COMPUTER VISION FOR AUTONOMOUS VISUAL INSPECTION IN INDUSTRIAL PRODUCTION LINES
DOI:
https://doi.org/10.47820/recima21.v6i3.6177Keywords:
Computer vision. Vision application. Industrial environments.Abstract
This article has the specific objectives of highlighting the computer vision techniques used in the IVAP project, discussing the stages of implementation of the system and analyzing the results obtained. Throughout this study, we hope to contribute to the existing literature on the application of computer vision in industry, citing examples from authors such as Gerald J. Agin (1980) and Rodrigo Barbosa Davies (2012), who explored the practice and effectiveness of these technologies in industrial settings. Computer vision has established itself as an essential tool for visual inspection in industrial production lines, promoting significant improvements in the quality and efficiency of manufacturing processes. This study addresses the implementation of an advanced computer vision system in the IVAP (Visual Autonomous Product Inspection) project, developed in partnership between the Conecthus Institute and Vantiva. The implementation of the Keyence CV-X Series system was motivated by the need to perform autonomous inspections, standardizing product quality without relying on manual evaluations, which are prone to errors. The development of the project focused on the use of sophisticated computer vision techniques, including defect detection and classification through machine learning algorithms for image processing and cosmetic defect analysis. The techniques used were studied and adapted to the industrial context, allowing the detailed inspection of cosmetic aspects of the products, with minimal exceptions. These techniques, which included lighting adjustments, camera sensitivity, and algorithms for detecting smudges and scratches, were crucial to the system's effectiveness.
Downloads
References
BENBARRAD, T.; SALHAOUI, M.; KENITAR, S. B. et al. Intelligent machine vision model for defective product inspection based on machine learning. Journal of Sensor and Actuator Networks, v. 10, n. 1, p. 7, 2021. DOI: https://doi.org/10.3390/jsan10010007
CHEN, Y. W.; SHIU, J. M. An implementation of YOLO-family algorithms in classifying the product quality for the acrylonitrile butadiene styrene metallization. The International Journal of Advanced Manufacturing Technology, v. 119, n. 11, p. 8257–8269, 2022. https://doi.org/10.1007/s00170-022-08676-5 DOI: https://doi.org/10.1007/s00170-022-08676-5
DAVIES, E. Computer and machine vision: Theory, algorithms, practicalities. 4 ed. [S. l.]: Academic Press, 2012. https://doi.org/10.1016/C2010-0-66926-4 DOI: https://doi.org/10.1016/C2010-0-66926-4
FORSYTH, D.; PONCE, J. Computer vision: A modern approach. 2 ed. [S. l..]: Pearson, 2012.
HÜTTEN, N.; GOMES, M. A.; HÖLKEN, F. et al. Deep learning for automated visual inspection in manufacturing and maintenance: A survey of open-access papers. Applied System Innovation, v. 7, n. 1, 2024. https://doi.org/10.3390/asi7010011 DOI: https://doi.org/10.3390/asi7010011
LI, Y.; HUANG, H.; XIE, Q. et al. Research on a surface defect detection algorithm based on MobileNet-SSD. Applied Sciences, v. 8, n. 9, p. 1678, 2018. DOI: https://doi.org/10.3390/app8091678
LIN, C.-Y.; CHOU, Y.-H.; CHENG, Y.-C. A deep learning-based general defect detection framework for automated optical inspection. In: 2023 IEEE International Conference on Industry 4.0, Artificial Intelligence, and Communications Technology (IAICT). 2023. p. 332–337. https://doi.org/10.1109/IAICT59002.2023.10205799 DOI: https://doi.org/10.1109/IAICT59002.2023.10205799
LIU, W.; ANGUELOV, D.; ERHAN, D. et al. SSD: Single shot multibox detector. In: LEIBE, B.; MATAS, J.; SEBE, N.; WELLING, M. (Eds.). Computer Vision – ECCV 2016. [S. l.]: Springer International Publishing, 2016. p. 21–37. DOI: https://doi.org/10.1007/978-3-319-46448-0_2
LIU, X.; WANG, J.; ZHANG, R. et al. Deep learning-based defect detection for printed circuit boards. In: SHEN, L.; ZHONG, G. (Eds.). Third International Conference on Computer Vision and Pattern Analysis (ICCPA). 2023.p. 127543I). International Society for Optics and Photonics SPIE. https://doi.org/10.1117/12.2684258 DOI: https://doi.org/10.1117/12.2684258
REDMON, J.; DIVVALA, S.; GIRSHICK, R. et al. You only look once: Unified, real-time object detection. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2016. p. 779–788. https://doi.org/10.1109/CVPR.2016.91 DOI: https://doi.org/10.1109/CVPR.2016.91
SERRA, J. Image analysis and mathematical morphology. [S. l.]: Academic Press, 1982.
SILVA, W. R. L. D.; LUCENA, D. S. Concrete cracks detection based on deep learning image classification. Proceedings, v. 2, n. 8, 2018. https://doi.org/10.3390/ICEM18-05387 DOI: https://doi.org/10.3390/ICEM18-05387
STEGER, C.; ULRICH, M.; WIEDEMANN, C. Machine vision algorithms and applications. [S. l.]: John Wiley & Sons, 2018.
SZELISKI, R. Computer vision: Algorithms and applications. [S. l.]: Springer, 2011. DOI: https://doi.org/10.1007/978-1-84882-935-0
THOLEY, N. Digital processing of earth observation images. Surveys in Geophysics, v. 21, p. 209–222, 2000. https://doi.org/10.1023/A:1006714832064. DOI: https://doi.org/10.1023/A:1006714832064
WU, X.; GE, Y.; ZHANG, Q. et al. PCB defect detection using deep learning methods. In: IEEE 24th International Conference on Computer Supported Cooperative Work in Design (CSCWD). 2021. p. 873–876. https://doi.org/10.1109/CSCWD49262.2021.9437846 DOI: https://doi.org/10.1109/CSCWD49262.2021.9437846
ZHU, L.; SPACHOS, P.; PENSINI, E. et al. Deep learning and machine vision for food processing: A survey. Current Research in Food Science, v. 4, p. 233–249, 2021. https://doi.org/10.1016/j.crfs.2021.03.009. DOI: https://doi.org/10.1016/j.crfs.2021.03.009
Downloads
Published
License
Copyright (c) 2025 RECIMA21 - Revista Científica Multidisciplinar - ISSN 2675-6218
This work is licensed under a Creative Commons Attribution 4.0 International License.
Os direitos autorais dos artigos/resenhas/TCCs publicados pertecem à revista RECIMA21, e seguem o padrão Creative Commons (CC BY 4.0), permitindo a cópia ou reprodução, desde que cite a fonte e respeite os direitos dos autores e contenham menção aos mesmos nos créditos. Toda e qualquer obra publicada na revista, seu conteúdo é de responsabilidade dos autores, cabendo a RECIMA21 apenas ser o veículo de divulgação, seguindo os padrões nacionais e internacionais de publicação.
