Re: Colorimetric Accuracy in the Field
Re: Colorimetric Accuracy in the Field
- Subject: Re: Colorimetric Accuracy in the Field
- From: Robin Myers <email@hidden>
- Date: Wed, 05 Jun 2013 13:58:27 -0700
The issue was what Bruce Fraser did in his test, which was to perform a relative colorimetric shift on the RGB coordinates for the ColorChecker neutral patches. He did not explain this, but it is evident.
Using non-equal RGB values for the neutral patches would be an absolute colorimetric case and does not apply to Bruce Fraser's test.
There are times where each of these cases apply. Screwdrivers for screws, hammers for nails.
Robin Myers
On Jun 5, 2013, at 1:41 PM, Ben Goren wrote:
> On Jun 5, 2013, at 1:30 PM, Robin Myers <email@hidden> wrote:
>
>> Since they are slightly yellow, their XYZ values lie on an axis defined by D50 but shifted slightly by their amount of yellow reflectance.
>
> Yes, exactly.
>
> But, to my mind, a more clear way of stating that is that the N{whatever} patches on the ColorChecker aren't actually neutral; they just have very low saturation.
>
> Which means that you want them to be rendered into an RGB file not as neutral R=G=B patches (as Bruce Fraser did) but instead as patches with (specific) RGB values very close to each other but not actually equal.
>
> Would you agree that a spectrophotometer is very unlikely to measure the N9.5 patch on Mr. Fraser's chart as L*=96.14 a*=0.00 b*=0.00, and yet Mr. Fraser's profile incorrectly rendered that patch as ProPhoto RGB = 241,241,241 = L*=96.14 a*=0.00 b*=0.00?
>
> Cheers,
>
> b&
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden