COMPUTER VISION-BASED QUALITY INSPECTION WITH COLLABORATIVE ROBOTS: AN AI-DRIVEN FRAMEWORK FOR SMART MANUFACTURING AND ZERO-DEFECT PRODUCTION
Keywords:
Computer Vision, Collaborative Robots, Quality Inspection, Deep Learning, Industry 4.0, Smart Manufacturing, Defect Detection, YOLO, CNN, Edge AI.Abstract
Computer vision-based quality inspection systems integrated with collaborative robots (cobots) represent a transformative shift in modern intelligent manufacturing. Traditional industrial inspection approaches rely heavily on manual evaluation or rule-based machine vision techniques, which are often limited by environmental variability, subjective judgment, and low scalability. With the rapid advancement of deep learning, edge artificial intelligence, and Industrial Internet of Things (IIoT), automated inspection systems have evolved into highly adaptive, data-driven frameworks capable of real-time defect detection and quality assessment. This study investigates the integration of computer vision and collaborative robotics for intelligent quality inspection in industrial environments, focusing on system architecture, deep learning methodologies, and real-time decision-making mechanisms. The research highlights how convolutional neural networks (CNNs), object detection models such as YOLO and Faster R-CNN, and transformer-based vision architectures enable high-precision defect identification and classification. Furthermore, the study analyzes the role of cobots in enhancing inspection flexibility through adaptive positioning, multi-angle visualization, and human-robot collaboration. The findings indicate that AI-driven cobot inspection systems significantly improve defect detection accuracy, reduce inspection time, enhance production consistency, and enable scalable deployment in complex manufacturing environments. Despite these advantages, challenges such as computational constraints, dataset scarcity, and environmental robustness remain critical issues. The paper concludes that computer vision-enabled collaborative robotic inspection systems are a key enabling technology for zero-defect manufacturing and next-generation smart factories under Industry 4.0 and Industry 5.0 paradigms.
References
1.He K., Zhang X., Ren S., Sun J. Deep Residual Learning for Image Recognition. CVPR, 2016.
2.Redmon J., Farhadi A. YOLOv3: An Incremental Improvement. arXiv, 2018.
3.Dosovitskiy A. et al. An Image is Worth 16x16 Words: Transformers for Image Recognition. ICLR, 2021.
4.LeCun Y., Bengio Y., Hinton G. Deep Learning. Nature, 2015.
5.Tao F., Qi Q., Liu A., Kusiak A. Digital Twins and Cyber-Physical Systems in Smart Manufacturing. IEEE TII, 2019.
6.Luo Q. et al. Deep Learning for Industrial Defect Detection: A Survey. Engineering Applications of AI, 2023.
7.Zhang W. et al. Collaborative Robots in Smart Manufacturing. Robotics and Computer-Integrated Manufacturing, 2022.
8.Wang J. et al. Edge AI for Industrial Inspection Systems. IEEE IoT Journal, 2024.