• Open Menu Close Menu
  • Apple
  • Shopping Bag
  • Apple
  • Mac
  • iPad
  • iPhone
  • Watch
  • TV
  • Music
  • Support
  • Search apple.com
  • Shopping Bag

Lists

Open Menu Close Menu
  • Terms and Conditions
  • Lists hosted on this site
  • Email the Postmaster
  • Tips for posting to public mailing lists
Re: Target for camera profiling?
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Target for camera profiling?


  • Subject: Re: Target for camera profiling?
  • From: David Iannarelli <email@hidden>
  • Date: Tue, 20 Jan 2004 05:37:06 -0500

Hi Bertho,

It sounds like I should play and practice with some of the 5" x 7" targets that
I have while waiting for Gretag's update since I already spent too much money on
the Gretag software that I bought.


I know what you mean. I thought for $4,00.00 I was getting the camera profiling option as well. Oh no, $700.00 more for that little ad on. I was told to buy the ColorEyes product because it was better at camera profiling. There was a $250.00 version of ColorEyes that used the regular Gretag color checker. I tried that version first. It built an okay profile but not good enough for a professional to use. You got color shifts in some areas of the image like saturated colors and skin tones. So, I bought the upgrade and have been happy with the profiles I have made. The regular color checker just does not have enough colors on the target for the profiling softwares to map out an accurate characteristic of every possible color the camera can capture then build that perfect profile from the data. I have not experimented much with the scanner targets for making a camera profile.

One thought: Is it possible and practical to buy the just the target from
ColorEyes and use it with the GM ProfileMaker Pro?


I do not believe they sell the target separately. I have spoke with the guys at integrated color and from their comments about the program there is some customized calculations going on when using their target and software together. No target has a black as black as the camera can record and many targets do not have enough saturated and skin color patches for building that perfect profile. The software has to help. Each software developer has their own philosophy on the best way to handle all the variables.

I came up with a neat trick for digital cameras when setting up the lights for
even target illumination:

I use the studio strobes in two soft boxes at 45 degree angle and replace the
target with a uniformly white sheet and I set the camera to show "flashing high
lights". It is very fast and easy to see the uniformity, or lack thereof, of
the lighting. The same can be done in tethered mode with the new curves display
in NC-4. Then if tethered, a final eye dropper check and it will be perfect.
Bertho


Is the color values the same in all the corners? If one of your lights has a slightly different color temperature it can effect how the profile is made. Say the patches on the left 1/4 of the target has a few points more red and blue light on them. The profile software thinks the camera is the reason for this. The profile will be made to compensate for the colors only from the effected area.

There is a chance that the camera may be less accurate in the outer edges of the sensor. Try not to fill the entire frame with the target to be on the safe side.

... the DC seems to have some
metamerism problems in some of the patches

Did you find that out through some CIE Metamerism Index determination?

The eastern regional sales manager confirmed the rumor when I asked him about it.


I like Bertho's idea of photographing one patch at a time. Then building the target back in Photoshop. Not much of a chance for unevenness with this approach. Not sure if the profile would be any better for all that work though. I have also herd photographers building custom camera profiles using paint swatches. If you were photographing paintings for artists that would be a very useful profile.


I guess there are differences in philosophy regarding the color temperature issue, or I'm missing something. If there is a big effect on the profile from the difference in strobe color temperature, is a profile made from illumination with a single strobe more useful than a profile made from an exposure averaging 2, 3, or 4 strobes, if that's what you typically use?


Try all the options. It makes sense if you have a lighting setup that you use enough to make a profile for that lighting. A general profile for every day work would be best made from one light. As Jack said the key is proper white balance. The custom profile will map the colors accurately in the areas of the image that match the white balance.

This ICC profiling technology has only just begun. The more we experiment and share our ideas the more we will all benefit.


Regards,

David Iannarelli
_______________________________________________
colorsync-users mailing list | email@hidden
Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/colorsync-users
Do not post admin requests to the list. They will be ignored.

  • Follow-Ups:
    • Re: Target for camera profiling?
      • From: "Bertho Boman" <email@hidden>
  • Prev by Date: Re: ColorSync utility question
  • Next by Date: Re: Target for camera profiling?
  • Previous by thread: Re: Target for camera profiling?
  • Next by thread: Re: Target for camera profiling?
  • Index(es):
    • Date
    • Thread