Re: "Visual Color Difference" chart [Re: (no subject)]
Re: "Visual Color Difference" chart [Re: (no subject)]
- Subject: Re: "Visual Color Difference" chart [Re: (no subject)]
- From: <email@hidden>
- Date: Tue, 16 Jan 2007 23:36:01 -0500
Terry,
Here is yet another way to define a target:
Define a color list such as:
L C h
83,8 65,4 116
85,8 65,4 116
87,8 65,4 116
85,8 57,8 116
85,8 65,4 116
85,8 73,6 116
85,8 65,4 112,5
85,8 65,4 116
85,8 65,4 119,5
In this example, the reference is Pantone 374 C with
L*C*h* coordinates of 85.8/65.4/116 (D50). I have selected
the L*C*h* representation because it is best suited for an
error target such as this one. (I will not elaborate here
on how to get Pantone L*C*h* values since this is not the
topic.)
The reference appears three times in the list. The first
three colors represent a variation in Lightness, the next
three a variation in Chroma, and the last three a
variation in hue.
The values have been adjusted to represent a DeltaE of
"almost exactly" 2 with reference to CIE94 (not CIELAB).
This file was made with Microsoft Excel and exported as a
TAB delimited format, but it could have been written with
Text Edit (Mac) or WordPad (Windows).
Copy the ten lines above and paste in TextEdit or WordPad
(the Tabs are now spaces, but do not change them, and no
need to change the decimal separator to periods). Open the
file with PatchTool, dowloaded from
http://www.babelcolor.com/main_level/download.htm
If you click on a patch in the middle row then move the
mouse over the other patches, you will see the DeltaE in
the "Info" window (select CIE94 for the color difference).
I have attached an exported images in JPEG(sRGB) but
TIFF-Lab 8 and 16 bit is also possible. (Note: you cannot
export an image in the trial mode).
In order to fine-tune the above list, I simply opened it
in PatchTool and looked at the CIE94 color difference. A
few iterations were required to get the final values.
Of course, when you get the target image you want, you
still need to print it and measure it for conformance. A
few iterations may be required there also, even for a
"calibrated" printer. And, as mentioned in a previous
post, you should have controlled viewing conditions as
metamerism can play games when comparing a printed target
with manufactured items.
Danny
email@hidden
www.BabelColor.com
On Fri, 12 Jan 2007 09:15:39 -0500
Terry Wyse <email@hidden> wrote:
On Jan 12, 2007, at 8:35 AM, Graeme Gill wrote:
Terry Wyse wrote:
It would be awesome if someone could come up with a
calculator
that would calculate all the possibilities for a delta
e of <x>
away from a reference. Humm...I may have to do some
poking around
and see what can be done.
How could you do that, given that are an infinite number
?
Well, I was thinking it could be limited to a certain
"resolution" or precision. For my purposes, limiting it
to whole Lab integers or one decimal place would be
enough.
Terry
_______________________________________________
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
Attachment:
target_jpg_sRGB.jpg
Description: image/pjpeg
_______________________________________________
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