In this paper, we reported the computational accuracy of various RGB encoding standards using a set of color patches printed by an inkjet printer. These patches were measured in CIELAB under D65 illuminant. The color spaces under study were Adobe RGB98, Bruce RGB, an extended RGB, CIE 1931 RGB, CIE 1964 RGB, a proposed inkjet RGB, Kodak RIMM/ROMM RGB, a proposed Laser RGB, NTSC RGB, Photoshop wide gamut RGB, sRGB, sRGB64, and SMPTE RGB. The computational path was from L*, a*, and b* via CIEXYZ to RGB. The conversion from CIEXYZ to a specified RGB space followed the definition of that space. The resulting RGB values were scaled to integers in various bit-depth. For comparison purpose, the scaled integer RGB values were converted back to CIELAB. The color difference between measured and reversed CIELAB values was used as the measure for computational accuracy.From this exercise, we were able to pinpoint major causes of the computational error. The remedies to the problem were recommended and preferred RGB color spaces were suggested.
Henry R. Kang, "Computational Accuracy of RGB Encoding Standards" in Proc. IS&T Int'l Conf. on Digital Printing Technologies (NIP16), 2000, pp 661 - 664, https://doi.org/10.2352/ISSN.2169-4451.2000.16.1.art00061_2