Help with transparency artifact

[billjanes1]

I am just getting into extreme macro and am not sure if I should post this here or in the beginner's forum, but will try to post here.

This is my first try. I captured a house fly and took 96 shots with the Mitutoyo 5x NA 0.14 using the Nikon d800e with a 200 mm f/4 AI Nikkor as a tube lens. Focus step was 25 microns with a Stackshot.

I stacked with both pmax and dmap. The DMap stack was unusable becuase of extreme transparency of the bristles on the fly, so I used PMax.

Here is a downsized view of the uncropped image. Since this is for proof of concept only, I made no attempt to clean the specimen. Slight vignetting is present.



On inspecting the full sized image at 1:1 many of the bristles show transparency artifact as shown here.



Stack selected does not appear to be usable here. If I downsize the image, the artifacts are less apparent. I was wondering if it might help to stack with an enlarger lens (I have the 50mm f/2.8 N Nikkor). I could use a smaller aperture to reduce the transparency artifact with some loss of resolution (the f/number of the Mitty is 3.5)

Any suggestions would be appreciated.

Bill Janes

[JH]

Flies are sometimes annoyingly difficult to stack. There could be hours of retouching. I usually end up with one or two sub-stacks for every large bristle.

In your second picture, there is a lot of DMap artefacts - a frequent problem with bristles - but also some areas that looks like they are out of focus or blurred. Therefore, I must ask if you had any problems with vibrations and/or if some of the steps became larger than intended.

Best regards Jörgen

[billjanes1]

Flies are sometimes annoyingly difficult to stack. There could be hours of retouching. I usually end up with one or two sub-stacks for every large bristle.

In your second picture, there is a lot of DMap artefacts - a frequent problem with bristles - but also some areas that looks like they are out of focus or blurred. Therefore, I must ask if you had any problems with vibrations and/or if some of the steps became larger than intended.

Best regards Jörgen

Jörgen,

Thanks for the input. The second picture was a 1:1 view of the first, which was stacked with PMax, so I don't understand why it shows DMap artifacts. Illumination was with diffused Nikon SB910 at 1/32 power, so I don't think vibrations were a factor. The focus step of 25 microns may have been too large since the DoF of the lens is 28 microns and more overlap may have been desirable, say 28*0.7 or 20 microns. Also the stack does not cover the entire depth of field of the subject.

The fly may have been too difficult of a subject for a beginner and I will try to build up experience with a simpler subject, but further suggestions would be most welcome.

Bill

[JH]

No problems - probably just an upload misstake. PMax does not give unsharp halos around bristles and handles cromatic aberattion better (less green and red bristles)

Try to do some retouching with sub stacks.

Regards Jörgen

[rjlittlefield]

In your second picture, there is a lot of DMap artefacts

The second picture was a 1:1 view of the first, which was stacked with PMax, so I don't understand why it shows DMap artifacts.

I agree with Jörgen's analysis. There's no way the posted 1:1 view was constructed by PMax. PMax does not give light blotches in the middle of dark bristles, or loss-of-detail halos around them. For that matter your original overview shot does not appear to be PMax either. The transparency artifacts produced by PMax are seamless interleaving of two focus planes, like shown at http://www.zerenesystems.com/cms/stacke ... foreground.

Unfortunately I also agree with Jörgen's idea that there is no solution to the problem of transparent bristles except for retouching. The software is just not smart enough to figure out that those low contrast dark areas are actually opaque things that should completely hide detail behind them, no matter how clearly that detail is seen by the lens when focused farther back.

--Rik

[billjanes1]

In your second picture, there is a lot of DMap artefacts

The second picture was a 1:1 view of the first, which was stacked with PMax, so I don't understand why it shows DMap artifacts.

I agree with Jörgen's analysis. There's no way the posted 1:1 view was constructed by PMax. PMax does not give light blotches in the middle of dark bristles, or loss-of-detail halos around them. For that matter your original overview shot does not appear to be PMax either. The transparency artifacts produced by PMax are seamless interleaving of two focus planes, like shown at http://www.zerenesystems.com/cms/stacke ... foreground.

Unfortunately I also agree with Jörgen's idea that there is no solution to the problem of transparent bristles except for retouching. The software is just not smart enough to figure out that those low contrast dark areas are actually opaque things that should completely hide detail behind them, no matter how clearly that detail is seen by the lens when focused farther back.

--Rik

Rik,

You and Jörgen are correct. I went back to the project, and the both images in my original post were stacked with DMap. The PMax stack is much better. The learning curve is rather steep, but excellent advice on this forum is very helpful.

Bill

[JH]

Two tips

Sensor dust can disturb DMap. If this is a problem retouch the source images before stacking.

PMax can build up noice. If this is a problem remove noice and avoid sharpening of the source images before stacking.

By the way I liked your picture.

Best Regards Jörgen