Colour matching experiments with RGB‐LEDs

Abstract

CIE colorimetry breaks down when lights produced by narrow band RGB-LEDs are matched with broadband lights. A colour matching experiment was set up and matches in a number of parts of the chromaticity diagram have been made, to determine the magnitude of the discrepancy. Differences between visual and instrumental matches increase as one moves in the chromaticity diagram from yellowish white lights toward greenish and bluish lights.CIE TC 1-36 recently suggested newly defined cone fundamentals: Applying a transformation of these to a space similar to the CIE XYZ space enables a much better prediction of the matches to be made. The difference between the visual match and its instrumental prediction decreases by a factor of two or even more.The use of a cone fundamental based colorimetric system is recommended for LED colorimetry.

Memory and preferred colours and the colour rendition of white light sources

“…The latter is known to be in error in the blue part of the spectrum, which would cause a considerable difference between the instrumental and visual colour matches for many of the LED light sources used in the visual experiments. 75 In contrast to Sanders' R p , Judd's flattery index R f and Thornton's CPI have a much better correlation with visual appreciation.…”

Minimizing observer metamerism in display systems

“…Said differently, we are interested in investigating the effect of observer metamerism-in which the same spectral stimulus generates different tristimulus values for different observers due to variations in their color matching functions. This study uses the CIE 28 observer as a reference observer and compares the colorimetric performance of following observers, as representative variations from this standard observer: 1) CIE 28 Standard Deviate observer, 12 2) CIE 108 observer, 13 3) Stiles-Burch observer DBJ, 1 4) Stiles-Burch observer JAR, 1 5) Shaw-Fairchild Observer, 14 and 6) CIE TC1-36 Observer, 15,16 for eight different displays (characterized purely by the spectral power distributions of their color primaries): a) CRT, b) HID lamp-based DLP 1 , c) HID lamp-based LCD, d) WCG-CCFL-based direct-view Bravia 1 LCD, e) Xenon-based DLP Cinema 1 , f) Radio-frequency based electrode-less plasma lamp (LIFI TM ), 17 g) LED (based on technology from Luminus Devices), 18 and h) LASER (primary wavelengths of 460 nm, 532 nm, and 635 nm). 19,20 The Stiles-Burch observers DBJ and JAR have been chosen as two extremes of the various Stiles-Burch 108 observers with regards to the location of their transition wavelengths-an indicator of the differences between observers.…”

“…The LMS cone fundamentals are converted to XYZ color matching functions using the matrices provided by Csuti and Schanda. 16 In this analysis, we are looking for the variation in A t j L i c i for different instances of observers j (Fig. 2).…”

Optimization of color reproduction based on chromatic properties of multiple real observers

“…The function of a color visualization device is to capture or reproduce target colors with limited color primaries. For digital cameras or LCD monitors, the primaries are red, green and blue colors and their reproduced colors can be predicted by the additive model [3,4]. For printing devices, their primaries are cyan, magenta, yellow and black colorants which follow the subtractive principle [3,5].…”