Research on spectral reflectance reconstruction algorithms for munsell color card

Abstract

XYZ chrominance parameters of color image have some limitations for the characterization of the color information of the image. Spectral reflectance can fully and accurately reflect the color information. It is critical to obtain accurate spectral reflectance using spectral reconstruction. In the paper, mathematical models have been constructed using Wiener estimation algorithm and Principal component analysis algorithm by the given spectral reflectance in 1269 Munsell and tristimulus values in D65 illumination, to find the relationship between spectral reflectance and tristimulus values by mathematical models. The average value 0.0522 of the root mean square error (RMSE) criterion and the mean value 0.9886 of goodness-of-fit coefficient (GFC) which have been obtained by Wiener estimation algorithm are better than the results of the Principal component analysis algorithm.