Enhanced YOLOv5 with SPC Architecture for Automated Quality Inspection and Robot‑Guided Grasping of Mechanical Parts
DOI:
https://doi.org/10.66069/ojspub.26820804Keywords:
mechanical parts inspection, deep learning, YOLOv5, robot grasping, hand‑eye calibration, industrial automationAbstract
Ensuring stable machinery operation is critical for both production efficiency and safety in industrial settings, with mechanical parts cleaning serving as a key pre‑assembly process. This study presents a vision‑based quality inspection and robotic grasping system for mechanical parts, integrating deep learning with automated control. A custom parts dataset was created, and the YOLOv5 network was enhanced by replacing its SPP structure with a novel SPC (Spatial Pyramid without Pooling) design, which eliminates pooling layers and achieves improved accuracy and recall over the baseline model. For automated handling, a camera‑based vision system, combined with camera and hand‑eye calibration, computes part coordinates in the robot base frame; these coordinates and part type information are transmitted to the robot via Modbus/TCP. The proposed system demonstrates a viable pathway toward higher automation levels in parts inspection and sorting, contributing to improved manufacturing quality and operational reliability.
References
JIN L S, JIA M, SUN Y O, et al. Detection of backside vehicle on freeway in daytime [i] . Journal of Southwest Jiao tong University ,2010 ,45(2): 231 – 237.
Meng , L. (2023). Research on the Evaluation System of Green Cabling of Cables Based on Neural Network. Innovation & Technology Advances, 1(2), 25–31. https://doi.org/10.61187/ita.v1i2.37
Junxi, Y., Wang, Z., & Chen, C. (2024). GCN-MF: A graph convolutional network based on matrix factorization for recommendation. Innovation & Technology Advances, 2(1), 14–26. https://doi.org/10.61187/ita.v2i1.30
Hu, H., Zhang, J., & Sun, Y. (2024). The Multiscale Deep Neural Networks: Unveiling New Directions in Text Sentiment Analysis. Innovation & Technology Advances, 2(2), 34–45. https://doi.org/10.61187/ita.v2i2.65
Li, W. Building an Intelligent E-commerce Platform: From System Design to Meeting User Emotional Needs. Int. J. Comput. Sci. Inf. Technol. 2024, 4, 449–455, https://doi.org/10.62051/ijcsit.v4n3.51.
C. Shi, G. Macdonald, B. Jalli, W. Lei, J. Zou, M. Jain, and J. Philip, “Think Less, Label Better: Multi-Stage Domain-Grounded Synthetic Data Generation for Fine-Tuning Large Language Models in Telecommunications,” in 2026 IEEE International Conference on Communications (ICC), 2026, pp. 1–6, doi: 10.1109/ICC59461.2026.11587874.
J. Choudhury, P. Ky, Y. Ren, and C. Shi, “Hypersphere for Branching Node for the Family of Isolation Forest Algorithms,” in 2021 IEEE International Conference on Smart Computing (SMARTCOMP), 2021, pp. 418–423, doi: 10.1109/SMARTCOMP52413.2021.00090.