Estimate RGB image color temperature

* Read in an RGB image, assuming it is sRGB display ready
* Convert them to XYZ
* Calculate the mean chromaticity of the top 10 percent of the Y
values
* Calculate the chromaticity of a white surface under different
color temperature blackbody radiators
* Find a close match and return the color temperature

See also sceneFromFile

Contents

Create a scene with a low color temperature

scene = sceneCreate('macbeth tungsten');
oi = oiCreate; oi = oiCompute(oi,scene);
sensor = sensorCreate;
sensor = sensorSet(sensor,'fov',sceneGet(scene,'fov'),oi);
sensor = sensorCompute(sensor,oi);
ip = ipCreate; ip = ipCompute(ip,sensor);
% ipWindow(ip);

rgb = ipGet(ip,'srgb');
cTemp = srgb2colortemp(rgb);

disp(cTemp)

        3500

Not quite perfect for the MCC because, we think

the white isn't really white

scene = sceneCreate('macbeth d65');
oi = oiCreate; oi = oiCompute(oi,scene);
sensor = sensorCreate;
sensor = sensorSet(sensor,'fov',sceneGet(scene,'fov'),oi);
sensor = sensorCompute(sensor,oi);
ip = ipCreate; ip = ipCompute(ip,sensor);
% ipWindow(ip);

rgb = ipGet(ip,'srgb');
[cTemp,cTable] = srgb2colortemp(rgb);

disp(cTemp)

disp(cTable(:,1))
disp(cTable(:,2:3))

        5500

        2500
        3000
        3500
        4000
        4500
        5000
        5500
        6000
        6500
        7000
        7500
        8000
        8500
        9000
        9500
       10000
       10500

    0.4766    0.4140
    0.4367    0.4044
    0.4051    0.3911
    0.3803    0.3771
    0.3607    0.3639
    0.3451    0.3520
    0.3325    0.3415
    0.3221    0.3322
    0.3136    0.3241
    0.3065    0.3171
    0.3004    0.3108
    0.2953    0.3054
    0.2909    0.3005
    0.2871    0.2962
    0.2837    0.2924
    0.2808    0.2889
    0.2782    0.2858

END

Published with MATLAB® R2024a