A three-dimensional imaging system for surface profilometry of moving objects

Non-contact optical imaging system design and the corresponding surface profilometry algorithm are critical components in various metrology applications, such as surface inspection of semiconductor components on the production line. For such challenging industrial applications, the most important considerations are often automation, precision and speed of the inspection. In this work, we propose a mathematical framework and a dynamic phase-shift algorithm (D-PSA) for a dense surface profilometry of moving objects. We also present a fringe pattern projection system with projector and camera arrays, with an aim to reduce the undesirable effects such as the uneven illumination and the perspective geometry effect on the reconstructed surface using a large field-of-view inspection system. This system is then applied to the inspection of the surface of moving printed circuit boards along a conveyor belt. Experimental results show that our approach can reconstruct the object surface effectively and efficiently.