The optical shape measurement methods relying on digital cameras require that the imaging geometry is known to high accuracy. One needs a camera model and a corresponding calibration procedure that establishes its parameters based on some dedicated dataset. We elaborate the recently suggested concept of the smooth generic camera calibration, emphasizing its utility for the metrological applications. First, we briefly outline the model and its advantages with respect to the state-of-the-art, then elaborate the ideas of consistently propagating the data uncertainties in the calibration process and utilizing the inherent smoothness in the imaging geometry, and finally present the experimental results of calibrating several non-trivial camera setups.
Alexey Pak, Alexey Pak,
"The concept and implementation of smooth generic camera calibration", Proc. SPIE 9960, Interferometry XVIII, 99600I (28 August 2016); doi: 10.1117/12.2237589; https://doi.org/10.1117/12.2237589