Helicon Focus Information

I've been asked about using Helicon Focus software by about half a dozen folks over the past month in separate emails or questions associated with image comments. I thought I'd post this here, if Guy has no objections, in order to try and get the content mostly in one place.

Helicon Focus is a software application that affords the photographer more creative control of the depth of field in an image than would otherwise be possible. Rather than making one image, a series of images is created and then input to Helicon and merged to create a single output image. I believe the application was initially created for macro photography to facilitate the creation of high detail images that also show extended depth of field. However, the software can also be used in landscape or product photography settings as well. It's worth noting that the software can be used not only to extend dof but also to minimize it in situations where it's desirable to leave some portion of the image out of focus.

Here's my approach based on about three years of using the product. I'm by no means an expert with it but the following seems to give me the best results:

Shoot RAW.

Use a tripod.

Shoot in manual or aperture mode.

I use the exposure meter display in my Nikon viewfinders and change the exposure compensation as I create the image stack to maintain consistent exposure in each image.

I shoot beginning with the nearest image elements in focus and move the focal plane towards the back of the subject. Since Helicon Focus will omit any part of the subject that's not included in every image in the stack, this lets me control what I keep in the image by selecting it in the first image in the stack.

In macro settings I use a rail to set my starting point for the stack, then proceed to move the focal plane through the portion of the subject I want in focus by making small adjustments, to the focus ring of the lens. The adjustment size will depend on the lens you use and the size of the image subject. True macro photography will require very small focus ring changes in each image.

Move the focal plane by using the focus ring on the lens and not by moving the camera on the rail, if you're using a rail. At least I get better results this way.

Macro photography may require a stack of fifteen or more images to attain the desired dof.

Don't rush the shots. Let the tripod and camera settle after each shot. And obviously you want to try to avoid moving the tripod as you make the lens ring adjustments.

Choose an F/stop that best suits what you're trying to accomplish in the image. In situations where you want maximum, sharp dof, use the sweet spot of the lens with regard to its IQ. However if you're trying to blur bg, then set the lens aperture to the widest F stop needed to make the subject sharp and the ground blurred.

Shoot all images in the stack at the same F stop.

Make sure that there's some overlap of in focus portions of successive images in the stack. I usually try for about thirty percent from one image to the next but again with true macro scales, it's impossible to be this accurate, so I just try for overlap.

It helps to plan out how many images you'll need in the stack by making a trial pass through the subject.

Ideally, I'd probably have seventy percent or so overlap in images but there's a tradeoff between the number of images in the stack and the likelihood of shifting the tripod accidentally at some point in the process. What you do should depend on how good your tripod technique is and what the shooting conditions are like. Even a slight wind will cause the process to fail because of subject movement.

As soon as the output image is created in Helicon, check it for artifacts. These can be ghosts of image elements or areas that have obviously incorrect tonal value. The ghosting is related to system movement while the latter is caused by changes in exposure during the process. That's why it's important to meter each stack image and adjust the exposure compensation accordingly. When I started using the product, I saw artifacts about 75% of the time. I've now got it down to about 10% of the time.

Sometimes you can remove one problem image from the stack to mitigate an artifact. The professional version of Helicon provides a tool for cloning out artifacts. I've never used it because I'm more familiar with the clone tool in PS.

The output image from Helicon will need sharpening.

My usual settings in Helicon Focus are Method B (depth map), radius 3 and smoothing one.

I use a RRS BH40 ball head and a Kirk macro focusing rail and I do my RAW conversions using Lightroom 3.

My experience indicates that stacking reduces image noise. So, assuming the wind permits, I now routinely shoot two or more images and run them through Helicon if I'm shooting under exposed as I normally do for flora images.

I use Helicon for about seventy percent of my work. It's a pain to create the stack but it affords a lot of control for dof. I haven't used a tilt and shift and would love to do so at some point but until I feel like I can afford that type of lens, I'll keep Helicon loaded on my pc.

Pat

Pat,
thank you very much for sharing your experience and for this useful explaination.

I think the only (minor!) correction is that Helicon Focus actually came from the microscope community, which is another step past macro.

Good summary and largely matches my experience with the program.

Hi Pat,

You put a lot of good work into this.

"I use the exposure meter display in my Nikon viewfinders and change the exposure compensation as I create the image stack to maintain consistent exposure in each image."

That's something I've not tried. Ah, the next experiment.

"Don't rush the shots. Let the tripod and camera settle after each shot. And obviously you want to try to avoid moving the tripod as you make the lens ring adjustments."

An extremely important point, and dependent on the subject, no A/C, no fans, no breathing. I used MLU and a lot of patience.

"The ghosting is related to system movement while the latter is caused by changes in exposure during the process."

I've read that the ghosting is also related to the transition between images. And you may well be right. The cause may indeed be due to uncorrected exposure changes.

"The output image from Helicon will need sharpening."

Interesting comment. Hasn't been my experience.

"My experience indicates that stacking reduces image noise."

Pat, I've read this: The improvement in noise would be proportional to the square root of the number of images stacked.
I need to reading further.
And, your statement does make sense, especially if you are correcting for exposure in your stack since shadow areas in particular have more noise if they are under exposed.

Helicon is now working on a software product they are calling Helicon Remote.

It is available for Windows XP, Vista, and 7, and they are currently working on the MAC version.

Oh, so much to learn...

Pat, thanks for taking the time to put this information together. I hope others who are working with stacking images reply to this topic too.

Carol

"The ghosting is related to system movement while the latter is caused by changes in exposure during the process."

I've read that the ghosting is also related to the transition between images. And you may well be right. The cause may indeed be due to uncorrected exposure changes.

When one is working in the close-up range or macro or macro range the extension built into the lens kicks in. So when you say you are focussing at different points you are essentially reducing or increasing the magnification all the time. So is the case with using focussing rails also.

The "leftovers" (call them artifacts or another name) are simply the inability of the software program to use all available information in the several frames to make all the images overlap perfectly.

I am willing to stand corrected, if my understanding is incorrect.

Great Article, Pat - I find it extremely useful.

Carol,

Just for clarification: In my work I've seen two different kinds of artifacts: the first is ghosting and the second appears as areas in the image where the tone and to a lesser extent hue may obviously be wrong. I think the second type of artifact, affecting tone is caused by variation in the exposure during the process.

Ram,

Your explanation for the first type of artifact makes sense to me. My problem is that most of the time the algorithm accommodates magnification changes without creating artifacts. I don't have any idea how that's done. I have done twenty image stacks at macro scales with no artifacts. So, I think there's more at work than just the magnification causing the artifacts but I don't know what it is. Possibly as you suggest, it may be the extent to which the magnification prevents the image registration in particular scenes. I'd love to know how the algorithm actually works.

Pat

You mentioned underexposing and I've found that to be important when shooting anything with white in which you want to maintain detail in those whites. For some reason I can properly expose the whites in the individual images but run them through Helicon and they often come out with blown highlites. I find if I underexpose around 1/2 stop or so this prevents that from happening.

Dan is correct, Helicon was initially developed for photomicroscopy. I find it invaluable for macro work....give it a try.

Nice job on the article, Pat.

Pat, Ram, Mark, et al,

Just did two experiments on a leaf.

Both were shot at f/8. Both were processed on HF using a radius of three and smoothing of 4 using a depth map. In addition the processing was accomplished using the Lanczos8 resampling technique. In the first, I used 15 images, and in the second, twice as many. I'm posting a B/W version of the first trial. In the second image I picked up a teeny-tiny touch of ghosting on a small section of the leaf close to the stem end near the center vein.
I'm making an assumption that it's from a plane I missed.
Without an instrument to gauge the plane of focus, no matter how careful we are, I think it's possible to miss a plane in which there's a variation of light and/or shadow and I think that's where a problem may lie.
Pat, I also have a question, I wonder if our cameras are sufficiently calibrated to catch a tiny shift in light, especially if it lies outside the point of focus? During both of these experiments, I kept a careful watch on exposure and nowhere did it shift and need to be corrected.
I'm assuming we all do one or two test images to check ISO, exposure, etc. before we actually shoot the stack.
Also, another question pops it's ugly head. Another variable that no one has mentioned is the Dynamic AF area. I'm using all 51 points. Would this come into play with our exposures?
And Pat, if I'm understanding the way you do your exposures, you're not really using the focusing rail, only the focusing ring. Is that correct?
In the interim, I sure hope we continue to share information. It sure helps, or may, simply leads to questions needing further clarification.
Certainly a combination of an art and a science.
Cheers to all,

Carol

Carol,

I use the rail to find my beginning image for the stack. After that, just the focusing ring on the lens.

Pat

Great article, Pat! Thanks a lot.

I am enjoying looking in on this thread, but when I think of doing all those shots for a stack, I begin to get pale and woozy! Goodness, all of you have oodles of patience.

C'mon in, Catherine. The water's fine. Give it a try



Carol