Correcting for Refraction (or not!) in alcohol

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

Moderators: rjlittlefield, ChrisR, Chris S., Pau

Sam Droege
Posts: 14
Joined: Thu Jun 07, 2012 5:34 pm
Location: Patuxent Wildlife Research Center
Contact:

Post by Sam Droege »

Some insects are so soft bodied that they simply shrivel and collapse when in the air (many aquatic insects for example as well as flies and spiders).

Additionally, very small insects, ideally, could be suspended for photography without having to deal with the occluding nature of a pin or other holding device. If the resolution could be worked out one can imagine the ability to layout groups of specimens and using good staging, stacking, and stitching layout nice "mini posters" of specimens exactly as you would like to see them.

I will play around with adding propylene glycol (about the same RI as the cuvette wall and we have plenty available) to see what that alters in terms of resolution and potentially lighting.

Many thanks for the suggestions

sam

Craig Gerard
Posts: 2877
Joined: Sat May 01, 2010 1:51 am
Location: Australia

Post by Craig Gerard »

Glycerin may also be an option; it's 'thicker' than PG and may assist with suspension.

http://www.youtube.com/watch?v=pNWCB_GoQA4


Craig
To use a classic quote from 'Antz' - "I almost know exactly what I'm doing!"

NikonUser
Posts: 2693
Joined: Thu Sep 04, 2008 2:03 am
Location: southern New Brunswick, Canada

Post by NikonUser »

I store much of my small insect stuff in drugstore glycerine and photograph these whilst still in glycerine. Allows for changing the orientation if needed - not possible for a Canada Balsam mount.

A fly genitalia here, scroll down

http://www.photomacrography.net/forum/v ... =genitalia
NU.
student of entomology
Quote – Holmes on ‘Entomology’
” I suppose you are an entomologist ? “
” Not quite so ambitious as that, sir. I should like to put my eyes on the individual entitled to that name.
No man can be truly called an entomologist,
sir; the subject is too vast for any single human intelligence to grasp.”
Oliver Wendell Holmes, Sr
The Poet at the Breakfast Table.

Nikon camera, lenses and objectives
Olympus microscope and objectives

Sam Droege
Posts: 14
Joined: Thu Jun 07, 2012 5:34 pm
Location: Patuxent Wildlife Research Center
Contact:

Post by Sam Droege »

Exquisite fly genitalia shots. An not taking anything away from the photography it would also appear to me to have the same loss of resolution compared to it the same structure were in air...

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

Post by rjlittlefield »

Sam, I've looked very closely at your images linked above (http://www.flickr.com/photos/usgsbiml/9736455115/ and http://www.flickr.com/photos/usgsbiml/9736465307/). I've even put crops of the Flickr "original" size images side by side on the same monitor so that I could directly compare corresponding anatomical features.

To be blunt, the differences I see simply do not look like what I would call "loss of resolution". The arista are equally well defined; so are the small hairs on the base of the antennae and around the eyes; and I see all the same textural features on that "rhinoceros horn" structure at lower left in the air image. (What is that thing, anatomically?)

If there were to be a loss of resolution, say from stopping down farther, then the arista, hairs, and texture on the "rhinoceros horn" would be more blurred, with slower transitions from dark to light. But I'm not seeing that.

Instead, I'm seeing a general loss of contrast, greatly increased pixel noise, and differences in the illumination pattern. Those are important, but they're not resolution losses, and I think it will be simpler to work out what's going wrong if we can first agree on what symptoms we're trying to correct.

On a related subject, I'm also seeing strong halos around features like the arista. That makes me think that both these images have been sharpened strongly. Obvious sharpening, combined with the great differences in what looks like pixel noise, makes me edgy about reading too much into the images as shown. I would be more confident if we had some unprocessed images to look at, or at least image pairs that I was sure had gone through exactly the same processing.

--Rik

Sam Droege
Posts: 14
Joined: Thu Jun 07, 2012 5:34 pm
Location: Patuxent Wildlife Research Center
Contact:

Post by Sam Droege »

Rik

Good observations. This is great CSI. You are correct. Both pictures would have gone through my standard default sharpening in Camera raw which is

Amount: 150 (the max)
Radius: 1.9
Detail: 25
Masking: 0

I have the original tif's in the lab (still on travel today) and propose to dig them out and process to jpegs without any additional processing...

After that we can design some experiments to explore hypotheses about lighting or other factors affecting pixel noise/resolution etc and I will shoot them.

How's that sound?

Thanks.

sam

Sam Droege
Posts: 14
Joined: Thu Jun 07, 2012 5:34 pm
Location: Patuxent Wildlife Research Center
Contact:

Post by Sam Droege »

Oh yes, I am not a fly biologist...so am not sure about the "horn" If it were a bee I would suspect a sheath section for the tongue.

sam

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

Post by rjlittlefield »

OK, I have some experimental data that may help to explain what's going on with this photographing-in-alcohol problem..

The experiment was conceptually simple. I took an old honeybee out of my freezer and pinned it into a vertical setup that allowed me to photograph it dry, then flood it with 99% isopropyl alcohol and photograph it again with no changes to the optics or illumination.

Shot with a Canon T1i, 15 megapixels APS-C, using Canon MP-E 65 lens at 3X and f/5.0, illumination with two Jansjö lamps diffused through facial tissue.

Here are the results.

First, full frame:

Image

Clearly there is a huge difference in appearance of this subject depending on whether it's wet.

Looking closer at selected 100% crops, we have these pairs:

Image

Image

Image

So, does shooting through the alcohol produce a loss of resolution? On careful study, I don't think it does. As we saw earlier with the chrome-on-glass target, shooting even through a very thick layer of liquid does not degrade resolution at this relatively small aperture. I see the same thing in these pairs, when I look specifically at dark details that are silhouetted against bright background.

However, it's clear that there's a big loss of detail. The image in air shows dust, hair, and cuticle texture pretty much everywhere, while the image in alcohol shows almost no cuticle texture and greatly reduced contrast for the dust and hair. Some of the dust particles and hair completely disappears.

Why is this happening?

I think it's basically the same issue noted earlier by g4lab: if you want to suppress surface reflections, then you immerse the specimen in liquid medium whose refractive index matches the specimen.

But of course the system doesn't care whether you want to reduce the reflections, it just does that regardless, depending solely on the RI's involved. Immersing the specimen in a medium whose refractive index does not quite match the specimen will not completely eliminate reflections but will significantly reduce them. That's what we have here. Refractive index for air is 1, for isopropyl alcohol is 1.38, and for chitin is about 1.55. Closer RIs, less reflection, less contrast, less detail.

The eye is an interesting case. I suspect this specimen had dried out quite a bit in my freezer. So in the eye, I think we're looking at a thin layer of air just underneath the cuticle. In that case, adding alcohol darkens the dark parts by reducing reflection from the chitin, but does not affect the bright parts because there is still an air/chitin interface just behind the transparent cuticle. At the same time, the partially occluding hairs and dust are made less reflective by the alcohol, so the net effect in that case is actually an increase in contrast and clarity. Interesting...

I hope this helps.

--Rik

ChrisR
Site Admin
Posts: 8668
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

Nice demo and workout Rik, I'd got as far as thinking about diamonds hidden in glasses of champagne...

johan
Posts: 1005
Joined: Tue Sep 06, 2011 7:39 am
Contact:

Post by johan »

Fascinating, super research
My extreme-macro.co.uk site, a learning site. Your comments and input there would be gratefully appreciated.

TheLostVertex
Posts: 317
Joined: Thu Sep 22, 2011 9:55 am
Location: Florida

Post by TheLostVertex »

Very nice demo Rik. So one solution to this issue would be to use a medium with as low of a RI as possible to immerse the subject in(clearly an impractical solution). Would another potential solution be to alter the lighting to try to create stronger reflections or contrast, since they are being reduces so heavily? Or can you think of another possible method for reducing the effect demonstrated?

Also would this mean that for the same set up, glycerin as suggested earlier, would show this effect to a larger degree? Since the RI of the chitin is 1.55, and the RI of glycerin is 1.47 compared to the 1.38 of isopropyl.

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

Post by rjlittlefield »

TheLostVertex wrote:Also would this mean that for the same set up, glycerin as suggested earlier, would show this effect to a larger degree? Since the RI of the chitin is 1.55, and the RI of glycerin is 1.47 compared to the 1.38 of isopropyl.
I expect that this is true.

I'm working on an experiment right now to check that experimentally by replacing the IPA with glycerin on this same specimen.

That effort is showing another issue.

I siphoned the IPA out of the container and started drizzling glycerin periodically over the specimen to flush off the IPA and (hopefully!) get almost complete replacement. But over an hour later, I'm still getting obvious refractive streamers coming off the specimen.

I don't know exactly what process Sam's people are using to place the specimen in hand sanitizer for photography. But if the RI of hand sanitizer happens to be different from the RI of whatever the specimen was in before, then it seems like there's some risk of getting refractive "bubbles" at the interface of the two media. That would be another possible source for degradation.
Would another potential solution be to alter the lighting to try to create stronger reflections or contrast, since they are being reduces so heavily? Or can you think of another possible method for reducing the effect demonstrated?

To be honest, I don't see any good solutions to this problem. Air is just a better medium for revealing surface texture. Sure you can adjust the illumination, but it seems like the best you'll ever get in dense medium won't be nearly as good as the best you can get in air. It's one of those "costs of doing business" with soft specimens.

--Rik

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

Post by rjlittlefield »

Here's an overview of results from the glycerin experiment.

Image

The experiment was not entirely successful because I ended up with quite a few small bubbles stuck to the specimen. But you can clearly see the general trend toward less reflection from the top surface of the specimen as the RI of the medium increases. (There's also a striking change in the appearance of the eye, but that's because the eye filled with liquid overnight.)

It occurs to me that these last two images, shot through alcohol and glycerin, look remarkably like they were backlit.

Let me be very specific that all the pictures were shot with the same front illumination setup. Here it is:

Image

The problem is just that a honeybee wet with glycerin looks very dark, almost black, because there are no significant reflections from the hairs or the surface of the bee.

--Rik

Sam Droege
Posts: 14
Joined: Thu Jun 07, 2012 5:34 pm
Location: Patuxent Wildlife Research Center
Contact:

Post by Sam Droege »

Rik et al.

Been hammered by cold the last couple of days so missed this last bit of work. Very interesting....and I think that this is nailing the issue down.

Hand Sanitizer does have bubble issues ... we usually soak the specimen in HS overnight or several days and then transfer to cuvette the next day. We use a thing syringe to remove the bubbles.

I do think that there might not be any easy solution to increasing the detail, and come to think of it, aquatic organisms appear to have more smoothed out, broader color patterns compared to the finer details of hairs, pitting, and such in terrestrial insects.

I imagine that lighting could help/hinder the creation of more detail, but that ultimately Rik is probably right that this is the cost of working with wet things.

On the other hand it would be nice to have a nice air gel.

Thanks everyone. I will pass this along to the alcohol crowd who will be very interested.

sam

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

Post by rjlittlefield »

One more interesting bit of info regarding photographing in liquid is that you get more DOF at the same resolution. How can this be??!

First, let's just take it as given that the apparent depth of the subject varies inversely with the RI of the surrounding medium.

As an experimental confirmation, I carefully trimmed the above stacks during processing so that they started and stopped exactly at the same points on the subject. All stacks were shot using a StackShot rail with focus step of 0.10 mm.

The observed apparent depths were:

Air: 4.3 mm (44 frames)
Alcohol: 3.2mm (33 frames)
Glycerin: 2.9mm (30 frames)

I have just now calculated what would have been expected based on RI=1.38 for isopropyl alcohol and 1.47 for glycerin.

Air: 4.3 mm (by observation from above)
Alcohol: 3.12 mm (by calculation)
Glycerin: 2.92 mm (by calculation).

Considering this aspect more carefully, I was surprised to find that it's one of those rare cases where there really is a "free lunch". While the apparent depth of the subject is reduced by immersion, the DOF and resolution of the optics remains unchanged. As a result, the effective DOF on subject is increased.

Here's the demonstration of that (3X, f/2.8 ):

Image

Image

You can see that the width of the OOF blurs is significantly larger in air than in water.

It took me a while to make complete sense of this from a theoretical standpoint. One of my standard mantras is that diffraction forces a strict tradeoff between DOF and resolution, and that's apparently violated here.

However, that mantra only applies when the wavelength of light is held constant. When the subject is immersed in dense medium, the wavelength at the subject shortens in proportion to the refractive index, and it's that shortening of the wavelength that allows the tradeoff to shift.

From a geometric standpoint, what happens is that for a fixed aperture the angle of the entrance cone is narrower in dense medium than it is in air. As usual, a narrower cone implies more DOF so that part is simple. The trick is to realize that the shortened wavelength plays off against the narrower cone in such a way that the resolution ends up being the same in both cases. I won't trouble you with the math.

--Rik

Post Reply Previous topicNext topic