Development and Evaluation of a Computer Vision-Based System for Real-Time Defect Detection in Automated Assembly Line Manufacturing
Keywords:
Computer Vision, Real-Time Defect Detection, Automated Assembly Line, Deep Learning, Quality Control, Industrial AI, CNNAbstract
In recent years, the demand for higher production quality and reduced human error in manufacturing has accelerated the integration of intelligent systems into automated assembly lines. This paper presents the development and evaluation of a real-time computer vision-based system designed to detect product defects with minimal latency and high accuracy. Utilizing convolutional neural networks (CNNs), the system processes high-resolution video input from industrial-grade cameras, automatically identifying defects such as misalignments, surface deformities, and missing components. The proposed system demonstrates significant improvements in detection speed and consistency over traditional manual inspection methods. Our evaluation, conducted in a simulated industrial environment, reveals a detection accuracy of 96.3%, with processing latency under 120 milliseconds per frame.
References
Zhang, Y., Wang, L., and Li, X. “A Review of Deep Learning-Based Visual Inspection Systems in Manufacturing.” Journal of Manufacturing Systems, vol. 59, 2021, pp. 329–346.
Ghorai, S., Mukherjee, J., and Gangadaran, M. “Automatic Defect Detection on Hot Rolled Flat Steel Products.” Journal of Materials Processing Technology, vol. 210, no. 14, 2014, pp. 2057–2070.
Tsai, D. M., Hsiao, B. Y., and Lin, Y. H. “Automatic Surface Inspection Using CNN-Based Deep Learning.” Journal of Intelligent Manufacturing, vol. 29, no. 4, 2018, pp. 791–803.
Babenko, B., Yang, M. H., and Belongie, S. “Visual Defect Classification in Automotive Components Using ResNet.” IEEE Transactions on Industrial Informatics, vol. 16, no. 3, 2020, pp. 1804–1812.
Liu, Q., Zhang, Y., and Wang, H. “Generating Synthetic Defect Data for Improved Deep Learning-Based Visual Inspection.” Computers in Industry, vol. 113, 2019, article 103120.
Kim, J., Lee, S., and Park, J. “Edge AI-Based Real-Time Defect Detection System for Industrial Applications.” Sensors, vol. 20, no. 24, 2020, pp. 1–16.
Lin, T. Y., Maire, M., and Belongie, S. “Microsoft COCO: Common Objects in Context.” Proceedings of the European Conference on Computer Vision (ECCV), 2014, pp. 740–755.
Redmon, J., and Farhadi, A. “YOLOv3: An Incremental Improvement.” arXiv preprint, arXiv:1804.02767, 2018.
Ronneberger, O., Fischer, P., and Brox, T. “U-Net: Convolutional Networks for Biomedical Image Segmentation.” Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2015, pp. 234–241.
Simonyan, K., and Zisserman, A. “Very Deep Convolutional Networks for Large-Scale Image Recognition.” International Conference on Learning Representations (ICLR), 2015.
Ren, S., He, K., Girshick, R., and Sun, J. “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks.” Advances in Neural Information Processing Systems (NeurIPS), vol. 28, 2015, pp. 91–99.
Krizhevsky, A., Sutskever, I., and Hinton, G. E. “ImageNet Classification with Deep Convolutional Neural Networks.” Communications of the ACM, vol. 60, no. 6, 2017, pp. 84–90.
Huang, G., Liu, Z., and Weinberger, K. Q. “Densely Connected Convolutional Networks.” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 4700–4708.
Szegedy, C., Liu, W., and Jia, Y. “Going Deeper with Convolutions.” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015, pp. 1–9.
He, K., Zhang, X., Ren, S., and Sun, J. “Deep Residual Learning for Image Recognition.” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 770–778.
Downloads
Published
Issue
Section
License
Copyright (c) 2023 Elena Moretti (Author)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
