New Article: Focus Stacking & Depth of Field

[McQ]

Just finished a new techniques article:

PART 1: Focus Stacking & Depth of Field

Focus stacking is a very powerful technique for extending depth of field beyond your camera's normal technical limits. It works by combining several photos taken at different focusing distances so that only the sharpest parts of each contribute to the final image. This is very useful for macro & close-up photography, but can also be used with landscape and low-light photos.

The article gives an overview of why it's needed, then walks through how it works conceptually. If you have Photoshop CS, then it also provides directions for how you can replicate the process. There's also a second half of this article here:

PART 2: F-Stop Stacking: Depth of Field & Corner Sharpness

As usual, please let me know if you feel anything is unclear, if you notice any typos or just want to add something from your own experience.

Thanks!

[carregwen]

Excellent tutorial, Sean (as usual). I use focus stacking a lot for floral images. Once you get used to it and develop a good workflow method, it's incredibly easy. I use Helicon stacking software, but you can use Photoshop almost as easily. This is one I did last week for the PAD. I deliberately left the right side (end of stalk) OOF to add some realism.

[Clactonian]

Thanks Sean, very clear and worthy of some experimentation.

[Peter Ryan]

Thanks again Sean for this addition. This is something I want to try when I get more time next year. I have it clearly in my mind to do so.

[Klickit]

Thankyou, Sean. This is very accessible and I was able to easily follow your explanation.

[jahcyber]

Cool idea, Sean. Being new to all these, I am in awe at what I'm learning here everyday. I have a fairly good Photoshop know-how, so this should be interesting to try out.

carregwen, that Toad Lily pic is breath-taking. Nice work.

[Donald]

carregwen, that Toad Lily pic is breath-taking. Nice work.

Did anyone else notice that if you saw this in the Latest Posts or Latest Threads window (where you just get the opening few words), it said, "carregwen, that toad"?

Well!

[McQ]

FYI, I've updated the original post to also made the second half of this article available:
PART 2: F-Stop Stacking: Depth of Field & Corner Sharpness

Did anyone else notice that if you saw this in the Latest Posts or Latest Threads window (where you just get the opening few words), it said, "carregwen, that toad"?

That's pretty funny -- no, I hadn't noticed that.

[rpcrowe]

This is really an excellent tutorial. Thanks for doing all the work to share your knowledge with us.

I need to clear up an area which somewhat confuses me.

When you use the term "focus distance", do you mean "the distance focused on" or the "lens to camera distance".

"Normal" images are usually focused (either manually or by suto-focus) by the camera/lens selecting a point upon which to focus. Changing the focus in a "normal" image is done by changing the focus point of the lens. Moving the camera would not provide much difference in focus in say a landscape (uless the movement was in meters or even kilometers which is unfeasible).

However, when shooting macro or close-up imagery, focus can be selected in two ways: First using the focus capability (most often manual but, sometimes auto-focus) OR changing the lens to subject distance... This is most effectively done with a focusing rail although sometimes it can be done without a rail, either hand-holding or tripod/monopod mounted

Obviously focus changing of "normal" images would be done by the focusing mechanism of the camera/lens. However, when shooting close-up or macro images, which is the preferred way to change the focus between the shots: using the lens' focus mechanism or using a focus rail and changingthe lens to subject distance.

Are my ideas of focus changing correct:

If I were changing the focus point with the lens focus mechanism, I would first focus on the closest area I desired to be in focus and then focus on the rear area. I would note where the focus for both points lie and then make several exposures from front to rear dividing the distance ito several focusing points.

I would do approximately the same thing when using a focusing rail but, the change in focus would be accomplished by moving the camera and shooting at several points along the distance between when the camera is focused on the front point to where I want the camera focused on the rear point.

[carregwen]

Did anyone else notice that if you saw this in the Latest Posts or Latest Threads window (where you just get the opening few words), it said, "carregwen, that toad"?

Well!

(Spoken in Monty-pythonesque voice...) Stop this sketch it's silly... I'm nothing like Toad, more like Mr Ratty.

[McQ]

I need to clear up an area which somewhat confuses me.

When you use the term "focus distance", do you mean "the distance focused on" or the "lens to camera distance".

I'm referring to the distance focused on (from lens to subject). So if a person were 10 feet away and you focused on their eyes, then for the purposes of this article the "focusing distance" would be approx 10 feet.

Obviously focus changing of "normal" images would be done by the focusing mechanism of the camera/lens. However, when shooting close-up or macro images, which is the preferred way to change the focus between the shots: using the lens' focus mechanism or using a focus rail and changingthe lens to subject distance.

This is a really interesting question, and could honestly be another topic unto itself. For most practical purposes they can be treated as having the same effect, but they're certainly not identical. More specifically, with each method you'd have the following consequences:

1. Internal Focusing. Focusing more closely works by increasing the effective focal length of the lens (the lens to sensor distance). This decreases the angle of view, but maintains the same perspective. The image will therefore appear to be zoomed in slightly, and the depth of field will decrease.
2. Focusing Rail. Focusing more closely works by increasing the lens to subject distance. This maintains the same angle of view, but changes the perspective (the field of view increases). The image will therefore appear to be zoomed out slightly (although no zooming out has occurred -- the lens is just further from the subject), and the depth of field will remain unchanged.

With the focus stacking technique, the primary advantage of using a focusing rail is that each focusing position can be equally spaced (such as every millimeter), since the depth of field remains the same for each position. When focusing using the lens itself, the focusing positions need to be more closely spaced for nearer subject positions, since these have a shallower depth of field.

The primary advantage of using the lens to focus is that the perspective remains unchanged. In theory this would mean that the photo merging process could be performed more accurately (and realistically), but in practice this may or may not be visible, and really depends on the subject layering.

However, the above differences are really nitpicking; you'd have to be taking an extremely high resolution and/or technical photograph to notice the difference. I'd just use whichever method you find easier to implement.

[rick55]

Hi, Sean;

Two beautiful tutorials, very clear. They keep to the key points very well, as usual.

Some picky comments, also as usual.

Part I

In the section, "WHEN TO USE IMAGE STACKING," "Low-light photography," starts, "one may wish to freeze motion in part of their image (such as with a moving subject)...." Moving subjects are pretty problematic for focus stacking, of course, as you point out during the tutorial. I suggest that the very long exposure is enough justification. Long exposure can bring noise concerns, and can be troublesome when there may be activity around the scene, like boats in a harbor with lights to blow out the shot. This would be a good place to reference Part II, perhaps.

Same section, paragraph that begins, "What's the solution?" Third sentence "... avoids many of the disadvantages of using a high f-stop, ...," maybe "using an extreme f-stop" or "using too high an f-stop."

In "OVERVIEW," step 3, last sentence, "Otherwise specialized software packages such as Helicon Focus, TuFuse, CombineZM or Adobe Photoshop CS4+ are required." Nitpicking, I wouldn't call CS4+ a "specialized software package." Maybe, "Otherwise specialized software packages such as Helicon Focus, TuFuse, CombineZM, or the stacking capabilities of Adobe Photoshop CS4+ are required."

Part II

First sentence, "Image stacking can be a powerful technique for improving the quality of a photograph." Isn't it more to capture a scene that would be hard to capture otherwise? Although I do see a quality improvement aspect in helping overcome corner sharpness issues. Maybe, "Image stacking can be a powerful technique for dealing with scenes that could be impossible to capture otherwise, and can even help overcome some of the typical design limitations of lenses."

Section heading "AUTOMATED SHARPNESS SELECTION IN PHOTOSHOP," should maybe be "AUTOMATED SHARPNESS SELECTION," since it references other tools. The section also starts by saying, "Alternatively, you can use specialized software to auto-select the sharpest regions of each photo (such as TuFuse, Helicon Focus, etc.)," but then the instructions are written only for Photoshop.

Cheers,
Rick

[Geoff F]

Of the suggested software for combining images; I tried Helicon (the free trial programme) but found it produced excessively oversharpened results; Combine ZP wouldn't work on my computer and I have been unable to successfully download and install Tufuse.

But the normal stupidity warning may well apply to me in each case!

I have on occasion successfully manually combined two photos to increase depth of field, but it does take a bit of very careful alignment so a workable automated system would prove useful to me.

[Andy]

....or just want to add something from your own experience.

It's been a while since I shot a stack* but I encountered a few issues that I see as limitations or things to be aware of:

1. Complex shapes with over-lapping areas. It's a bit like shooting through bars at the zoo.....objects in front of what's in focus are larger when they are out of focus so obscure detail and lower contrast behind. You can end up with weird fuzzy bits round the legs, palps and hairs on spiders for example....

There was something like 30 shots went into the above....shot at about 3:1 I think.

2. Perspective. If you make everything sharp by stacking there's none....I think....either way I reckon this looks a bit weird....

It was a flat bit of machined alloy that I shot from an angle but I feel stacking kinda flattened it out and the image doesn't read right to me
(I picked an easy subject to mess about with one night There was 45 shots went into the above)

3. Unnaturally abrupt drop off in focus:


* I tend to prefer moving subjects and less time at the PC so rarely stack other than for the sake of trying out the technique.

[McQ]

Part I . . .

Hi Rick -- thanks for all of the feedback. I've updated the two pages to address almost everything that you mentioned. In some cases I still left parts of the text in the article (such as with freezing motion in low-light), but changed the order/emphasis.

[McQ]

Of the suggested software for combining images; I tried Helicon (the free trial programme) but found it produced excessively oversharpened results; Combine ZP wouldn't work on my computer and I have been unable to successfully download and install Tufuse.

But the normal stupidity warning may well apply to me in each case!

I have on occasion successfully manually combined two photos to increase depth of field, but it does take a bit of very careful alignment so a workable automated system would prove useful to me.

Hi Geoff -- thanks for sharing your difficulties with each software package. I've generally found that Helicon produces the best results, but this often requires fine-tuning the settings before it comes out just right (such as the radius and smoothing values, for example). However, unless the focus stack is really complex, Photoshop CS4+ is by far the easiest to implement and achieve adequate results. Otherwise manually combining these also works, as you say, but that can be *very* time consuming if the layering isn't simple . There might also be other software options, but the ones I've listed are probably the most popular...

[rick55]

In some cases I still left parts of the text in the article (such as with freezing motion in low-light), but changed the order/emphasis.

Hi, Sean;

Since you left this in the focus stacking section, I realized that I must not have gotten it, so I went back and reread it. I see the concept now: in your swan example from Cambridge, you could have done it with focus stacking. When I pointed out that motion is a problem in focus stacking, I missed the case where the motion takes the subject out of the frame for the rest of the set.

It might be helpful, in case someone joins who's as dense as me, to reference the swan example in Part II, and point out that focus stack will work as well as f-stop stack. Alternatively, perhaps note in the "capturing the photos" section that in the "freezing the motion" case, you're looking for one frame with the moving subject.

Cheers,
Rick