Chromatic Aberration - DPChallenge Forums

03/19/2004 02:31:21 PM · #1

Has anyone found a solution to "repairing" this other than the "Hue/Saturation-Magenta" fix I have read about?

How To Repair Chromatic Aberration with Photoshop

Before:

After:

Is there a lens filter to help reduce this effect?

I have read articles that say even the higher end lenses can give the same results to some extent.

Note:
I said "repair".
I don't want the $1,500-$15,000 Canon lens solution LOL.

P.S. A big thanks to Crabappl3 for helping me find this fix and info :)

Message edited by author 2004-03-19 14:41:04.

Gordon

RIM BlackBerry Curve

Canon EF 85mm f/1.8 USM

03/19/2004 03:14:16 PM · #2

If you shoot RAW, and use Photoshop CS, there is adjustable CA repair built in to the converter. If you have some RAW samples I'd be happy to have a play with it to show/ see if its any good.

Mostly its a function of the quality of the lens used, and in particular the air interfaces - so the $1000 lens 'fix' does work :)

03/19/2004 03:26:52 PM · #3

I got this seal shot from a friend that emailed it to me as a jpg.
I do not think he shot it in RAW mode.

I do not have a RAW file of my own at this time but if I come across one I will let you know.

What do you mean by "and in particular the air interfaces"?

Originally posted by Gordon:

Gordon

RIM BlackBerry Curve

Canon EF 85mm f/1.8 USM

03/19/2004 03:37:17 PM · #4

Originally posted by Calvus:

What do you mean by "and in particular the air interfaces"?

My poorly explained understanding is:

CA is mainly caused by light of different wavelengths (colour) being refracted more or less than other wavelengths. Typically this happens the most at air/glass boundaries, where the refractive index is higher - all sorts of funky coatings and glass doping is used to adjust these indices in the more expensive lenses - which reduce the amount that different wavelengths diverge.

Lenses are made up of elements, that are packed together in groups. A group of elements are sandwiched together, so less air/glass boundaries to cause CA. Fewer groups is generally better for CA, for equivalent coatings/ glass composition etc.

I'm sure someone can explain it better than I can today.

Or as an alternative visualisation - consider a prism spliting a beam of white light - e.g., cover of a Deep Purple album.

Light refracts at each of the glass/ air boundaries (the beam spreads more) You want your lenses to do that as little as possible - when it happens, CA occurs.

Message edited by author 2004-03-19 15:38:23.

Olympus OM-D E-M5 Mark III

03/19/2004 03:52:12 PM · #5

I have read and understood the wavelength portion I just wasn't sure how you associated the "Air" portion into it, now I do.

Thanks.

Originally posted by Gordon:

Originally posted by Calvus:

What do you mean by "and in particular the air interfaces"?

My poorly explained understanding is:

CA is mainly caused by light of different wavelengths (colour) being refracted more or less than other wavelengths. Typically this happens the most at air/glass boundaries, where the refractive index is higher - all sorts of funky coatings and glass doping is used to adjust these indices in the more expensive lenses - which reduce the amount that different wavelengths diverge.

Lenses are made up of elements, that are packed together in groups. A group of elements are sandwiched together, so less air/glass boundaries to cause CA. Fewer groups is generally better for CA, for equivalent coatings/ glass composition etc.

I'm sure someone can explain it better than I can today.

Or as an alternative visualisation - consider a prism spliting a beam of white light - e.g., cover of a Deep Purple album.

Light refracts at each of the glass/ air boundaries (the beam spreads more) You want your lenses to do that as little as possible - when it happens, CA occurs.

GinaRothfels

Olympus 8mm f/1.8 M.Zuiko Digital ED Fisheye PRO Lens

03/19/2004 03:56:59 PM · #6

I didn't know there was a way to repair it. Thanks for the link.

dadas115

Canon EOS-1000D Rebel XS

Canon EF-S 60mm f/2.8 Macro USM

03/19/2004 04:24:57 PM · #7

I am not even sure if I would call what you had in your pictures chromic aberration, at least not in the sense that most people use the term when talking about digital photos. When I think about CA it is more along the lines of what is seen in this picture:

//www.pbase.com/image/25120162

There are several things that can cause this to happen. One of the most common causes is due to light of different wavelengths being refracted to differing degrees as they pass through the various glass elements. The effect is usually more obvious with longer focal length lenses. Usually lens manufacturers use fluorite and other low refraction and low dispersion elements to combat this problem. This is also one of the functions of diffractive optics. In digital cameras it can frequently be caused by sensor blooming. It is my understanding that the amount of glass-air interfaces leads to problems like light loss, ghosting and flare.

Greg

kirbic

Canon EOS R5

Canon RF 24-70mm f/2.8 L IS

03/19/2004 04:27:01 PM · #8

First part of this reply has to do with the repair of CA. In many cases you can reduce CA in a previously-captured image if you know how the CA is distributed spatially. It is usually worse at corners (varies radially), however shooting at an angle through a pane of glass, e.g. into an aquarium, can introduce CA that varies across a frame. It's usually not possible to completely eliminate it, but it's often possible to reduce to acceptable levels.
You can investigate CA in an image in photoshop by going to the "channels" tab and clicking on the inidvidual color channels. by rpaidly clicking through R, G & B channels, you can note how the position of a detail shifts. by doing this in various areas of an image, you can get an idea of what rescaling and/or movement of the channels will be required to minimize CA.
A more sophisticated approach:

//www.dolabs.com/pressroom/release_dxo_opticspro.html

In the image shown, I can find no significant CA. The large color fringe effects on the OOF rocks are caused, IMO, by mixing of the non-neutral tones of the rocks with the more neutral, lighter background color. It might be best to just desaturate selectively to eliminate this.

dadas115

Canon EOS-1000D Rebel XS

Canon EF-S 60mm f/2.8 Macro USM

03/19/2004 04:37:45 PM · #9

I donâ��t know if it is just my eyes playing tricks on me but the after picture doesnâ��t appear to be quite as sharp as the before picture.

Greg

dwoolridge

03/19/2004 05:47:54 PM · #10

Originally posted by Gordon:

Or as an alternative visualisation - consider a prism spliting a beam of white light - e.g., cover of a Deep Purple album.

Does Gordon mean Pink Floyd's Dark Side of the Moon.?