|
Cool, I was hoping it was correct.
What Ive done is write a small function where I
pass a Lab value as cmsCIELAB and a profile handle and it returns the Delta E,
It converts to 16 bit PCS then dows a double transforms and the checks the Delta
E.
Is there any change this could be included in
little CMS as a stnadard function, I would assime you could just to the
transforms on a single profile using the A2B and B2A tables .. and not what I
have to do and create a LAB profile plus two transforms
bool cmsLab2Device2Lab(cmsHPROFILE
hProfile,LPcmsCIELab LabIn, LPcmsCIELab LabOut);
hProfile - pointer to profile
LabIn - Desired Lab colour
LabOut - Resulting Lab colour on output device
Intent ???? - Specify rendering intent to be used
This allows the freedome to choose the DeltaE
method
----- Original Message -----
Sent: Saturday, March 27, 2004 7:41
AM
Subject: Re: [Lcms-user] Gamut Checking
for creating colour charts
Hi,
This is more or less the algorithm littlecms is
using for gamut checking.
The differences would be, I'm using relative
colorimetric instead and the "Lab2" value is checked again across the profile
to track white point displacements, but the basics are same.
It works to me. If anybody knows a better method,
and want to share it, I would be glad to include it in the CMM.
Hue is an angle, so the range goes 0-360. Chroma
can be any positive number.
Regards,
----- Original Message -----
Sent: Thursday, March 25, 2004 11:56
AM
Subject: [Lcms-user] Gamut Checking for
creating colour charts
Im in the process of creating some custom
colour charts for work, Theis is be based on LCH. I dont want to print
all the colours in the LCH colourspace as this is too many and most are
out of gamut (3600 colours at 360 step 10 and L/C step
0.1) I want to limit the range to
colours inside the gamut of the device. I have working code based on
the below steps but I just want to check if this is the correct
way
Firstly im converting from LCH to D50 LAB and
calling this LAB1
Then im Converting from LAB1 to Device Colours
via Absolute Colourmetric transform, Im then
taking this Device Colour and converting it back from the Device
colour via Absolute Colourmetric transform and calling this LAB2.
Im then discarding all colours where the Delta
E between LAB1 and LAB2 > 20 (The ensures im getting colours just outside
the gamut)
Is this the best way?
Oh BTW is the range for Chroma and Hue H
0.0 to 1.0?
| 