Automated quality inspection of fastening screws for furniture using 3D computer vision and data analysis techniques

In this work, we present a feasibility study on the use of 3D analysis for quality control of fastening screws in furniture assembly. 3D computer vision techniques based on laser profilometry were employed to replace manual sampling inspections with automated quality checks. A suitable dataset of point clouds representing the upper layer of the inspected containers was first collected. The 3D data was then analyzed by automatically isolating the screws through unsupervised clustering and performing accurate measurements against the reference CAD model of each component. This analysis enabled the evaluation of the inspection technologies’ effectiveness in supporting predictive maintenance and monitoring quality drift. The results confirm the potential of this approach for assessing the quality of the manufactured screws.