Using http://www.easyrgb.com/index.php 
CIE-L*ab   =   48.000  -14.000   25.000
equals
RGB 0-255  =   105.30   119.53    70.87
RGB 0-0.1  =  0.41293  0.46875  0.27791

XYZ        =   13.654   16.794    8.461
Illuminant =  D65
Observer   =  2° (1931)

So Ryan was correct to say the values are scaled between 0 and 1 (I was using the wrong row in my re-conversion to 0 to 255 scale)
This is the 1st time I've come across RGB scaled to 0 and 1. Is its common in python and does the other scikit-image functions handle the 0 -1 scaled values or should I work with 0 to 255 values to be safe?



On Mon, 27 Mar 2017 at 14:36 Michael O'Brien <mob.irl@gmail.com> wrote:

I'm not sure what illuminant colormine uses but if you multiple by 255 you get 151,154,141 which it getting closer to what I expect (greenish colour) with 105,119,70 still being much closer


On Mon, 27 Mar 2017, 14:22 Ryan Anderson, <randerson@resonon.com> wrote:
It looks to me as if lab2rgb is returning colors scaled as floats between 0 and 1 while colormine.org is returning 8 bit integers scaled 0-255.

Try multiplying by 255 and comparing the outputs.

On Mar 27, 2017 7:02 AM, "Michael O'Brien" <mob.irl@gmail.com> wrote:
Hi all,


If I use the following array of LAB values 
array([[[ 48.07851346, -14.69400135, 25.2170507 ],
 [ 63.21232753, -3.65444198, 6.67734193],
 [ 45.80027308, -2.47952304, 11.5073341 ],
 [ 46.84900559, -10.37025167, 19.92527996],
 [ 44.92476952, 19.36861848, 2.97238555],
 [ 35.50132805, 5.89086564, 1.95259827]]])
and use the following code to pass the array to lab2rgb
rgbcolours =color.lab2rgb(ClusterColours.values[np.newaxis])
I get the array
array([[[ 0.40929448, 0.47071505, 0.27701891],
 [ 0.59383913, 0.60611158, 0.55329837],
 [ 0.4393785 , 0.4276561 , 0.34999225],
 [ 0.4159481 , 0.4516056 , 0.3026519 ],
 [ 0.54449997, 0.36963636, 0.4001209 ],
 [ 0.36970012, 0.3145826 , 0.315974 ]]])
but on colormine.org if I enter the 1st of the LAB values I get 105,119,70 for RGB and not colours so close to black
Where am I going wrong?

_______________________________________________
scikit-image mailing list
scikit-image@python.org
https://mail.python.org/mailman/listinfo/scikit-image