Focus Stacking Artifacts with EM1 II

[Harvey Melvin Richards]

I tried some in camera focus stacking with the EM1 II and Oly 60 f/2.8. The camera was hanging upside down from a tripod, remote shutter and the photo was of the bottom of a mushroom. The final .jpg from the camera has little rows of almost pixel sized white dots. Not all of the final images that I got have these, but several do. The attached image is an extreme blow up and 5 of the 6 dot lines are oriented in the same direction, but the cicled blue image is more horizontal than the other circled red dots. Any idea what is going on here?

Subscribe to see EXIF info for this image (if available)

 

[Harvey Melvin Richards]

A full sized color photo. Could these be a hot pixels? The lines tend to go towards or away from the center. The number of dots in the line is the same as or slightly less than the number of stacked images. The dot size is about 6 pixels square.

Subscribe to see EXIF info for this image (if available)

 

[Carbonman]

What happens when you shoot an even dark surface? Do you see any evidence of stuck pixels? The only time I noticed a stuck pixel was against a black background, and it showed up as a red spot, sort of like you have with a few of the rows of dots. It seems to be a lot, maybe beyond what can be fixed by pixel mapping.

 

[Acraftman]

This is probably completely the wrong suggestion because it does seem a pixel issue but have you tried the focus stacking software used in this video? He mentions it within the first couple of minutes.

 

[gtm]

I have to think this is some strange artifact of the stacking process. Something really odd would have had to have happened to cause that number of hot pixels. From what Olympus told me when I had one stuck pixel (E-M1.1), they are always there once they "go bad" and it is usually one pixel at a time. Olympus JPEG software will try and even out the surrounding pixels to blend the hot pixel which could explain why you see it effecting 5 pixels (in reality this only seems to accentuate the bad pixel). Raw files should show only the single hot pixel. As Carbonman suggested take a black picture (lens cap on works). Hot pixels should be obvious. Pixel mapping will shut down the bad pixels, but from what you show that would be a significant number of pixels to lose and too many to lose all at once absent something happening to the camera/sensor.

 

[djtaylor7]

Do you know how many frames were in the focus stack? Wonder if it is the same number of dots?

 

[Harvey Melvin Richards]

Thanks for all the suggestions, I am away from the computer and camera so I will try so try to get to it tomorrow. I used 15 stacked images for this photo, and most of the dot paths have 12 to 15 dots. Paths that are close to each other have a similar path also. I also have other stacks done at the same time and I have’t found any problems with them.

 

[Bushboy]

It has got something to do with the stacking, I think.
I sometimes get red lines...

Subscribe to see EXIF info for this image (if available)

Check my weird anomaly’s. Above the mushroom, slightly left of centre, a red line.
I get these sometimes.

 

[Richard_M]

Have you checked the individual images to see if they appear on them also?

I've only done one in camera stack, and while I didn't noticed any anomalies like on your image, I was not overly impressed with the result.

 

[Bushboy]

Good idea Richard.
Next time I’ve got my big computer going. I will check the individual pics.

 

[Bushboy]

Actually, it could just be a bit of something stuck to my sensor...

Subscribe to see EXIF info for this image (if available)

Same place look. Up top in the middle. Bit of red fluff off my jersey!
Nah, I don’t know.
Sorry about high jack your thread Harvey!

 

[Harvey Melvin Richards]

Actually, it could just be a bit of something stuck to my sensor...
View attachment 761676
Same place look. Up top in the middle. Bit of red fluff off my jersey!
Nah, I don’t know.
Sorry about high jack your thread Harvey!

I’m just glad that it’s not exclusive to my camera. Mine seems a little random, I did a dozen stacks with very minimal camera relocating and the lines were only there on 2 or 3 of the stacks.

 

[rloewy]

Affinity's stacking software is real easy to use - might be worth to try it again with focus-bracketing and see if the problem is fixed when using an external software to stack.

 

[Growltiger]

Helicon Focus and Zerene Stacker are the most highly regarded stacking programs.

The answer will be clearer once the original files are examined. I suspect some are stuck pixels and some are just weird. Well worth doing the pixel mapping.

 

[Harvey Melvin Richards]

I took this stacked photo a day earlier, also posted in the Spiders thread. I don’t see any artifacts on this one. Whatever the problem is, it does seem to be random.

Subscribe to see EXIF info for this image (if available)

EM121186-1 by Harvey Richards, on Flickr

 

[rloewy]

I took this stacked photo a day earlier, also posted in the Spiders thread. I don’t see any artifacts on this one. Whatever the problem is, it does seem to be random.

View attachment 761736 EM121186-1 by Harvey Richards, on Flickr

This is more likely than not the stacking algorithm not handling certain issues well - and well worth considering an external stacking program.

 

[Harvey Melvin Richards]

This is more likely than not the stacking algorithm not handling certain issues well - and well worth considering an external stacking program.

When I get a chance I will try the stack in Photoshop. The nice thing about in camera is that the results are almost instant.

 

[Harvey Melvin Richards]

Here is the photo stacked in PS. Not as many lines of dots, but some are still there. The topo effect is really pronounced along the top of the shroom.

 

[Growltiger]

Here is the photo stacked in PS. Not as many lines of dots, but some are still there. The topo effect is really pronounced along the top of the shroom.

Have you checked to see if those dots appear in any or all of the original photos?

 

[Harvey Melvin Richards]

Have you checked to see if those dots appear in any or all of the original photos?

There are single dots in a few of the photos that I have checked. I'm sure they are just being repeated in the stacks, but they end up being lines because of the different focal planes.