Re: Color Measurement
Re: Color Measurement
- Subject: Re: Color Measurement
- From: Robin Myers <email@hidden>
- Date: Wed, 15 Jul 2009 11:24:59 -0700
On Jul 15, 2009, at 8:12 AM, Mike Eddington wrote:
<snipped>
What the instrument then reads on substrates *other* than the
tiles is "part of the game". What is the alternative, in your
opinion?
Other than using the same make and model (and rev level) I'm not sure
there is one yet. Perhaps a model of instrumental measurement
correlation or "normalization" as suggested here:
http://www.npes.org/pdf/WhitePaper.pdf, but I'm afraid my propeller
doesn't spin quite that fast this morning. Would it be feasable to
have
corrections factors for instruments based on media? One would need a
reference instrument to align everything toward.
This is an excellent paper and well worth reading for anyone trying to
improve inter-instrument agreement. Take special note of Table 4, the
results, which show a marked improvement, and Sections 5 and 6 which
discuss the differences for which the model cannot account and the
final results.
I'm more a realist on the issue, and I don't want to make a mountain
out of a mole hill, just to emphasize that using tight tolerances with
different instruments can lead to issues like the one that originated
this thread.
The way I see it, in dealing with tolerances and differeing
measurement
devices (assumed to be within manufacturer's specs), as in the origial
post, there are only a few options without having to cause undue
cerebral stress.
1.Adjust your target baselines for the new device (as Jon Crook
suggested)
2.increase your tolerances to account for instrument differences, or
perhaps use dE2000
This should produce no improvement. Changing the metric will not
improve the inter-instrument agreement. It would be like changing from
an Imperial to a metric tape measure, the units change but the item
being measured does not.
3.Re-calibrate the output device using the new measuring device.
Likely
there won't be significant visual differences...probably...maybe.
Actually, recalibrating the new measurement to the older instrument
should improve the inter-instrument agreement, as long as the
instruments have similar measuring geometries, etcetera.
4.ignor it
Robin's suggestion is a good one, and might certainly help the issue,
but I just don't think alignment to reference tiles will automatically
equate to alignment on differing surfaces/substrates.
If you are referring to similar measuring geometries in the
instruments and the substrates are similar (e.g. paper), then a model
can be developed that will greatly improve the inter-instrument
agreement, as noted in Dr. Rich's paper. However, attempting to get
0/45 degree measurements to agree with integrating sphere specular-
component-included measurements for any surface is extremely difficult.
If you refine your requirements for measurement situations, it might
be possible to create a set of calibrations that will allow you to
refer one instrument to another, or both instruments to a reference
standard.
Robin Myers
_______________________________________________
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