Excessive Chromatic Aberration + Inexperienced Photographer

A forum to ask questions, post setups, and generally discuss anything having to do with photomacrography and photomicroscopy.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

bralex
Posts: 129
Joined: Mon Jan 22, 2018 11:34 am
Location: New Mexico, USA

Excessive Chromatic Aberration + Inexperienced Photographer

Post by bralex »

I've been trying my hand at photographing snowflakes as opportunity permits. I'm not a very experienced photographer, and am at a bit of a loss to improve one particular problem - excessive CA in these stacks.

First, equipment. I have an Olympus OM-D EM-5 Mk II, Olympus 40-150mm PRO, 1.4x teleconverter, and a Raynox 150. Shooting on a tripod, outdoors (on my porch). Using only the (overcast) sun for lighting. Collecting flakes on a piece of felted paper mounted on a board.

Settings: I've messed with different settings; the pics below were shot at f/4, 1/13 second, iso-800. Shooting JPEG. If other info would be helpful please let me know.

Software/processing: I ran the .jpg files through Photoshop elements for auto-levels (90% of the time I'm happy with auto-levels in my stuff).

Some sample pics. These aren't the best, I just grabbed a sequence to show the CA issues I'm having. I recognize a number of other problems :)

The first pic has the background in focus, flake out of focus - CA on the out of focus portion is very apparent.

Image

Second pic has much of the flake in focus, no CA [edit: *less* CA].

Image

Third pic is the Zerene DMAP stack (14 pics). The CA, accumulated largely from the out-of-focus pics, is pretty drastic. This is clearly not a Zerene problem, though if there is a useful setting there I'd be interested.

Image

My thoughts run largely to 1) camera settings or 2) I ought to be shooting RAW and there's some software I could use to preprocess out the CA. I found the Oly Viewer software to be quite unfriendly to use, especially for stacks, but maybe I need to give it another go.

I'm very interested in any suggestions or help.

Lou Jost
Posts: 5948
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

I think these are largely lens problems. A zoom plus a teleconverter plus a close-up lens (even a very good one) could add a lot of CA.

Scarodactyl
Posts: 1618
Joined: Sat Apr 14, 2018 10:26 am

Post by Scarodactyl »

It does kind of look like something is wrong with the stack, though, as much of thr flake is still out of focus. The center seems to be in worse focus than the second picture shows.

zzffnn
Posts: 1896
Joined: Thu May 22, 2014 1:25 pm
Location: Houston, Texas, USA
Contact:

Post by zzffnn »

bralex,

So did you use that "first pic" in that "14 pics" stack? Since snow flake is mostly not in focus in "first pic" and there is lots of CA there, maybe simply don't use it in the stack?

Transparent crystal against black background seems ruthless in revealing CA. Would you be willing to use dark blue or purple background?

My Panasonic 100-300mm f/4-5.6 lens at 300mm + Raynox 250 also produces CA in high contrast areas under harsh light.

Lou,

Which budget 200mm or 300mm prime would you recommend to be used with micro four thirds sensor and Raynox 150 or 250?

I know Oly 60mm marcro is quite good and many vintage 135mm primes are very decent. But long working distance from a 300mm prime + Raynox would be nice. Native m4/3 200mm and 300mm primes are kind of expensive, though automatic focus bracketing is nice.

rjlittlefield
Site Admin
Posts: 23563
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: Excessive Chromatic Aberration + Inexperienced Photograp

Post by rjlittlefield »

bralex wrote:The first pic has the background in focus, flake out of focus - CA on the out of focus portion is very apparent.
I agree -- the out-of-focus area in front are blueish.
Second pic has much of the flake in focus, no CA [edit: *less* CA].
To my eye the CA is still pretty intense, but here it's a yellowish cast in areas behind focus. (Longitudinal CA has complementary color casts on either side of focus.)
Third pic is the Zerene DMAP stack (14 pics). The CA, accumulated largely from the out-of-focus pics, is pretty drastic.
The phrase that I've highlighted reflects some misunderstanding. DMap works by selecting pixel values from individual frames, or sometimes weighted averages of pixel values in adjacent frames. So, DMap cannot really accumulate anything. It's very different from PMax, which definitely does accumulate noise and contrast, and could conceivably accumulate color casts from out-of-focus frames while still preserving detail from focused frames. If we're seeing colored pixels that DMap selected from out-of-focus frames, then that's an indication that there's some problem with the stacking process, perhaps a matter of haloing, because ideally no pixel values would come from out-of-focus frames.
My thoughts run largely to 1) camera settings or 2) I ought to be shooting RAW and there's some software I could use to preprocess out the CA.
The problem shown here looks like straightforward longitudinal CA caused by the optics. You might try stopping down, though that would work better if you had brighter light to work with.

Unfortunately there cannot be any software that can fix longitudinal CA in general, because the information that would be needed to do that is not present in the images. So the only way software can "fix" longitudinal CA is by use of one or another heuristic that happens to do something useful for the scene at hand. One common strategy is to desaturate some range of hues that is assumed to be absent from the original scene, because those hues must the the result of CA. But if the scene could contain any hue, you're basically stuck.

For this particular subject, simply converting the result to gray-scale would not be crazy.
I'm very interested in any suggestions or help.
There are no settings in Zerene Stacker that will specifically affect its handling of CA.

However, with troublesome scenes it's always a good idea to try both DMap and PMax. Although PMax can accumulate CA from out-of-focus frames, sometimes it goes the other way and diminishes CA instead. I don't know any simple and effective way to predict, so just try them both. But at ISO 800, noise accumulation may be a killer for PMax so be prepared to see that.

--Rik

Chris S.
Site Admin
Posts: 4042
Joined: Sun Apr 05, 2009 9:55 pm
Location: Ohio, USA

Post by Chris S. »

Bralex,

Considering your situation from a wider view, I don't think a conversation about chromatic aberration really serves your needs. Instead, it looks to me as if you're facing a situation that many of us (myself included) have faced: You are looking to increase the magnifications at which you shoot, and your optics simply aren't up to the job.

From what I read about snowflake photography, most snowflakes can be photographed with a 4x or 5x lens. Over the past decade or so, many of us have adopted microscope objectives for this and higher magnifications. You wouldn't need a very expensive microscope objective to give a vast improvement over the results you're currently getting. Adding a microscope objective to your rig needn't be difficult or expensive. With an "infinite" objective, you could probably mount it with an adapter on the end of your 40-150mm zoom.

If this direction interests you, you might want to read the following posts:

FAQ: How can I hook a microscope objective to my camera?
Lenses for use at 4-5X on an APS-sized sensor

--Chris S.

Lou Jost
Posts: 5948
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

I haven't systematically tested the Raynox with very many MFT lenses, so I can't answer your question. On the 60mm, I have found that the Precision Optics closeup lens has better corners than the Raynox.

Adalbert
Posts: 2427
Joined: Mon Nov 30, 2015 1:09 pm

Post by Adalbert »

Hello Bralex,
CA is a property of your optical setup. In case of the microscope lenses mostly caused by the missing eyepiece for its correction.

I remove the CA in the following way:
1.) SET ISO to the lowest value (in my case 100)
2.) TAKE photographs in the RAW format
3.) REMOVE the CA from each picture during the conversion from RAW into the TIFF (in my case by LR)
4.) STACK in two steps (first substacks with PMax) then DMap from the results
5.) REMOVE the remaining CA (by PS ) from the created stack as follows:
. Put image in Photoshop Layer 1
. Copy to Layer 2
. Layer 2, Filter>Gussian blur. Set amount to 20 (+-10)
. Layer 2, set Layer mode to Color

BTW, my LU Plans (5x; 10x; 20x) from NIKON produce a lot of CA :-)
BR, ADi

bralex
Posts: 129
Joined: Mon Jan 22, 2018 11:34 am
Location: New Mexico, USA

Post by bralex »

I usually don't get tags right the first time, but let's go! I really appreciate all the feedback!
Scarodactyl wrote: It does kind of look like something is wrong with the stack, though, as much of thr flake is still out of focus. The center seems to be in worse focus than the second picture shows.
It's not a good stack, I only meant it to demonstrate the CA problems I've been having.
zzffnn wrote: So did you use that "first pic" in that "14 pics" stack? Since snow flake is mostly not in focus in "first pic" and there is lots of CA there, maybe simply don't use it in the stack?

Transparent crystal against black background seems ruthless in revealing CA. Would you be willing to use dark blue or purple background?
I did, though this was just a stack to show the problems, not a serious effort. I was trying to have the background in focus, otherwise it looks like tentacles arising from the mist :)

I hadn't considered changing the background color - my only thought had been to get the blackest black I could find. I'll hit Hobby Lobby and pick up a couple more colors, see how it goes.
rjlittlefield wrote: The problem shown here looks like straightforward longitudinal CA caused by the optics. You might try stopping down, though that would work better if you had brighter light to work with.
*snip*
For this particular subject, simply converting the result to gray-scale would not be crazy.
Next time I get the chance, I will try stopping down. I'd been tweaking the settings a bit, but kind of flailing - that's where my lack of experience shows most. Gray-scale feels like cheating but I've done it a few times. *shifty eyes*

Chris S. wrote: Adding a microscope objective to your rig needn't be difficult or expensive. With an "infinite" objective, you could probably mount it with an adapter on the end of your 40-150mm zoom.
I routinely use a Mitty 10x mounted to the 40-150 with the teleconverter in place. I hadn't considered using it for the snowflakes; I could probably pull the teleconverter and drop the zoom level. I'll give that a try. Reducing the number of optical components, as Lou implied, might help as well.
Adalbert wrote: I remove the CA in the following way:
I'll have to do some experimentation. I haven't sprung for PS/LR because I detest the subscription model, but I do need to step up my game and there are other options.

bralex
Posts: 129
Joined: Mon Jan 22, 2018 11:34 am
Location: New Mexico, USA

Post by bralex »

I should also mention that those pics have an FOV of about 4 mm, cropped down from about 16 mm or so, and are low quality .jpg to meet the board size requirements.

bralex
Posts: 129
Joined: Mon Jan 22, 2018 11:34 am
Location: New Mexico, USA

Post by bralex »

As should have been obvious to me, the Mitty 10x was unsuited to this particular application. Even a 3 mm snowflake completely filled the FOV. The quest continues.

I have a Raynox 250; my next experiment will be to use that with the 40-150 lens and no teleconverter.

Chris S.
Site Admin
Posts: 4042
Joined: Sun Apr 05, 2009 9:55 pm
Location: Ohio, USA

Post by Chris S. »

bralex wrote:. . . the Mitty 10x was unsuited to this particular application. Even a 3 mm snowflake completely filled the FOV. The quest continues.
Quoting Kenneth Libbrecht, who has published several books of snowflake photographs:
  • "I use three microscope objectives - the Mitutoyo M Plan Apo 2X, 5X, and 10X objectives. . . . I use the 5X most of the time, the 2X when the crystals are large, and the 10X only rarely."
When Libbrecht wrote this, his camera had an APS-C sensor. For your 4/3 sensor, a 4x objective would give a similar FOV.

--Chris S.

bralex
Posts: 129
Joined: Mon Jan 22, 2018 11:34 am
Location: New Mexico, USA

Post by bralex »

I'd seen his site before, but not read his techniques. Very interesting. He makes no mention of stacking; given that he's photographing them flat his rig must have adequate depth of field. I'll have to consider - I could probably pick up a 4x or 5x sometime. Thanks!

Lou Jost
Posts: 5948
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

I think that is the most important thing to fix your CA's: replace some of your optics (TC and Raynox) with a good microscope objective. You might also try using a larger-format (FF) zoom lens, which may allow you to push the objective down to lower magnifications. The MFT zooms are excellent but the cheaper ones already suffer in the corners. FF zooms have much larger image circles, so the image is more limited by the size of the objective's image circle than by the zoom's optics.

Post Reply Previous topicNext topic