Radiometric calibration of vision cameras and intensity uncertainty estimation

The irradiance measurement performed by vision cameras is not noise-free due to both processing errors during CCD fabrication and the behaviour of the electronic device itself. A proper characterization of sensor performance, however, allows either removing the resulting noise from the image or accounting for it within image processing algorithms. This paper proposes a robust algorithm named R2CIU for characterizing the noise sources affecting CCD performance with the aim of estimating the uncertainty of the intensity values yielded by vision cameras. R2CIU makes use of sets of images of plane calibration cards such that, for every set, the average intensity changes. As it is shown in the paper, the (spatial) average and variance of (temporal) average images of every set lie in a parabola, while the (spatial) averages of the (temporal) average and variance images of every set lie in a straight line. R2CIU exploits these two facts to compute the noise source distribution parameters by least-squares fitting. Intensity uncertainties are estimated next by a sum in quadrature of the variances of the different noise sources involved in the process of image formation. Experimental results and application examples can be found at the end of the paper.