Photostacking

[chauncey]

It works quite well in macro work and it truly sucks when taking forest images...
how might it work in images of the Grand Canyon?

[arith]

I do not think it will work because change of focus is a change of perspective; I believe they use a rail in macro work. But then I've only stacked for noise reduction.

[DanK]

I think it should work fine. The perspective change is present regardless of whether one uses a rail (I stack a lot without a rail), but stacking software can deal with that. I've read about using it for landscapes but haven't for the reason you mention: it won't work if you have trees or anything else that moves in the images. That shouldn't be a problem with the Grand Canyon. On the other hand, won't you have enough DOF without it?

If you try it, please post your results.

[Manfred M]

I don't think you would need it for shooting the Grand Canyon. The place is quite large. I'm trying to figure out why you would even consider focus stacking in that type of shot. Landscape photographers have been shooting that landmark for decades using normal camera techniques and have been getting sharp images.

The only time I use a rail in landscape work is for panos and even there, it is only necessary if there something in the foreground that is relatively close to the camera, otherwise, just rotating the camera on a tripod (or even hand holding) works well enough.

Shoot at a moderate focal length, at a reasonable aperture (f/8 to f/11) and focus based on the hyperfocal distance and everything should be sharp from things that are quite close to the camera right through to infinity.

[xpatUSA]

I do not think it will work because change of focus is a change of perspective

I didn't know that, is there a reference?

Here it says "Many will say that focal length also determines the perspective of an image, but strictly speaking, perspective only changes with one's location relative to their subject"

https://www.cambridgeincolour.com/tu...era-lenses.htm

So I'm a bit confused

[chauncey]

Shoot at a moderate focal length, at a reasonable aperture (f/8 to f/11) and focus based on the hyperfocal distance and everything should be sharp from things that are quite close to the camera right through to infinity.

That presupposes that you can shoot at those f/stops...in my case noise at anything above ISO 200 is verboten.

[Manfred M]

That presupposes that you can shoot at those f/stops...in my case noise at anything above ISO 200 is verboten.

But you shoot with a tripod, so slow shutter speeds should not be an issue for you.

[arith]

I'm just thinking of something I saw about taking HDR images; where it said a change in aperture was a change in perspective. You could always choose a wide focal length and small aperture, then crop, or maybe even a tilt & shift lens but I don't know how to use the latter.

[mikesan]

I didn't know that, is there a reference?

Here it says "Many will say that focal length also determines the perspective of an image, but strictly speaking, perspective only changes with one's location relative to their subject"

https://www.cambridgeincolour.com/tu...era-lenses.htm

So I'm a bit confused

That has always been my understanding as well. However, I have found that the formal definition of perspective may vary depending on the "perspective" of the individual, and has been the subject of many an argument.

[george013]

That has always been my understanding as well. However, I have found that the formal definition of perspective may vary depending on the "perspective" of the individual, and has been the subject of many an argument.

Perspective is the suggestion of a third dimension in a 2 dimensional plane.
When talking about perspective in the above used context, the line perspective is meant. That is a different magnification of 2 equal subjects situated in a different distance. The angle of view is determining the line perspective.
And the angle of view is determined by sensor size and image distance. Last one means distance sensor to the optical centre. And that changes when focusing.

Some one understands?

George

[Dave Humphries]

In a rather gross simplification, but some thoughts on what various people have read and why different sources may appear to contradict, especially if some of the contexts are overlooked;

a) I would agree that 'perspective' only changes with a change of shooting distance (not focal length per se)
b) Focus stacking (to increase DoF) is usually discussed in the context of macro shooting
c) Most lenses 'breath' - i.e. are subject to a focal length change when focus is changed, this usually isn't obvious in normal photography
d) This phenomena is likely to be more significant at close focus distances; e.g. especially macro shooting distances
e) Also at macro shooting distances; I suspect the 'nodal' lens to subject distance is also likely to change when focus is changed and this may account for any observation of perspective apparently changing
f) You'd think that using 'a rail' (to focus on a macro subject) must change perspective
g) I believe that some macro lenses (old manual ones) focus by effectively working like a rail; i.e. they move all optical elements with respect to the image sensor (or film plane)
h) While other modern macro lenses use 'internal focusing', where only some elements, (deep within the lens) move, so the front element and sensor stay put relative to the subject - I wonder if focusing one of these exhibits any perspective change?


In writing the above, I now realise that the definitions of certain phrases above are not nearly well defined enough for a deep discussion. Also that what applies to one design of lens may not apply to another.

e.g. I stand by "a)", but what do we really understand by "shooting distance"; is that the sensor to subject distance, or the lens nodal point to subject distance? (that will impact 'perspective')
These differentiations are insignificant for landscape shooting, but become very relevant in macro shooting

I think I'm more confused now than when I started trying to help!
Hopefully, subsequent replies may enlighten me

[Shadowman]

I think it should work fine. The perspective change is present regardless of whether one uses a rail (I stack a lot without a rail), but stacking software can deal with that. I've read about using it for landscapes but haven't for the reason you mention: it won't work if you have trees or anything else that moves in the images. That shouldn't be a problem with the Grand Canyon. On the other hand, won't you have enough DOF without it?

If you try it, please post your results.

Except for the clouds but that might make for an interesting capture.

[revi]

In reply to Dave's post:

If you change the apparent focal length of your lens, you might not change the perspective, but you do change the size of your subject on the sensor. So if a lens breathes, you might have trouble aligning your stack if you change the distance at which you focus (without moving the sensor). Of course, changing the position of the whole assembly will also change the image size on the sensor...

In macro it is not at all evident to change the focal point when you are at the limit of your lens' focusing distance. There are cases where you just can't shift the focus to the required point without also adjusting sensor position:
- At a magnification of 1x, a variation in the lens-subject distance requires an equal change in the lens-sensor distance. That's ok, it's what happens when you focus with an 'old' macro lens.
- Slightly under 1x magnification, the variation in lens-sensor distance will be less than the variation in lens-subject distance. That's fine as well, it just means that the focal point will move a lot for a small displacement of the lens.
- However, at magnifications > 1x, you have a problem: if you change focus, your lens-sensor distance will have to change more than your lens-subject distance...
Much easier to use a focussing rail when you get to magnifications of 1x and more.
(This is for a single lens, it might change for a complex objective, and it probably will change for an objective with internal focusing.)

And in landscape photography, using a focusing rail for stacking is obviously out of the question... Manfred's use of a focusing rail in landscape panorama's is not to allow adjustment of the focusing distance, but to make sure the camera rotates around the 'nodal point' (or more pedantically, the entrance pupil) when taking a panorama with both foreground and background objects. This avoids shifting of the foreground relative to the background between individual images. The position of this 'nodal point' differs between lenses, and lies usually within the lens.

[rpcrowe]

Chauncey,

What would you want to gain by photo stacking of an image of the Grand Canyon.

Photo stacking is usually done in macro shooting when the DOF is minuscule.

I shooting an area like the Grand Canyon, you have a very large DOF to work with.

I just grabbed a couple of examples using a 28mm lens on a crop camera and the equivalent 44.7 mm lens on a full frame camera:

28mm on crop camera at f/16: focus on hyperfocal distance of 8.55 feet and the DOF is 4.8 feet to infinity

44.7mm on a full frame camera at f/16: focus on hyperfocal distance of 13.8 feet and your DOF is 6.9 feet to infinity.

These DOF should cover almost any shot of the Canyon.

OTOH: A use for photo stacking in shooting pictures of scenics like the Grand Canyon would be the creation of "superresolution imagery".

Although, I have not done so yet (I soon will!), this site gives a step by step instruction on the creation of the superresolution images.

http://petapixel.com/2015/02/21/a-pr...ith-photoshop/

I was surprised to learn that the author recommends doing this hand-held rather than on a tripod and was not at all surprised at the warning that moving subjects will not be a good choice for this type of shooting...

[chauncey]

Although, I have not done so yet (I soon will!), this site gives a step by step instruction on the creation of the superresolution images.

A quick YouTube search indicates that there is more than one way to skin a cat...
https://www.youtube.com/results?sear...with+Photoshop

That said, rest assured that I will be giving the technique a try, in spite of not really
understanding how it works.

[DanK]

I think this thread is now discussing two different techniques, sensor shift and focus stacking.

Focus stacking with a stationary camera is just a means of combining in-focus areas from photos with different focus points. I have read that it is often used with landscape work, but I have never done so, even though I stack images all the time in my macro work. The only time I can see it helping in landscape work is if you need either a wide open lens for some reason or a very deep depth of field--e.g., very close foreground detail and background detail at infinity. I would be very surprised if it had any benefit for a shot of the Grand Canyon at a reasonable aperture.

[chauncey]

OK, what is sensor shift...how does that work?

[DanK]

Check out the link in Richard's post 14


Sent from my iPhone using Tapatalk

[Dave Humphries]

The only time, apart from macro, I have considered using 'focus stacking' was when twilight shooting part of a church with a full moon behind, the disparate distances meant I couldn't get both in focus - as the relative distances were something like 20m and 380,000,000m! There were just the two subject planes, nothing in between.

That said, the exposure also needed to be changed between shots, since one was sunlit and the other not (it was twilight, as mentioned), so arguably I would have been using an HDR technique.

I don't think I ever bothered to process an image from the two shots, not sure why.

[chauncey]

OK, what is sensor shift...how does that work?

OK, what is sensor shift...how does that work?