[argyllcms] Re: using a digital camera as a colorimeter

  • From: "Alastair M. Robinson" <blackfive@xxxxxxxxxxxxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Fri, 07 Oct 2005 00:29:19 +0100


Stephan Bourgeois wrote:

I then used a digital camera (Kodak DX6440) under direct sunlight and got an *excellent* result, with almost no color cast. This camera doesn't have a "raw" mode (only sRGB JPEG), and there is some clipping on the primary colors, as one would expect in the sRGB space. No doubt I will do this again with a camera that gives more control and more range... I have only tested the profile with "photographic" images, I haven't tested it against very saturated color patches that may reveal errors due to the clipping.

That's very interesting.
Did you photograph the IT8 target and build a profile for the camera, or "trust" the camera's sRGB space?

If the former, did you have the IT8 target and Argyll target in the same shot?

My interpretation for those results is :
a. the light source could be a more important factor in color rendition of these devices than the spectral response of the RGB filters. The scanners I used offer the worst light source i.e. cold cathode tube, that looked very "spikey" in a handheld spectroscope.
b. the scanner seems to clip highlights when using a paper with fluorescent whitening agents. The clipping is different with a camera.

I'm told that using Vuescan can help to avoid the clipping problem (due to better exposure control), which in turn helps minimise the colour cast.

I had a very interesting conversion with Stephen Tweedie a couple of months back, and he showed me a program he'd knocked up to do some jiggery-pokery on an Argyll .ti3 file and perform a rough FWA correction, using just XYZ sample data.

Here are my questions:
1. Please comment on this.

See above :)

2. What would be the difference between using a colorimeter (eg. Color mouse CM2C) and a digital camera with a good quality light source (xenon strobes)?

I imagine that the colourmouse would have a much wider gamut. I have one myself, and have had good results generally.

The colourmouse, AFAICT has a fairly simple light sensor (possibly even monochrome), and cycles between several different coloured LEDs. Using the amount of light reflected from the patch for each colour, the driver software figures out an XYZ value for the measured patch.

3. Can I use the peak error and average error reported by the "profile" command as an indicator of the accuracy of the color measuring device?

No - the only way to do that is to measure some patches of colour, convert them to the printer's colour space using absolute colorimetric intent, print them, and compare the printed patches.

The peak and average error merely tell you well Argyll's generated tables fit the original data in the .ti3 file. With the colourmouse, and targets of 120 - 150 patches, I seem to get a peak error of about 1.3, and average of 0.8.

What sort of values do you see using your digital camera trick?

All the best,
Alastair M. Robinson

Other related posts: