Encyclopedia of Color Science and Technology

2016 Edition
| Editors: Ming Ronnier Luo

CIE Guidelines for Mixed Mode Illumination: Summary and Related Work

  • Suchitra Sueeprasan
Reference work entry
DOI: https://doi.org/10.1007/978-1-4419-8071-7_4



Mixed mode illumination refers to an image comparison between softcopy and hardcopy with successive binocular viewing. When comparing the softcopy images on self-luminous displays with the hardcopy images under ambient lighting, an observer’s eyes move back and forth between the images. Under such circumstance, the state of adaptation is unfixed and the human visual system partially adapts to the white point of the softcopy display and partially adapts to the ambient illumination. The term mixed chromatic adaptation is defined as a state in which observers adapt to light from sources of different chromaticities [3].


In 1998, the Technical Committee 8-04, Adaptation under Mixed Illumination Conditions, was formed in Commission Internationale de l’Eclairage (CIE)/Division 8 (Image Technology), with the aim to investigate the state of adaptation of the visual system when comparing softcopy images on self-luminous displays and hardcopy images viewed under various ambient lighting conditions. A number of experiments were conducted with the goal to achieve color appearance matches between softcopy and hardcopy images under mixed illumination conditions. Katoh [5, 6, 7] developed the mixed adaptation model, namely, S-LMS, for such application. In his study, softcopy images on a CRT were compared with hardcopy images under F6 illumination. Under equal luminance levels of softcopy and hardcopy, it was found that the human visual system was 60 % adapted to the monitor’s white point and 40 % to the ambient light. The same adaptation ratio was also found for unequal luminance levels. It is concluded that the adaptation ratio was independent of image content, luminance, and chromaticity of the monitor’s white point and the ambient illumination.

The study by Berns and Choh [1], in which softcopy images were compared with hardcopy images under F2 illuminant with equal luminance levels, showed that an image with a chromatic adaptation shift of 50 % was most preferred as the closest color match and the best stand-alone image. The color model tested in this study was the RLAB color space. Shiraiwa et al. [9] proposed a new method in which the mixed chromatic adaptation was applied in CIE xy chromaticity coordinates. The best adaptation ratio was between 50 % and 60 %, which is similar to the previous studies. In the visual experiments, where the illuminants of softcopy and hardcopy images were different, their proposed method and the S-LMS model, both incorporating the mixed adaption, generated better color appearance matches than the conventional color management systems.

Henley and Fairchild [4] tested the performance of various color models with the inclusion and exclusion of mixed adaptation. Observers made appearance matches of color patches on a CRT to hardcopy originals under six different matching methods. The results were reported in terms of color differences between the actual match and the predicted match by color models. The models incorporating the mixed adaptation improved the results in all conditions over their corresponding conventional models.

The studies by Katoh and Nakabayashi [8] and Sueeprasan and Luo [10] closely followed the experimental guidelines put forth by TC8-04. In the experiments, softcopy and hardcopy images were compared using the simultaneous binocular matching technique. Various color models were tested. In Katoh and Nakabayashi’s study, the linear transformation matrix in the S-LMS model was replaced by different chromatic adaptation transform matrices. The results showed that the S-LMS model with the Bradford (BFD) matrix performed best. They also investigated whether incomplete adaptation was needed in the mixed chromatic adaptation model. RLAB method and D factor resulted in much better score than complete adaptation, indicating that the incomplete adaptation was essential.

In Sueeprasan and Luo’s study, the performance of the promising chromatic adaptation transforms (CMCCAT97, CMCCAT2000, and CIECAT94) and the S-LMS mixed chromatic adaptation transform was compared. The state of chromatic adaptation was also investigated. The results showed that the incomplete adaptation ratio was crucial in producing color matches. The human visual system was between 40 % and 60 % adapted to the white point of the monitor regardless of the changes in illumination conditions. CMCCAT2000 outperformed the other models.

The results from the previous studies are in good agreement for the chromatic adaptation ratio, which is in the range of 40–60 % adapted to the white point of the monitor. The adaptation ratio is consistent over various viewing conditions and regardless of the chromatic adaptation transform and incomplete adaptation formula used. Both incomplete and mixed chromatic adaptations are required in the mixed adaptation model for predicting color matches under mixed illumination conditions. Based on these findings, TC8-04 recommends the mixed adaptation model for use in cross-media color reproduction when mixed illumination conditions are employed.

Recommended Model

CIE TC8-04 recommends the S-LMS mixed adaptation model for achieving appearance matches under mixed chromatic adaptation. The S-LMS model is fundamentally a modified form of the von Kries transformation with incorporation of partial adaptation. The compensation for chromatic adaptation includes incomplete adaptation and mixed adaptation.

The first step of the S-LMS model is to transform XYZ tristimulus values to the cone signals for the human visual system (Eqs. 1, 2, and 3). X n Y n Z n values are tristimulus values of the reference white. MCAT02 is the chromatic adaptation transformation matrix used in CIECAM02 [2].
$$ \left[\begin{array}{c}\hfill L\hfill \\ {}\hfill M\hfill \\ {}\hfill S\hfill \end{array}\right]={M}_{\mathrm{CAT}02}\left[\begin{array}{c}\hfill X\hfill \\ {}\hfill Y\hfill \\ {}\hfill Z\hfill \end{array}\right] $$
$$ \left[\begin{array}{c}\hfill {L}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}\hfill \\ {}\hfill {M}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}\hfill \\ {}\hfill {S}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}\hfill \end{array}\right]={M}_{\mathrm{CAT}02}\left[\begin{array}{c}\hfill {X}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}\hfill \\ {}\hfill {Y}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}\hfill \\ {}\hfill {Z}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}\hfill \end{array}\right] $$
$$ {M}_{\mathrm{CAT}02}=\left[\begin{array}{ccc}\hfill 0.7328\hfill & \hfill 0.4296\hfill & \hfill -0.1624\hfill \\ {}\hfill -0.7036\hfill & \hfill 1.6974\hfill & \hfill 0.0061\hfill \\ {}\hfill 0.0030\hfill & \hfill 0.0136\hfill & \hfill 0.9834\hfill \end{array}\right] $$
Then, compensation is made for the change in chromatic adaptation according to the surroundings. The human visual system changes the cone sensitivity of each channel to compensate for the change in illumination. In the calculation, the signals of each channel are divided by those of the adapted white. There are two steps of calculation to obtain the adapted white of a monitor.
The first step is the compensation for the incomplete adaptation of the visual system to the self-luminous displays (Eq. 4). Even if the monitor is placed in a totally dark room, the chromatic adaptation of the human visual system to the white point of the monitor will not be complete. That is, the reference white of the monitor does not appear perfectly white. Chromatic adaptation becomes less complete as the chromaticity of the adapting stimulus deviates from the illuminant E and as the luminance of the adapting stimulus decreases. The D factor from CIECAM02 is used (Eq. 5). F is 1.0, and L A is the absolute luminance of the adapting field.
$$ \begin{array}{l}{L}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}^{\prime }={L}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}/\left\{D+{L}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}\left(1-D\right)\right\}\\ {}{M}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}^{\prime }={M}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}/\left\{D+{M}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}\left(1-D\right)\right\}\\ {}{S}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}^{\prime }={S}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}/\left\{D+{S}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}\left(1-D\right)\right\}\end{array} $$
$$ D=F\left\{1-\left(\frac{1}{3.6}\right){e}^{\left(\frac{-{L}_A-42}{92}\right)}\right\} $$
The next step is the compensation for mixed adaptation. In cases where the white points of the monitor and the ambient light are different, it was hypothesized that the human visual system is partially adapted to the white point of the monitor and partly to the white point of the ambient light. Therefore, the adapting stimulus for softcopy images can be expressed as the intermediate point of the two (Eqs. 6 and 7). It should be noted that incompletely adapted white is used for the white point of the monitor. Yn(CRT) is the absolute luminance of the white point of the monitor, and Yambient is the absolute luminance of the ambient light.
$$ {L}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}^{{\prime\prime} }={R}_{\mathrm{adp}}\cdot {\left(\frac{Y_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}}{Y_{\mathrm{adp}}}\right)}^{1/3}\cdot {L}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}^{\prime }+\left(1-{R}_{\mathrm{adp}}\right)\cdot {\left(\frac{Y_{\mathrm{ambient}}}{Y_{\mathrm{adp}}}\right)}^{1/3}\cdot {L}_{\mathrm{ambient}}{M}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}^{{\prime\prime} }={R}_{\mathrm{adp}}\cdot {\left(\frac{Y_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}}{Y_{\mathrm{adp}}}\right)}^{1/3}\cdot {M}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}^{\prime }+\left(1-{R}_{\mathrm{adp}}\right)\cdot {\left(\frac{Y_{\mathrm{ambient}}}{Y_{\mathrm{adp}}}\right)}^{1/3}\cdot {M}_{\mathrm{ambient}}{S}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}^{{\prime\prime} }={R}_{\mathrm{adp}}\cdot {\left(\frac{Y_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}}{Y_{\mathrm{adp}}}\right)}^{1/3}\cdot {S}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}^{\prime }+\left(1-{R}_{\mathrm{adp}}\right)\cdot {\left(\frac{Y_{\mathrm{ambient}}}{Y_{\mathrm{adp}}}\right)}^{1/3}\cdot {S}_{\mathrm{ambient}} $$
$$ {Y}_{\mathrm{adp}}={\left\{{R}_{\mathrm{adp}}\cdot {Y_{ n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}}^{1/3}+\left(1-{R}_{\mathrm{adp}}\right)\cdot {Y_{\mathrm{ambient}}}^{1/3}\right\}}^3 $$
When the luminance of the monitor, Yn(CRT), equals the ambient luminance, Yambient, the adapting white can be calculated by Eq. 8.
$$ \begin{array}{l}{L}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}^{{\prime\prime} }={R}_{\mathrm{adp}}\cdot {L}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}^{\prime }+\left(1-{R}_{\mathrm{adp}}\right)\cdot {L}_{\mathrm{ambient}}\\ {}{M}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}^{{\prime\prime} }={R}_{\mathrm{adp}}\cdot {M}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}^{\prime }+\left(1-{R}_{\mathrm{adp}}\right)\cdot {M}_{\mathrm{ambient}}\\ {}{S}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}^{{\prime\prime} }={R}_{\mathrm{adp}}\cdot {S}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}^{\prime }+\left(1-{R}_{\mathrm{adp}}\right)\cdot {S}_{\mathrm{ambient}}\end{array} $$
Radp is the adaptation ratio to the white point of the monitor. When the ratio Radp equals 1.0, the human visual system is assumed to be fully adapted to the white point of the monitor and none to the ambient light. Conversely, when the ratio is 0.0, the human visual system is assumed to be completely adapted to the ambient light and none to the white of the monitor. These two extreme cases assume that the human visual system is at single-state chromatic adaptation. For mixed chromatic adaptation, 0.6 is chosen for Radp in the S-LMS model.
With the newly defined white points for the softcopy images, the von Kries chromatic adaptation model is applied. The cone signals after adaptation are calculated as Eq. 9.
$$ \begin{array}{l}{L}_s={L}_{\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}/{L}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}^{{\prime\prime}}\\ {}{M}_s={M}_{\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}/{M}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}^{{\prime\prime}}\\ {}{S}_s={S}_{\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}/{S}_{n\left(\mathrm{C}\mathrm{R}\mathrm{T}\right)}^{{\prime\prime}}\end{array} $$
For hardcopy images, the simple von Kries chromatic adaptation without incomplete chromatic adaptation and mixed chromatic adaptation is used (Eq. 10). The paperwhite is chosen as the reference white, because the eye tends to adapt according to the perceived whitest point of the scene.
$$ \begin{array}{l}{L}_s={L}_{\left(\mathrm{Print}\right)}/{L}_{n\left(\mathrm{Print}\right)}\\ {}{M}_s={M}_{\left(\mathrm{Print}\right)}/{M}_{n\left(\mathrm{Print}\right)}\\ {}{S}_s={S}_{\left(\mathrm{Print}\right)}/{S}_{n\left(\mathrm{Print}\right)}\end{array} $$


The S-LMS mixed adaptation model is designed to integrate into the CIECAM02 model, which is developed for color management applications. Hence, the S-LMS model should be applied to extend the CIECAM02 model for use in cross-media color reproduction when mixed mode illumination is employed. The input parameters required are XYZ tristimulus values of the white point of the monitor, the adopted white for the ambient light, and the paper white of hardcopy.



  1. 1.
    Berns, R.S., Choh, K.H.: Cathode-ray-tube to reflection-print matching under mixed chromatic adaptation using RLAB. J. Electron. Imaging 4, 347–359 (1995)ADSCrossRefGoogle Scholar
  2. 2.
    CIE Publication 159: A colour appearance model for colour management systems: CIECAM02. (2004)Google Scholar
  3. 3.
    CIE Publication 162: Chromatic adaptation under mixed illumination condition when comparing softcopy and hardcopy images. (2004)Google Scholar
  4. 4.
    Henley, S.A., Fairchild, M.D.: Quantifying mixed adaptation in cross-media color reproduction. In: Proceedings of the 8th IS&T/SID Color Imaging Conference, pp. 305–310. (2000)Google Scholar
  5. 5.
    Katoh, N.: Practical method for appearance match between soft copy and hard copy. SPIE Proc. 2170, 170–181 (1994)ADSCrossRefGoogle Scholar
  6. 6.
    Katoh, N.: Appearance match between soft copy and hard copy under mixed chromatic adaptation. In: Proceedings of IS&T 3rd Color Imaging Conference, pp. 22–25. (1995)Google Scholar
  7. 7.
    Katoh, N., Nakabayashi, K.: Effect of ambient light on color appearance of soft copy images –color management on the network. In: Proceedings of AIC Color 97 Kyoto 2, pp. 582–585. (1997)Google Scholar
  8. 8.
    Katoh, N., Nakabayashi, K.: Applying mixed adaptation to various chromatic adaptation transformation (CAT) models. In: PICS Conference Proceedings, pp. 299–305. (2001)Google Scholar
  9. 9.
    Shiraiwa, Y., Hidaka, Y., Mizuno, T., Sasaki, T., Ohta, K., Usami, A.: Matching of the appearance of colors in hard-copy and soft-copy images in different office environments. SPIE Proc. 3300, 148–158 (1998)ADSCrossRefGoogle Scholar
  10. 10.
    Sueeprasan, S., Luo, M.R.: Applying chromatic adaptation transforms to mixed adaptation conditions. Color Res. Appl. 28(6), 436–444 (2003)CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2016

Authors and Affiliations

  1. 1.Department of Imaging and Printing TechnologyChulalongkorn University Intellectual Repository, Chulalongkorn UniversityPathumwanThailand