2
\$\begingroup\$

Many different software (including Lightroom) can correct lateral chromatic aberration without having a special profile for the lens. How is the correction done?

Adobe's method for doing it when a lens profile is available is described here. A slightly different distortion correction is applied to the red, green and blue channels. Generally, distortion correction is applied to the green channel as normal and additional small distortion corrections are used for the red and blue ones.

But what about the case when there is no lens profile or no CA correction information included in the profile?

\$\endgroup\$
1
  • \$\begingroup\$ There are two types, they accompany each other. You might succeed adjusting the red, green, and blue images making them superimpose. However now you deal with longitudinal chromatic aberration. It is a plague, if you succeed making a color-correct lens in the sense that all colors have to the same focal length, you still get a rainbow effect. It’s a drawn-out rainbow radially from the axis. The length is proportional to image size. The circles under a microscope, reveal they are made up of sub-circles of color. This results in a variation of colors that cannot be corrected by a resizing. \$\endgroup\$ Commented May 2, 2016 at 21:55

2 Answers 2

3
\$\begingroup\$

It is only a very slight over-simplification to describe lateral chromatic aberration as the situation in which different colours produce pictures of slightly different sizes. Seen in that light, it's merely a matter of adjusting (slightly) the image size in one or more channels so that everything comes back into registration. In this article on The Online Photographer, the master printer Ctein describes the adjustment process he used to correct chromatic aberration in a film image when making a dye transfer print. With digital, it's a matter of scaling rather than adjusting the height of an enlarger head, but the end result is the same.

Incidentally, it's because the scaling/size difference is different in front of and behind the plane of sharp focus that longitudinal chromatic aberration is difficult to correct. You need to know what lies where in the picture's depth in order to make the appropriate changes. It's also why lateral CA is much more apparent at the edgs of an image than nearer the centre.

\$\endgroup\$
-1
\$\begingroup\$

Transverse chromatic aberration: The focal length a measurement, lens to image plane, when imaging a far distant subject. The camera lens is a converging lens. Light from an object at infinity enters the lens as parallel rays. The lens bends these inward. The path of the image forming rays trace out a cone. The camera is focused. This adjusts the distance, lens to image. The image will be in focus when the apex of the cone just kisses the imaging chip (film). Now all lenses suffer from 7 aberrations. These are mitigated by the optician by constructing the lens using several different shaped lenses, some dense glass, some less dense. Chromatic aberration is mitigated by combining a positive power lens with a weak negative. They have opposite chromatic aberrations so they almost but not quite cancel each other.

Transverse chromatic aberration is also called chromatic variation of focal length. What happens is: The distance lens to image is a function of light frequency. Violet comes to a focus near the lens while red’s focus is further downstream. All other colors such as green, yellow, and orange come to a focus in an intermediate position. The further the image forms from the lens, the bigger it is. Thus the red image is largest, violet the smallest. What we see is a color fringing, a rainbow of color surrounding the boundaries of each object being imaged. Again we can mediate but not eliminate with transparent lenses. The mirror lens, with a first surface mirror with slivering on the surface shows no chromatic aberrations as the light does not transvers the lens.

As to how software can mitigate. The rainbow pattern surrounding the subject is identifiable and thus the software can mitigate. The software looks at adjacent pixels and interpolates between. In a way the results are fabricated as this is just a best possible guess as to the colors near the periphery of objects.

\$\endgroup\$
3
  • 2
    \$\begingroup\$ The question is about C/A correction, not C/A itself. You're burying the lede in your 3rd paragraph, and you're not particularly detailed on HOW the correction is accomplished. And can you cite a source for the information? \$\endgroup\$
    – inkista
    Commented May 1, 2016 at 19:15
  • \$\begingroup\$ @inkista I feel strongly that the point of this website is to educate. The source is me Alan Marcus. \$\endgroup\$ Commented May 1, 2016 at 21:13
  • \$\begingroup\$ You do get that 'cause I said so doesn't actually buttress your case...? I've been challenged to produce citation, and was able to do it easily. \$\endgroup\$
    – inkista
    Commented May 7, 2016 at 18:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.