Spectral Gamut Mapping using LabPQR

Abstract

Spectral color management requires inversion of printer spectral characterizations and necessarily involves the concept of spectral gamut mapping. A printer was spectrally characterized and the spectra were transformed to an interim connection space (ICS), a spectral description space with low dimensionality useful for building lookup tables (LUTs) of feasible sizes. LabPQR is the ICS used. It has separate dimensions describing colorimetry (CIELAB) and a spectrum's metameric black difference from a standard metamer (PQR). The relationship between digital value and LabPQR was inverted using a single stage objective function combining colorimetric and spectral criteria. The objective function's colorimetric criterion minimized CIEDE2000 under chosen conditions and its spectral criterion minimized Euclidian distance in PQR coordinates. A weight series was performed to find the optimal trade-off between colorimetric and spectral error. A 1:50 weighting ratio, CIEDE2000 to PQR difference, was deemed best. For the GretagMacbeth ColorChecker, the proposed single stage objective function showed equivalent levels of the performance to a full 31-dimensional unmodified spectra approach, resulting in an average RMS error of 4.18% and an average CIEDE2000 of 0.03. The single stage objective function for spectral gamut mapping using LabPQR proved to be promising for spectral reproduction.