameri, R., Saberirad, S., & Shokouhandeh, H. (2024). A systematic review of deep learning approaches for surface defect detection. Engineering Applications of Artificial Intelligence, 123, 106123.
Cheng, Y., & Yu, L. (2026). A comprehensive survey of industrial surface defect detection: From closed-set to open-set recognition. Robotics and Computer-Integrated Manufacturing. https://doi.org/10.1016/j.rcim.2025.102XXX
He, Z., Zhang, X., & Li, Y. (2025). A comprehensive review of machine vision-based surface defect detection for PCB manufacturing. Array, 20, 100321. https://doi.org/10.1016/j.array.2025.100321
Liao, X., Chen, Y., & Wang, H. (2026). Defect-aware unequal network for industrial surface defect detection. Engineering Applications of Artificial Intelligence. https://doi.org/10.1016/j.engappai.2026.107XXX
Xia, T., Liu, J., & Zhao, Y. (2025). Deep learning-based surface defect detection for stamped parts in manufacturing. Mechanics & Industry, 26(1), 43.
Frydrych, K., & Nowak, M. (2025). Deep learning-based inspection of steel surface defects: A review and challenges. The International Journal of Advanced Manufacturing Technology.
Leng, Y., Zhang, H., & Wu, Q. (2025). Improved Faster R-CNN for steel surface defect detection. Scientific Reports, 15, 12740.
Jha, S. B., & Singh, R. (2023). Deep CNN-based visual defect detection: A survey of current trends. Computer Vision and Image Understanding, 225, 103571.
Chen, W., Li, X., & Zhao, H. (2022). Machined surface defect detection using improved superpixel segmentation and aggregation. Journal of Manufacturing Processes, 80, 287–301.
Park, K. B., Kim, Y. J., & Lee, S. H. (2022). Industrial surface-defect detection using deep nested convolutional networks. Journal of Computational Design and Engineering, 9(6), 2466–2478.
Shaikh, S., & Patil, R. (2022). Surface defect detection using convolutional neural networks in real-time industrial systems. Journal of Engineering Research and Sciences, 1(5), 14–20.
Martin-Ramiro, P., Sainz de la Maza, U., Singh, S., & Orús, R. (2023). Boosting defect detection in manufacturing using tensor convolutional neural networks. arXiv preprint arXiv:2401.01373.
Arikan, S., Varanasi, K., & Stricker, D. (2019). Surface defect classification in real-time using convolutional neural networks. arXiv preprint arXiv:1904.04671.
Downloads
How to Cite
Issue
Section
License
Copyright (c) 2026 International Journal of Engineering Science & Humanities
This work is licensed under a Creative Commons Attribution 4.0 International License.
