Calculation of the Color-Matching Functions of Digital Cameras from their Complete Spectral Responsitivities

©2002, IS&T—The Society for Imaging Science and Technology initial processing stages, any digital image capture device is an additive color reproduction system because the optoelectronic sensor captures or sums photons. Therefore its raw spectral sensitivities must all be positive.6–8 However, when the raw RGB space is electronically varied, the resulting spectral sensitivities may have negative parts. This hinders the fundamental analysis of the color reproduction quality among apparently similar performances of different trademark devices. This subject is so important that ISO 173219 (Color characterization of digital cameras) proposes that the direct digital data, like RGB space, should not be matrixed except when they have been previously transformed using the following processes: • linearization (if necessary); • dark current/frame subtraction; • shading & sensitivity (flat field) correction; • flare removal (scene-dependent); • white balancing (adopted white has neutral code values; i.e. equal RGB values or no chrominance); • demosaicing (color pixel reconstruction).