Miniature flowers, focus-stacked

[Kodiak]

Is this not the case also Kodiak with a macro lens where focal length changes whilst focusing, or at least it does on mine?

Not exactly Grahame, since the film plane is always at the same distance
from the subject. The rail will not refocus but move the combo. The only
situation I would recommend using the rail is when one does not use an
AF lens. In such case, I would opt for DanK's manual solution since it is
a lesser evil.

Some will argue that since the focus slice is further… I understand but
the resulting stack and the working time in PP are not negligible!

[Stagecoach]

Not exactly Grahame, since the film plane is always at the same distance
from the subject. The rail will not refocus but move the combo. The only
situation I would recommend using the rail is when one does not use an
AF lens. In such case, I would opt for DanK's manual solution since it is
a lesser evil.

Some will argue that since the focus slice is further… I understand but
the resulting stack and the working time in PP are not negligible!

I have not yet had anywhere near enough experience using the two methods Kodiak to have a clue about the relative processing times using the two methods.

But I have just thought of a project for tonight to refresh my use of the auto stacking focus control

[FlyingSquirrel]

Thanks for the comments everyone. I appreciate your feedback, suggestions, etc.

Izzie, I grow all of my miniature plants in the one paludarium that I linked to in the first post.

Dan, I'd love to shoot snowies with you

I'd like to comment on the focus stacking methods being discussed. There will always be those that will disagree with each other, and I say the following with that in mind: I don't believe that changing focus vs moving the focal plane has as big of a difference on the quality or difficulty of the results as is being touted by some. Frankly, the stacking software algorithms and technology are so good these days, there is not a noticeable difference in the results, that I've noticed personally. Helicon Focus Pro is very good, and it seems to handle any capture method quite well, assuming that you make good captures, and you adjust the software parameters to find the best settings.

Additionally, on the MF vs macro rail topic, my experience is that I simply cannot accurately, evenly, and consistently turn the MF ring, especially when dealing with small subjects (1/2" flowers). However, I CAN use a macro rail to make minute, relatively even, consistent adjustments to the focal plane. Thus, for me, the macro rail trumps the MF method. Let us not also forget that, the increments at which you need to focus, will vary depending on what aperture you are shooting at (in general, a wider aperture is preferable, but that leads to less DOF per image, thus narrower focus increments being required)

Finally, regarding the AF control method via computer, I would have to agree that it is the most accurate, easiest method and will produce more consistent, higher quality results. The only disadvantage is that you have to have a computer with you to shoot the stacks. I recently got a CamRanger, which I can control with my KindleFire HD tablet. I've not yet done a stack with it, but I think it will suffice for field work, albeit being more weight and bulk to pack and fuss with.

Anyway, those are just my opinions based on my own experience.

[DanK]

Is this not the case also Kodiak with a macro lens where focal length changes whilst focusing, or at least it does on mine.

Yes.

I think this thread may be confusing for people who have not done this. So, let me try my understanding:

-- there is only one parameter that always must be changed between images to be stacked: the point of focus.
--this can be changed in either of two ways: moving the camera, or changing the focus of the lens. Either can be done by hand or by computer, but the choice between manual and automatic only matters as a matter of accuracy and convenience.
--As Matt wrote, either method will change the apparent magnification quite substantially when one is close to 1:1, as in the OP's images. This effect is large with my lenses if I am near minimum focusing distance.
--This change in perspective does not cause problems for modern stacking software in most cases. It causes problems (generally, in the form of halos or blurry areas) where there is a substantial distance between an edge and the surface behind it. This problem is basically one of geometry. Kodiak's car is an example of an image that does not have this sort of issue. When you have that issue, either method of changing focus will cause difficulties. This requires extra work in PP, and it isn't always possible to fix it entirely.
--When you don't have this problem, post-processing can be quite simple, and it would be identical for the two methods (moving the camera or changing focus). Here is the simplest workflow I use for flower macros, when there are no complications, such as uneven lighting or halos:

1. Import to Lightroom
2. Set WB on the images selected for stacking.
3. Export the selected set as 16-bit prophoto TIFFs to Zerene. I do this with a plug-in, so I just tell LR to export to Zerene, set the format to TIFF, and let it go. It will then export the images, start Zerene, and load the TIFFs.
4. In Zerene, select "Stack all-DMap" or "Stack all--PMax'. There are important differences between these methods, but I won't go into that here.
5. Save the composite.
6. Go back to LR and edit the composite, sometimes as little as tonality adjustments and sharpening.
7. Export.

Of course, there are times when I have to do a lot more. The point is that the stacking process itself, even with the difference in effective magnification between slices, does not add much complication to postprocessing.

Matt, it sounds like you and I are pretty much on the same page. Re the use of a rail: I was not arguing against it. If you find it easier, by all means, that's the way to go. My point was just that many people will not find one necessary at this level of magnification.

I do have two questions for you:

Re snowies: was that a response to my comment about the small amount of reflection in the first picture? If so, for what it is worth, this is one reason I use ambient light. It allows me to bounce a light source against a reflective umbrella, giving me a much larger light source. That helps when a flower produces harsh reflections, although it is not always ideal. Of course, you could do that with a flash too; you just wouldn't know if you had it as you want it until you took some pictures. I'm also curious why you say that a wider aperture is better. That may be why you prefer a rail. I generally do flower stacks with the lens set around f/8. That leaves some room for error, and it is about the lens's sweet spot anyway. Am I losing something by not opening wider?

Dan

[Anne22]

Beautiful shots, really love #3

[NorthernFocus]

...Re snowies: was that a response to my comment about the small amount of reflection...

Wrong Dan, Dan. I made a comment in my post about shooting snowy owls. They tend to show up during winter around the PNW where Matt resides. I was wrangling for a invite to come down and shoot with him

[DanK]

Wrong Dan, Dan. I made a comment in my post about shooting snowy owls. They tend to show up during winter around the PNW where Matt resides. I was wrangling for a invite to come down and shoot with him

Ah, too many Dans. On that front, this particular Dan is incompetent. I have only managed a handful of decent bird shots ever.

[Kodiak]

•

That's why I preferred to stay Kodiak!

[NorthernFocus]

Ah, too many Dans. On that front, this particular Dan is incompetent....

We each have our fields. I'm clueless regarding the focus stacking. I can barely stand using a tripod much less rails, computers, etc.

[ajohnw]

The person who maintains Zerene pops up on a forum I am a member of at times. He's also often around on photomicrography.net. I've collected a few of his answers to questions but they are a few years old now. A couple of them might help.

Firstly stack good clean images. Take care of noise, colour, balance and any spotting that might be needed and don't over sharpened.

Not so easy on a camera but very easy on a microscope. Too few images will be a disaster and within reason too many helps. He suggests 70 to 80% of the dof. In camera terms that means 70 to 80% of the depth where the image looks sharp at it's full resolution. On a microscope the figures can be obtained from tables. This relates to another comment that he made - help the program to stack. The out of focus fuzz wants to build up as rapidly as possible which suggests using the fastest aperture available where the lens gives adequate resolution. The shortest possible distance should also help.

He suggests this for D-Map

The default parameters for DMap are chosen to work well with images that are
sharp when viewed at 100% = actual pixels. Images that do not look sharp at
100% generally require larger settings for the two radius parameters. A good
strategy for setting those radii is to find a frame that contains the level of
detail you care about, reduce the display Scale until that frame appears sharp
on screen, then make the radii larger than default by the same ratio. For
example if you find that the image looks sharp when displayed at 50%, then make
the radii 2X bigger than default. If it has to be reduced to 25% to look sharp,
then make them 4X bigger. This same strategy works well in other programs that
use radii also.

And on the 2 methods but bear in mind these comments are now 4 years old and the program will have moved on.

In Zerene Stacker, the most popular stacking method is PMax. It has no
parameters at all, and it's also the best method around for finding and
preserving even low contrast or fuzzy detail. However, PMax also has the
downsides that it tends to increase noise, increase contrast, and may alter
colors. DMap is just the opposite -- it faithfully preserves contrast, colors,
and noise, but may miss low contrast or fuzzy detail and requires tuning to
give the best result.

Another

If stacking mush is what's being described, then the best fix I know is to use
Zerene's PMax method or CombineZP's Pyramoid Maximum Contrast. (The complex
wavelet method would probably give good results too, but I haven't specifically
tested it.) Most of the other methods are vulnerable to stacking mush,
including Zerene's DMap, both methods A and B in Helicon Focus, and the depth
map methods like Do Stack in CombineZP. (The fact that ********** mentions the
interactive threshold setting tells me that he's using DMap, which is vulnerable
to mush. PMax has no threshold setting.)

All a bit microscope orientated but similar thoughts may help with conventional camera lenses.

John
-

[DanK]

The out of focus fuzz wants to build up as rapidly as possible which suggests using the fastest aperture available where the lens gives adequate resolution.

I don't understand this. The faster the aperture, the more out of focus there is. I'll have to go back to Rik's online documentation. John, did Rik actually suggest wider apertures himself?

Re this:

If stacking mush is what's being described, then the best fix I know is to use
Zerene's PMax method or CombineZP's Pyramoid Maximum Contrast. (The complex
wavelet method would probably give good results too, but I haven't specifically
tested it.) Most of the other methods are vulnerable to stacking mush,
including Zerene's DMap, both methods A and B in Helicon Focus, and the depth
map methods like Do Stack in CombineZP.

One of the nice things about Zerene is that you can paint from any source image onto the composite. A very handy way to do that, when you have 'mush' (what I am calling halos) is to tell Zerene to use both PMax and DMap. PMax will almost always have much less of it, but DMap is in my opinion superior for flowers in other respects, as it does a better job of preserving colors. So, you stack both ways and then, for the mushy areas, paint from the PMax composite to the DMap composite. It is much faster than finding the appropriate image from the stack and painting from there, although I do that as well.

[Wavelength]

Lovely images

[ajohnw]

No I inferred that from his comments in a little tutorial he did on some work he was doing himself as an example but as I said it's microscope related where the fuzz really does build up at a very fast rate and 100 shot stacks aren't unusual. Focus steps can be in um. I'd be inclined to encourage a rapid fuzz build up if that is feasible. These packages do work by effectively selecting sharpness so a more rapid build up should help.

Stacking mush - some one was complaining about quote "> One effect I notice in the stacks is a muddiness of the image and I would like to minimize that effect.". I copy pasted his response. He asked for a sample image but none was forthcoming.

His tutorial on macro with a camera makes no mention of ideal aperture. I suspect he feels people should study this page and know that lenses generally give their best resolution at certain apertures and to use that when maximum detail is needed, another aspect. He does mention 70% of dof here.

http://zerenesystems.com/cms/stacker.../macromicrodof

The Canon 100mm is best at F4 purely on central resolution on a crop body, The Sigma as well. The resolution evens up more across the frame as they are stopped down more so F5.6 and F8 may be more desirable. Past that resolution drops. This may or may not matter but where this software is mostly used it can be crucial as the F ratio's are a lot faster and fixed except in the case of lenses intended for some form of micrographic use.

I'd try it if I were you. There is even a possibility that decent clean stacks could be produced with the lens wide open in many cases. It all depends on the detail that needs to be recorded. Gaining extra detail via stacked shots taken at higher F ratios is what this software is about really but that aspect isn't as marked when using conventional camera macro gear.

Removing images from a stack when there are too many has been mentioned as a method of improving a result. On the other hand talking to some one who does a lot of this professionally things may have "improved" in some packages.The more the better. I hoped to quiz him more this year but didn't go to the places where he would be around. Maybe next year.

John
-

[DanK]

John,

interesting. A couple of thoughts.

First, we are probably talking about two different kinds of mush. I was talking about halos that appear when edges are far from the surface behind them. There is also a more general mushiness that can can occur in other cases, when the program can't differentiate between meaningful detail and noise. You can see this in the section under DMap here: http://zerenesystems.com/cms/stacker/docs/howtouseit. Rik notes that this is particularly a problem when using a microscope, but I have occasionally had it in my photos. I generally use DMap for flowers, for reason Rik describes on that page, but I use PMax when halos are a serious problem, and I sometimes combine the two.

Re the resolution of the 100L: I just looked some up some tests on my lunch break, and I found varying results, even for people using Imatest. Photozone gives results similar to what you describe for crops. Others show more of a fall-off as you open up wider than 5.6. All of them seem to agree that 5.6 is about as good as one will get. However, I would wager that anything from f/4.0 to f/11 would give results similar enough that most of us mere mortals would be unable to tell the difference without pixel peeping, and perhaps even with it.

(New England) Dan

[ajohnw]

He has mentioned using all methods and then combining the best results from each into a single image recently on photomicrography.net to some one that was stacking flower shots. They were concerned about artefacts on the end of petals which he feel is a current state of the art problem.

I do feel that shallower depth of field will help the software pick out the right bits more correctly so would feel that I would have to try it.

On m 4/3 I am more or less bound to have to worry about diffraction effects. On the other hand I have already taken shots that I feel loose detail at web sizes even at F11 where in theory diffraction will be having some effect. I reckon it all comes down to the debayering software producing "better" results than might be expected.

What the tutorial I mentioned was effectively saying was leave some "sharp" overlap in each image which is where I suspect his 70% of dof comes from. Dof in camera terms though depends on image size - shots can be made "sharper" by reducing the size.

There are some free macro photography books about on the web but the author contradicts himself. One point around F6 results in best resolution and in another he often uses F11. It seems that the slower one might have been mentioned if work is being done out of doors, even F16, due to possible movement. My approach would be the same as when using a microscope - minimum mag needed to show detail which can be judged through the viewfinder with a bit of practice and an F number that will not interfere with pixel resolution if the images are to be stacked. But I will still wonder about getting the smallest possible dof in the hope it helps the stacking - it should at the expense of more shots.

John
-

[FlyingSquirrel]

@Dan K, I have to applaud you for your extremely well written explanation of the basics of focus stacking and the differences in capture methods. Very, very nicely done!!

Re snowies: was that a response to my comment about the small amount of reflection in the first picture? If so, for what it is worth, this is one reason I use ambient light. It allows me to bounce a light source against a reflective umbrella, giving me a much larger light source. That helps when a flower produces harsh reflections, although it is not always ideal. Of course, you could do that with a flash too; you just wouldn't know if you had it as you want it until you took some pictures.

As you saw, the snowies comment was re: NorthernFocus. I will reply to your comments about flash vs ambient. I used to love ambient light for everything. But I used to be afraid of flash. Very afraid. Wouldn't pick one up if you paid me to. But recently I decided that I needed to learn how to use a flash. Once I really got into the concepts, and started experimenting, I have grown to really enjoy using flash, when it is reasonable to do so. I still love ambient light, as well. When shooting macro, I always have my 5-in-1 popup reflector and diffuser set. With reflectors, diffusers, and flash, you can pretty much pull off whatever kind of macro lighting you can dream up.

The three biggest benefits of flash that I see are:
1. You can freeze motion easier than with ambient (irrelevant for stacking, but relevant for insects and single capture shots)
2. You can shoot at the native base ISO (100 for the 7D), thus resulting in much cleaner images. Less of an issue if you have a newer FF camera, but I don't
3. You can take pics at any time of day or night, indoors or out, regardless of ambient conditions.

In learning flash, I've been experimenting a lot with diffusers, and various other things. The images I shared in this thread were taken a few months ago, when I was just getting started. Recently, I came across a very promising diffuser material at a plastics shop (basically just a white plastic in a scrap bin). The results are incredible. I am still a relative novice compared to yourself or many others on here, in terms of my overall experience with macro, and flash. But I will be working on mastering it as much as I can. I look forward to sharing more macro pics in the future.

I'm also curious why you say that a wider aperture is better. That may be why you prefer a rail. I generally do flower stacks with the lens set around f/8. That leaves some room for error, and it is about the lens's sweet spot anyway. Am I losing something by not opening wider?
Dan

When first learning about stacking, I read, in a few sources, that the shallower DOF from a wider aperture yields better results when using software to combine captures. This is because the software can detect the more abrupt transition between the apparent focus / oof areas. A smaller aperture creates a broader and less distinct DOF transition, thus causing possible errors in the software's processing. Additionally, wider aperture results in less diffraction, though at the apertures you are talking about (f8) diffraction is not an issue, and you are correct that it makes sense if the lens is sharper in that range. I shoot most of my stuff at 5.6

I want to thank everyone for your comments, re my images, as well as adding value to the thread.

[DanK]

Matt,

You've given me some things to think about. I'm going to explore the issue of aperture. Re lighting: like you, I use diffused flash for things that move. When I said "ambient light" for flowers, I didn't mean natural light. I use halogens with diffusers and umbrellas, with long exposures so that I can shot at base ISO. This has one big disadvantage-- the risk that the flower will move from vibrations or even wind. The big advantage is that you can move the lights around and get a fairly good idea of how the image will look. However, your shots have me intrigued. I may play around with reflected flash as an alternative.

[rudi]

I am a bit in the same position as Matt, what ambient light vs flash is concerned. I never used flash until some ligthening
threads at CiC convinced me to try it out. Still a lot to learn but my keeper rate is going up (from 1 to 2 per session ...)

About stacking I am with DanK and generally do flower stacks at f8.
I am no expert at all, but I have done some experiments f4 vs f8. The stacks at f8 came out better than f4.
Keep in mind that: Pictures were taken outside (on tripod luckily), stacked in CS5, that I am just trying out Zerene/HeliconFocus, and will restack them later (as the days become shorter and shorter...).

I have done some stacks indoors lately and came to the conclusion that the sligthest breese can cause a movement off the flower resulting in an image shift at the pixel level, that can cause the 'mush'....

I was planning more experiments during wintertime.
Looking forward to your results.

[FlyingSquirrel]

Matt,

You've given me some things to think about. I'm going to explore the issue of aperture. Re lighting: like you, I use diffused flash for things that move. When I said "ambient light" for flowers, I didn't mean natural light. I use halogens with diffusers and umbrellas, with long exposures so that I can shot at base ISO. This has one big disadvantage-- the risk that the flower will move from vibrations or even wind. The big advantage is that you can move the lights around and get a fairly good idea of how the image will look. However, your shots have me intrigued. I may play around with reflected flash as an alternative.

Ah, I see. What you describe as ambient light is what I was taught in school is called continuous lighting. I can only assume that you do all of this in studio, as it seems it would be difficult or impossible to carry and use that setup in the field (probably would require some hefty batteries!)

Don't forget that you can use a diffuser AND a reflector with flash.

I look forward to seeing what kind of results you get after you do some experimenting with things.