Improving 3D photogrammetry models through spectral imaging: Tooth enamel as a case study

by Aurore Mathys, Patrick Semal, Jonathan Brecko, Didier Van den Spiegel

Reflective or translucent materials are a challenge to digitize in 3D. Results are better with a matt coating although objects from museum collections are often too fragile or too valuable to be treated in this way. It is therefore essential that alternative solutions are found. This study analyzed spectral photogrammetry as a possible solution. Spectral photogrammetry is an emerging technique which uses images at different wavelengths to create 3D models. Tooth enamel is a challenging material to digitize. Six sets of teeth were photographed at different wavelengths. The results showed that the quality of the models enamels parts improved when taken with ultraviolet wavelengths whilst models were less accurate when photogrammetry was performed with the red and infrared spectrum. This can be explained by the optical properties of enamel. This study demonstrates that knowing the optical properties of a material beforehand could help future photogrammetric digitization of challenging materials.