3D reconstruction of transparent objects using visible light tomography

Abstract

We demonstrate the feasibility of using visible light tomography in 3D reconstruction of transparent objects. Reconstructing transparent objects remains a challenging problem in optical metrology due to the complex interaction of light with refractive surfaces. In this study, a low-cost smartphone-based CT setup was used to capture projection images of thin-walled transparent objects. Using the projection data, a filtered back-projection algorithm was implemented to obtain the reconstruction of the object. Results showed that we were able to recover the overall geometry of thin-walled transparent objects, while thicker surfaces produced inaccurate reconstructions caused by strong refraction and transmission effects.