X rite colorchecker passport magenta tint

Started 4 months ago | Discussions thread
Mark Scott Abeln
Mark Scott Abeln Forum Pro • Posts: 18,137
Re: X rite colorchecker passport magenta tint

Steve BB wrote:

Mark Scott Abeln wrote:

Steve BB wrote:

Are colour profiles meant to be 'exact'


ok good to know

The biggest problem is that any particular visible color may be made up of a large number of different varieties of wavelengths. For example, a dark gray may be made up of all visible light frequencies equally, or it may be made up of only three narrow frequencies, or any combination of frequencies that balance so that it *appears* gray. If we are talking about gray paints, if they all appear to be the same under a given light source, then we say that the paint colors are metamers of each other, even though the precise combination of spectral frequencies they reflect happen to be different. As colors become more saturated, the range of potential metamers becomes quite narrow, until you get to spectral colors where there is only one.

The trouble is, real-life objects may very well not have the same spectral reflectance as the color patches on the ColorChecker target, even though they *appear* to be the same color, at least under a particular lighting. It's entirely possible that under a different light source (let's say in shade versus incandescent lighting) that these two colors won't match, and that's called metamerism failure. In consumer products, where components may be made of different materials, but it is desired that they always appear to be the same color, then it is important that there is a metameric match at least under common lighting conditions.

Cameras don't have the same spectral sensitivity as human vision, and this can lead to mismatch of colors, which is called device metamerism failure. Two materials, identical to the human eye, may appear different to the camera, while two materials identical to the camera may appear different to humans. This is the basic color difference between different models of camera, even if you take great care to shoot them under controlled conditions with custom processing. You can force colors to be anything that you want, but there will be metamerism failure that varies between models.

All of this color matching must be done under controlled conditions. A small neutral patch surrounded by a large brightly colored surrounding will *appear* to be different from the same patch surrounded by a large black area, which will in turn appear to be different from the same patch surrounded by a white. Also, colors, say in a print, will look different depending on the light levels and color of light falling upon the print. These effects are the source of a lot of optical illusions, but unfortunately all of this is hard to quantify and put into a general model of human vision. Cameras don't take "color appearance models" into account, and how could they? They can't predict the final viewing conditions of the image, and instead that is a part of the art of photography.

As mentioned, cameras don't have the same spectral sensitivity as does human color vision, and converting the raw values from the sensor to human color is not something that can be done precisely, at least without adjusting spot colors, which has its own problems. It's something like trying to fit a line to data in statistics: some values are going to fit well, and some are not, and it is something that requires a bit of engineering and artistic judgement. Do you want to fit most of colors as close as you can, even if some will poorly match, or do you want to minimize the worst match, even then no color is really good? Or do you want to force all colors on the ColorChecker to match exactly? This last solution is very likely to lead to problems with *other* colors not found on the ColorChecker, or it may lead to major hue shifts with exposure (this is called a 'twisted' profile). The only way to get a good match for all colors is to have a sensor that satisfies the "Luther-Ives condition" and none of them do, nor is this foreseen anytime in the near future.

or do they just help, and manual color corrections are to be expected afterwards?

Yes, unfortunately.

It helps if you use a "spectrally flat" light source, such as skylight, incandescent bulbs, and flash, which is close enough. You'll definitely have problems if fluorescent or LED bulbs are used.

Im using Bowens XMT 500s

Strobes are a "gray body" light source, but they are close enough to a spectrally flat light source for government work.

 Mark Scott Abeln's gear list:Mark Scott Abeln's gear list
Nikon D200 Nikon D7000 Nikon D750 Nikon AF-S DX Nikkor 35mm F1.8G Nikon AF Nikkor 28mm f/2.8D +4 more
Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow