Sub-stacks

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

Moderators: rjlittlefield, ChrisR, Chris S., Pau

descall
Posts: 178
Joined: Sat Oct 01, 2011 3:12 pm
Location: United Kingdom

Post by descall »

Fantastic! Thanks very much, Des :D

svalley
Posts: 343
Joined: Sun Dec 03, 2006 7:07 pm
Location: Albany, Oregon

Recent Slabbing Experiences

Post by svalley »

I have been experimenting with slabbing since Rik told me about it last November.

I have been doing it by hand and also a bit differently than the techniques described so far.

My workflow has been:
  • Launch ZS
    Load all source images
    Stack > Align All Frames
    Select images 1-10
    Stack Selected (DMap)
    Select images 11-20
    Stack Selected (DMap)
    repeat previous two steps for all other slabs (DMap only).
    Then, Select images 6-15
    Stack Selected (PMax)
    Select images 16-25
    Stack Selected (PMax)
    repeat previous two steps for all other slabs (this time PMax only).
    Select all output images
    File > Save Output Image(s) to some scratch directory.
    Exit ZS
    Re-launch ZS
    Load the images I just saved
    Align & Stack All (PMax).
So, my overlapping slabs are PMax and and DMap, with the top slab and the bottom slab being DMaps, then those are all aligned and stacked using PMax.

As others have reported for their technique, I am getting a very clean final stack that needs very little retouching. This is fairly tedious for deep stacks . So I am definitely, going to try figuring out the ZereneBatch.xml scripting.

I guess I need to try the overlap method you guys are using and see if it yields better results than my method.

Great job all of you! This is a major step forward in stacking!

Steve
"You can't build a time machine without weird optics"
Steve Valley - Albany, Oregon

ofarcis
Posts: 278
Joined: Sun Jan 27, 2013 10:29 am
Location: Madrid - Spain
Contact:

Post by ofarcis »

Hi to all,

Chris proposed to me that my program can do substacks. For me it is quite easy to implement but I wonder if it really is used or not, since the processing time is extended to have the final stack.

Right now in the two years I've been photographing minerals i don't use this. I do not know if it is useful or not. Any suggestion?

Regards, Oscar.

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

Post by Chris S. »

Oscar,

I don't know which Chris suggested that your program do substacks (also called "slabs"), but I'll assume it was ChrisR, since you're posting this in his thread. There are several of us Chris's around here. So my comments, though from a "Chris," are a second vote, not a repetition of the first vote. (Unless I'm even more forgetful than think I am, and said something I don't remember).

Slabbing is very, very, very important to me. It's "deal-breaker" important, meaning that I would not, and often could not, use stacking software that does not support slabbing (sub-stacking). I use Zerene Stacker for all my stacking, and its robust scripting capability allows me to do slabbing. (Wonderful thing about scripting capability--it lets users do things not directly built-into the software.)

I'm not surprised that you have not yet encountered the need to slab, with micro-minerals. But if you start doing deep stacks (I sometimes do 1500 images or more), or work with complex subjects with overlapping features using objectives with very small depth of field, you'll likely find slabbing very useful, and perhaps mandatory.

Slabbing can also make retouching much easier and faster. Without slabbing, overlapping features that confuse stacking software might be spread across dozens of frames, and the job of retouching the desired element, without slabbing, can be incredibly tedious. In this case, judicious slabbing can place these features--completely rendered--in different slabs, which makes retouching them pretty easy.

Also, slabbing commonly speeds up, rather than slows down, my workflow. I assemble the slabs in PMax, and as you say, this step takes longer than it would without slabbing. But the time spent is unattended computer time--I can be off doing other things. I then assemble the slabs using DMap, and, since my slabs are much fewer in number than my initial images (five to ten percent as many), I can quickly try different estimation radii in Zerene Stacker. The result of enabling quicker trials is that I do more trials, and doing more trials leads me to do better work.

I suspect that not all that many people do a lot of slabbing, but this may be because, so far, it's been a tedious chore. When slabbing becomes quick and convenient, photographers will find it easier to do deep stacks, high-magnification stacks, stacks of complex, overlapping elements, and combinations of the above. Slabbing is like a magic wand--and amazingly, seems so far to be used by only a few of us.

All that said, I must not really understand the purpose of your software. It has seemed to me that your software is intended to provide an additional interface for a stacking controller such as the StackShot. If so, I don't see why slabbing capability belongs in your software. Whether I slab or not, I shoot the stack exactly the same way. My slabbing is performed after image acquisition, in Zerene Stacker. How would I benefit from doing my slabbing in your software? (This is an honestly meant question, not an attack!)

Cheers,

--Chris

ofarcis
Posts: 278
Joined: Sun Jan 27, 2013 10:29 am
Location: Madrid - Spain
Contact:

Post by ofarcis »

Hi Chris S.,

Yes, I meant ChrisR. Sorry, I had not noticed that there are several Chris.

My program does is control a rail focus accurately. It can be any simple or complex rail focus. Currently works with anyone moving through stepper motors unipolar and bipolar. Soon when you receive a test Stack Shot from Paul DeZeeuw (Cognisys) I can implement the interface with the Stack Shot too. Also I have the list of things to do use servomotors.

Moreover, the program is capable of handling SLRs from Canon, Nikon and Olympus. There is a version for each manufacturer.

Therefore, the program moves the focus rail and photography. Once this process, I called to Helicon Focus, CombineZP or Zerene Stack to stack the pictures.

If implemented the slabs, which could do is put an advanced option to stack each slab indicated and next stacking the slabs.

My question you've already answered is whether it is really interesting to implement it or not, for now "two Chris" say yes. A view that says the rest.

Regards, Oscar.

ofarcis
Posts: 278
Joined: Sun Jan 27, 2013 10:29 am
Location: Madrid - Spain
Contact:

Post by ofarcis »

It could even apply one of the filters you have my app as contrast or focus on the slab before the start final stack.

What is the usual number of photos that make up a regular substack?

I have not used it because we only have problems with the "hairs" of minerals, the rest does not usually cause problems. You can see some pictures of me from minerals in http://macrorail.com/GaleriaEng.php (sorry yesterday start the develompen of this Web page an only some functions works, the pictures haven't description jet).

Regards, Oscar.

ofarcis
Posts: 278
Joined: Sun Jan 27, 2013 10:29 am
Location: Madrid - Spain
Contact:

Post by ofarcis »

Hi to all,

I am developing the interface to the advanced mode of stacking (Sub-Stacks) as you can see in the following image:

Image

Basically it indicates the number of images that comprise each sub-stack, the number of photos that overlap between each sub-stack, program and method used for stacking the sub-stacks and the program and the method used to create the final stack.

Currently already works with Helicon Focus, CombineZP and Zerene Stacker. The latter only works in pmap mode, I'm trying to see how to solve one problem of dmap mode with Rik.

My intention is to facility the use of this advanced option without detriment to its power.

If anyone has any suggestions, this is a good time.

Regards, Oscar.

ofarcis
Posts: 278
Joined: Sun Jan 27, 2013 10:29 am
Location: Madrid - Spain
Contact:

Post by ofarcis »

Hi to all,

I give an example of Cornwallita mineral (used this always for optics and functions test), where I stacked 140 pictures in a range of 5 microns to move 0.7 mm. Then I proceeded through this advanced mode and as you can see in the picture to make sub-stacks of 18 photos, 6 photos overlapping between stacked, with a total of 12 sub-stacks by Helicon Focus Method A, to make the final stack using Helicon Focus to method B.

Image

The full stack of 140 photos from the Helicon Focus takes 16:56 minutes. Keep in mind that the Helicon images displayed on the screen one by one.

The stacked by sub-stack system takes 7:41 minutes. It takes less because it has no picture on the screen except the progress bars.

Why do takes much less in the sub-stack? because apart from not having to present any image (Helicon say that without it earns a 30% speed) by having to treat small groups of pictures, it takes less time to stack, all images are loaded into memory while when many this is impossible.

I put pictures of the results for yourselves whether it to value or not worth using this process. I had to reduce by 1/4 the size of the resulting images.

Keep in mind this is a quick test without bothering too light, stacking, etc.., Since the objective is to make a complete test not a work of art.

Helicon Focus stack. Normal stacking.
Image

Sub-stacking method. My app.
Image

I painted red arrows here so you can see the differences quickly: http://www.tuplaneta.es/foros/indexSinV ... 8783#58783

Regards, Oscar.

Rylee Isitt
Posts: 476
Joined: Fri Apr 13, 2012 3:54 pm
Location: Canada
Contact:

Post by Rylee Isitt »

I ended up trying "slabbing" this weekend, using Zerene Stacker.

I got good results after a bit of practice. I have a subject with lots of spines and low-contrast areas. Although the initial DMAP stack was still full of little problem areas, touching it up was so much easier with only 15 source images to worry about. However, PMAX struggled, too, by producing results with background details that bled through foreground details. I'm still working on the image but I should be posting the result soon. It's pretty cool :)

I'm definitely sold on the technique, and in general the idea of hybridizing DMAP and PMAX somehow to dampen their weaknesses.

So, Rik, what is the likelihood that we can get a new mode call AIDMAP that uses pattern matching to recognize structures such as spines, convex and concave surfaces, and so on... and uses these to refine the standard depth map? It would be revolutionary if even something as simple as spines could be recognized for what they are, and with some pre-conceived idea of their shape, given a better representation in the depth map...

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Rylee Isitt wrote:So, Rik, what is the likelihood that we can get a new mode call AIDMAP that uses pattern matching to recognize structures...
Beats me. That's been on the list for quite a long time. Unfortunately it falls in the category labeled "multiple breakthroughs required", and I swore off scheduling {-o< such things several years ago.

--Rik

Rylee Isitt
Posts: 476
Joined: Fri Apr 13, 2012 3:54 pm
Location: Canada
Contact:

Post by Rylee Isitt »

Rik,

Haha, yeah, it seems a fairly ambitious goal. Kind of up there with me wanting to create AI that does all my boring jobs for me ;)

crayfish74
Posts: 148
Joined: Tue Sep 04, 2012 6:11 pm

slabs

Post by crayfish74 »

Dear Rik.-

I have tried to work using slabs, but i don´t know if Im doing something wrong, but i don´t see any big difference doing slabs and normal stacking.

I have done this workflow.

A workflow that is tedious but simple to understand goes like this:
Launch ZS
Load all source images
Stack > Align All Frames
Select images 1-20
Stack Selected (PMax)
Select images 11-30
Stack Selected (PMax)
repeat previous two steps for all other slabs.
Select all output images
File > Save Output Image(s) to some scratch directory.
Exit ZS
Re-launch ZS
Load the images you just saved
Uncheck everything at Options > Preferences > Alignment
Stack > Align & Stack All (DMap).

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: slabs

Post by rjlittlefield »

crayfish74 wrote:I have tried to work using slabs, but i don´t know if Im doing something wrong, but i don´t see any big difference doing slabs and normal stacking.
The big difference is in retouching, not in the computed outputs.

--Rik

pierre
Posts: 288
Joined: Mon Jan 04, 2010 12:37 pm
Location: France, Var, Toulon

Sub-stacks

Post by pierre »

For information, the last Version of Helicon Focus 6.2.2 now includes the capability of sub-stacking.

Image

Image
Regards

Pierre

ofarcis
Posts: 278
Joined: Sun Jan 27, 2013 10:29 am
Location: Madrid - Spain
Contact:

Post by ofarcis »

Good news and not good news. Now can use Sub-stacks but stack ever with the same type of stack method. See few post ago, my proposal that works very well use differ method to stack each sub-stack and use other method to stack the sub-stack results.

This is very convenient to obtain the final stack with more details or quality, etc.

Regards, Oscar.

Post Reply Previous topicNext topic