Arnstein, welcome aboard!
Sorry to say, there is no automatic method to get rid of those halos in Zerene Stacker or any other focus stacking software that I'm aware of.
This is a question that I field frequently, because I'm the fellow who created ZS and answers all the email support requests.
The underlying difficulty is explained in more detail at
http://www.photomacrography.net/forum/v ... 557#102557 and following. The key issue is addressed in these sentences:
the basic problem is that when foreground goes OOF [out of focus], its blur expands to cover nearby parts of the background. As a result, those portions of the background are never seen both focused and uncontaminated. They're either uncontaminated but out of focus (when focus is on the foreground), or they're focused but contaminated (when focus is on the background).
An even worse version of the same problem occurs in the V-shaped corners where a front part of the subject overlaps a back part of the subject, against contrasting surround. In that case the surround in the middle of the V is never seen clearly but instead is always contaminated by some part of the subject that is OOF. The problem is particularly bad when the surround is black and the subject is light, because it takes only a little OOF light contaminating what logically should be black surround, to turn it dark gray as seen by the camera. But the problem occurs even when the surround is light and the subject is dark, because very near the center of the V, the contamination is always at least 50% and often 75% or higher.
In your case, if you look through all your source images, I expect you'll find that in the regions you've marked, that black surround is never seen as black. It's always some shade of gray, contaminated by some OOF part of the subject.
So, the solution to your problem does not lie in stacking algorithms, which at this time are limited to combining the
images that are provided as input. Instead, what you need is a version of the "painter's algorithm" in which a smart process first figures out what the
3D structure is, then draws what that structure would look like if photographed by an imaginary pinhole camera that is magically invulnerable to diffraction. As a matter of practice, the simplest approach is to use your human understanding to figure out what should be black surround, then use selection and area filling tools in Photoshop or some similar tool to implement that vision.
You're correct that the problem gets worse with higher magnification objectives. That's because higher mag objectives always come with wider entrance cones, which create wider contaminated regions near edges. But the fundamental problem occurs all the time, with any lens. People who shoot white lilies against black backgrounds run into it also, just in smaller parts of the V's.
You asked if it matters which substacker software you use. it does not. At the end of the day all of them just end up doing a Stack Selected on the images in each slab.
BTW, I would have tried a higher contrast threshold setting for the DMap, with the goal of eliminating those separated light blobs in the dark surround. But there is no setting for the contrast threshold that would give a clean rendering in the centers of those V's, because the camera never saw those areas clearly.
--Rik
