Improving My Microscopy Technique

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

Moderators: MacroMike, nzmacro, Ken Ramos, twebster, S. Alden

Locked
lacerta
Posts: 139
Joined: Sat Feb 12, 2005 10:16 am
Location: Georgia, USA

Improving My Microscopy Technique

Post by lacerta »

I have learned a lot since becoming a regular visitor on this forum. But so far I feel that I am not squeezing the most out of my camera's resolving power when taking stacked images. I know that there are inherent limits of performance with any camera, but I know that my Sony F717 (5 mp) can take some very sharp pictures when shooting garden variety photographs such as landscapes, portraits, etc. What might I be doing wrong when shooting through the microscope ?? I have made the following improvements and am open to suggestions:
1) To reduce camera shake I am now using microscope adaptor (M99 from Martin Microscope) that weds my camera tightly to the eyepiece.
2) I built a much more solid table to place my setup upon.
3) I am now using an electronic cable release for hands-off operation.
4) To remove vignetting, I use only optical zoom and avoid the digital zoom range
5) I use the manual focus mode to lock the focus on the camera. Otherwise the camera struggles at times to achieve what it thinks is focus by zooming in and out.
6) I use dual head fiber optic lighting to provide adequate light and avoid digital noise.
7) Built a set of diffusion domes (thanks Charlie for your ideas) to reduce specular reflections.
8) Use only mounted/immobile specimens.

Despite vast improvement over my earlier attempts at using a cheaper camera on a tripod and looking into the eyepiece, I feel that I could be getting better results. I was wondering if most digital cameras have optimum settings that provide the best resolution? I have been shooting with an aperture close to wide open to allow for faster shutter speeds. But with the camera solidly affixed to the microscope and with a dead immobile subject and with plenty of light I question if fast shutter speeds are really needed. I wonder if stopping down the aperture would improve image quality. I know that shooting a small aperture improves DOF, and that a large aperture allows more light, but isn't there a sweet spot on most cameras that provide the optimum optical resolution? Provided that there is sufficient light, do most cameras shoot better at F8 than at F2.4?? How about ISO settings? Provided sufficient light, wouldn't the slowest possible setting provide the best image quality? I know with film that Kodachrome 64 has much better color and finer detail that a high speed film such as ASA 400. Is this not also true with digital cameras ??
To improve the technical quality of my images I have concentrated on three areas: reducing camera shake, reducing noise, and using the best optical equipment I can currently afford. Other than taking private lessons from Charles Krebs, is there anything more I can do, or is there something I am overlooking? Your thoughts?
George

User avatar
Ken Ramos
Site Admin
Posts: 4809
Joined: Tue May 04, 2004 7:58 pm
Location: Western North Carolina

Post by Ken Ramos »

Read over your list George. Looks like you have covered it quite well. Good photographs are essentail to scientific study and I too strive to get the best that I can and would undoubtedly like to have them better but I am limited by finances. I too have learned a lot from this site and the photographers on it. Hopefully one day I may put what I have learned to use but at the moment I just make the most of what I have.

You mentioned wide aperatures, slow shutter speeds, and ISO/ASA. From what I have gathered here and from stumbling around on my own, I have come pretty much to your conclusions. A wide aperature really, to me now, makes some difference but not of any significance with the digital camera that "I" use, a Sony P-200. But I have found out that a slow shutter speed contributes some to noise in the image which affects the clarity of the image and that the lower the ISO, just as in camera film, the better. Less grain in relation to film, less noise in relation to digital with a low ISO. Cable or remote releases are pretty much a must and so is a rock solid platform for both camera and microscope. Also shoot at the highest megapixel that you can. This is what Charlie advised me to do when I was using my Canon G5. I found out most of the time when I resize the image, I need to do less digital sharpening. My set up is similar to Charlies, but not like it, as far as camera mounting goes. However it is on a much smaller scale but provides good stability using the average commercial fixed lens digital cameras.

Grahams set up, the last I looked, is more like what I am using but I do not have a flash unit installed and Graham gets some excellent results from his using a flash. Charlie definitely made a statement by showing how much a flash improves the quality of your images and microscope objectives.

That is about all I can add George. Your questions are undoubtedly going to get a lot of good responses in the days to come. :D
Site Admin.
Kenneth Ramos
Rutherfordton, North Carolina
Kens Microscopy
Reposts of my images within the galleries are welcome, as are constructive critical critiques.

rjlittlefield
Posts: 727
Joined: Fri May 06, 2005 11:57 pm
Location: Richland, WA, USA
Contact:

Post by rjlittlefield »

George,

I've looked over your previous posts, and I see a huge range of subject, method, and equipment. I also see a lot of really good images. But I can guess that you're wrestling with that new Reichert (sp?) Stereo Star Zoom, especially if you're comparing against Charlie's images ;-)

So...

Can you clarify what is the current setup or setups that you're uncomfortable with? (I might guess that you're using a Sony DSC-F717 through a Martin Microscope MM99-58 to a Reichert Stereo Star Zoom, but that could be wildly wrong!)

And can you post out some pictures and talk through your concerns?

For example, are you losing resolution across the whole field, or more toward the edges? Are you having problems with curvature of field? Chromatic aberration? At all zoom levels, or just some?

You mention stacking. Do your problems appear in single images, or is it the processed stack that is disappointing?

Have you done a careful comparison of what your camera sees, against what you can see directly through the microscope? Sometimes a direct look-at-one, look-at-the-other will point you in the right direction.

You asked if stopping down the aperture would improve image quality. I think the answer is a resounding "maybe", and the only way to tell for sure is to try it. Yes, every optical system has some kind of "sweet spot" that trades off diffraction against lens aberrations and geometric blur. But the more complicated the system, the more difficult it is to understand & predict. If you're talking about stopping down the camera lens, which is being fed by a relay lens, which is being fed by a microscope objective, which is being fed by a condenser system, then you're in an area that I'm still trying to figure out too. Try it, and if the results don't make sense, show us the pictures and we'll puzzle over them too.

--Rik

lacerta
Posts: 139
Joined: Sat Feb 12, 2005 10:16 am
Location: Georgia, USA

Post by lacerta »

Thanks Ken and Rick for your informative responses. I guess what I really need to do is more experimenting with different camera modes( aperture priority, speed priority) with different combinations of settings and do some comparative analysis of images of the same subject. That is something that I have yet to do.
I think some of my disappointment comes from inspecting my images in Photoshop using the "actual pixels" and getting a blown up view of detail. I have some images I have taken (portraits, some macros) that retain some very impressive resolution. In comparing these to my microscopy photos, I definately notice a degradation in image quality.

And certainly what I can see in the microscope is stunningly more detailed than what is recorded in the digital image. I guess no recording media can beat the human eye but this tells me that my microscope optics are pretty good. I have looked through many microscopes during the last few years as a biology student and have a pretty good feel for what is good and bad.
I have two microscopes. My compound is a National Optical DC5. For a Chinese made scope (don't laugh, I also balked at first!) it is really a very good value and I am quite happy with it. It is equipped with upgraded objectives (plan achromatics) and an integral CMOS for video feed to computer via USB2. The on board digital is a low-res 2 mp but records decent video clips and seeing the live image on my 17" computer screen relieves me of some of the fatigue during long sessions of looking directly into the binoculars. Having the image on the monitor also works well for attaching the Sony F717 to one of the eyepieces. The camera also has a pivoting body that allows easy viewing of the image on the camera's LCD viewfinder. With the the focus set to "manual" on the camera, I can get the image parfocal to what I can see on my computer monitor, at least close enough for stacking images. I recently acquired an electronic cable release. It turned out that reducing camera shake, especially when shooting wet mounts, was probably my biggest improvement, along with fixing the camera to the eyepiece with the MM99 adaptor. The MM99 also allowed me to upgrade my camera from my wife's pocketbook digital (P100) to the camera I am currently using (F717). The adaptor has stepdown optics (like a reversed lens) that finally solved the problem of have a huge chunk of glass trying to peer down a 23mm hole.
My second microscope is the newly acquired Reichert Stereo Star Zoom. Reichert I have found was once a subsidiary of Leica. This stereo scope is a spitting image of the Bauch & Lomb Stereo Zoom 4, in fact the parts and lenses are supposedly interchangeable. After playing with it for a week or so I wonder how I got along without it. It is great for preparing permanent mounts for the compound, and manipulating specimens. It is so nice to be able to reach in with a pair of fine forceps or a pipette and know that "up" is really "up", and "left" is really left. Also great working distance between specimen and objectives to allow use of diffusion domes and fiber optic lighting, etc. My recent post of the buprestid beetle was taken using this scope and the same mm99 with F717.
So Rik, that is my modest set up so far. I still plan to build a flash (Tom Webster has a great article on this website). I hope to use that mostly for motile protozoa, and some of the parasitology work I have been doing. I guess flash is the only way to freeze rapidly moving cilia.
I have one more question for you or Ken if you don't mind. I was reading the basic micrsocopy manual on microscopy-uk website and they seemed to indicate to me that compound microscope objectives have been optically corrected for the spherical abberation of a coverslip. If that is true then veiwing anything other than a wet mount, or mount under a coverslip will result in a degraded image when using a compound scope. Is that true?? Looking at my objectives they are labeled, for example, "Plan 40X/0.65 160/.17". I read that to mean it is a Plan Achromat with a 40X magnification and an N.A. of .65, but what the heck is a "160/.17"?? The .17 is suspiciously suggestive of a .17mm coverslip, is it not? Hmmmm. If that is true than that will explain why I never got very good results using my compound like a stereo to photograph naked (to mean uncovered) aphids and insect parts. Everything looks so much better when viewed under a coverslip and suspended in some liquid medium.
Sorry long winded. Thanks again fellows.
George

rjlittlefield
Posts: 727
Joined: Fri May 06, 2005 11:57 pm
Location: Richland, WA, USA
Contact:

Post by rjlittlefield »

lacerta wrote:Looking at my objectives they are labeled, for example, "Plan 40X/0.65 160/.17". I read that to mean it is a Plan Achromat with a 40X magnification and an N.A. of .65, but what the heck is a "160/.17"??
George,

I'll be pretty busy for the next couple of days, and I hope/presume some other people will chime in before I can get back with you.

Regarding the objective labeling, see http://www.microscopyu.com/articles/opt ... specs.html . The objective you have is designed to work with a 160mm tube length (that's 160mm of empty tube, or an optical system that acts the same). Yes, it is corrected for a 0.17mm cover slip. See http://www.microscopyu.com/tutorials/ja ... index.html for some further discussion. Considering that the focal length of a 40X objective is roughly 4mm, you can imagine why inserting 0.17mm of glass would matter. It's roughly equivalent to setting up to photograph a portrait at 6 feet, and having somebody stick a 3" thick piece of glass in the way.

I am particularly bothered by your report that what you see in the microscope is much more detailed than what you capture in the digital image. If that's true, and also what you see at "actual pixels" looks fuzzy, then something about your photographic setup is causing the fuzziness. It is conceivable that the MM99 relay lens, combined with your camera lens, is not behaving well with your objective. (I understand that sometimes objectives and eyepieces are co-designed so that the eyepiece compensates for aberrations introduced by the objective.)

Do you have a friend with a digital SLR (DSLR)? If so, then the following experiment might be informative. Pull out the eyepiece, pull the lens off the DSLR, and set up the DSLR so that the microscope objective projects directly onto the sensor of the DSLR. If you don't have a microscope adapter for the DSLR, then you can just set up on tripod and use some black fabric as a light shield. Turn off the lights, focus through the camera (parfocal is most unlikely with this experiment!), and take some pictures. If the DSLR pictures come out significantly better than with the MM99 plus your camera, that'll be a clue. If they look about the same, then I'd suggest another look-at-one, look-at-the-other test to check that for sure the direct view shows more detail than the photos.

Regarding the stereo scope, I'm a bit conflicted about what to admit/recommend. The fact is, I am the very happy first owner of a circa-1970 Bausch & Lomb 10-45X stereo zoom. I use it very frequently, but I long ago gave up trying to photograph through it because the image quality was far inferior to what I could get by sticking a microscope objective on bellows on an SLR. On the other hand, a few months ago I was in the office of a professional biologist acquaintance, and I saw his collection of very nice images shot through the eyepiece of his Nikon (I believe it was...) stereo scope. So I think the answer there is, "it depends on the scope", and I have no experience with your Reichert or its Leica / Bausch & Lomb equivalent. Sorry...

One final thought... In terms of pixels per detail, it's pretty unlikely that your microscope images will ever compare well against portraits and even macros. Diffraction imposes some pretty stringent limits! But it certainly should be the case that what you get from your camera, particularly for a static subject and a stacked image, compares well with what you can see directly through the eyepiece of the microscope. Keep shooting for that goal.

--Rik

Charles Krebs
Posts: 1200
Joined: Tue Jul 13, 2004 10:50 am
Location: Issaquah, WA USA

Post by Charles Krebs »

George... a few thoughts, and a few questions to get more info. Rik touched on some good points that I would like to examine in greater detail.

Your compound microscope (National/Motic) is a "finite" tube length model (160mm). This was the most common "standard" before manufacturers starting moving to "infinity" optical systems. (The objectives I use are also finite, 160mm tube length). In almost every case, with this type of microscope, the objectives and eyepieces were designed to be used together. The objectives themselves did not fully correct all aberrations, and the "final" corrections were accomplished with the eyepiece. So the objective and the eyepiece formed, in essence, the "complete" optical package. Eyepieces for these systems usually (but not always) have a "k" or "c" marked on them someplace to designate them as "compensating" (or corrective, if you will) eyepieces. There was no standardization, and the degree of correction provided by the eyepieces varied between manufacturers.

(Some interesting reference to this can be found on these pages:

http://www.couger.com/microscope/Ted-Cl ... larke3.pdf

http://www.microscopy-uk.org.uk/mag/ind ... tccdm.html )

I am not familiar with the adapter you have, so one question is this... When you use the MM99 adapter you mentioned, are you still using one of your original microscope eyepieces, or do the optics in the adapter replace the eyepiece?

I suspect your objectives require a compensating eyepiece for best quality (the only finite tube length objectives I'm aware of that do not are the Nikon CF objectives). If this is the case, and you are not using an eyepiece designed by the microscope's manufacturer, and if the optics in the MM99 adapter are not appropriately "compensating" this may be one source of imaging problems. Hard to say how much effect this may have ... could be minor or huge! (When I started looking into photomicrography and asked a lot of questions, the "optics", and their implementation, that would best take the image from the microscope and "relay" it to the film or sensor was described to me by an experienced individual as the "Holy Grail". I learned that he was right! The best match is not always easy to find.)

My second question is this... how do you arrive at the best focus for the camera? Do you set the camera focus to manual and at"infinity"? The F717 is a great camera, but I can't imagine the electronic viewfinder being too easy to focus in photomicrography. And maybe I'm wrong, but I've always been doubtful about the accuracy of determining focus from a monitor. I'm at a disadvantage here because I have very little experience using anything other than a SLR on my microscope. But I do know that with my set up, the slightest mis-focus causes a huge hit on image quality.

As far as the aperture to use in the F717, my guess would be somewhere around f4 to f5.6. I would avoid the smallest and largest apertures. (But this is something you should be able to settle with some testing.)

A .17mm coverslip is "ideally" a #1.5. The amount an image deteriorates if you deviate from this number is rather dependent on the na of the objective. It really should not matter too much with most 4X and 10X objectives, as the numerical apertures are not particularly high. At the other extreme, say a "high, dry" objective with a na of 0.95 the image quality drops like a rock. This is why these objectives have correction collars to accommodate slight variations in cover glass thickness.

My experience (as Rik also mentioned) is that the detail I get with the pictures through the microscope is not as "impressive" as can be obtained in conventional landscape or closeup photography.

And for me, I can always see finer detail through the eyepieces than I have yet been able to record in the camera... but I'm always working on it!

Charlie

rjlittlefield
Posts: 727
Joined: Fri May 06, 2005 11:57 pm
Location: Richland, WA, USA
Contact:

Post by rjlittlefield »

Charlie,

Thanks for the links and discussion. I had no idea that most finite tube objectives require compensating eyepieces for best results. This is serious food for thought...

--Rik

Charlie Krebs

Post by Charlie Krebs »

Rik...

Yup, always more food for thought :?

There's a little more on this at the end of this article (also by Ted Clarke):

http://www.microscopy-uk.org.uk/mag/ind ... scope.html

I always thought it curious how this is not mentioned much when looking at adapters. I realize it should not be an issue with "infinity optics" scopes, but there sure are a great many "finite" microscopes still being used. But it needs to be examined on a case by case basis. I have a Nikon CF 2.5X photo-eyepiece, and an Olympus NFK 2.5X photoeyepiece. (My phase objectives are Nikon CF and should not need a compensating eyepiece , my others are Olympus, and were designed to be used with a compensating eyepiece). I must admit to occasionally using them with the "wrong" objectives without objectional results. Perhaps I should scrutinize it more carefully.

I would think that some testing with the appropriate green filter would be the easiest way to see if this issue that needs attention.
Charlie

lacerta
Posts: 139
Joined: Sat Feb 12, 2005 10:16 am
Location: Georgia, USA

Post by lacerta »

Thanks Charlie for the informative response. This is all new information to me. My adaptor replaces the eyepiece. I am now prompted to contact the manufacturer (Martin Microscope) and ask about the use of their product in lieu of a compensating eyepiece. I was sold on the MM99 because of the images I saw in their gallery http://www.martinmicroscope.com/Microgr ... dgeDIC.htm that compared several different cameras. My F717 scored consistently well and I was certainly impressed. Perhaps it was the Aus Jena Jenaval Apo scope that made all the difference. Sheeesh.
Regarding setting up parfocality. I find that I have to be in manual focus mode, and focus the camera to the closest possible distance which is .90 meters. At that point it "appears" that I am parfocal to what I view on the monitor, or what is seen looking in the other eyepiece. If I leave the camera focused on infinity, the focus is way off!
Most of my photography lately has been of taking stacked images so focus hasn't been that critical. I usually rack the objective up till everything appears out-of-focus and start shooting as I incrementally drop the lens. I then stop the sequence when everything again appears to loose focus. Before I run the stack in HF, I will usually weed out the first couple of images and likewise for the last couple. I then clean up the noise in each image using Helicon Filter or NeatImage. This seems to work alright. I know that taking pictures of living motile organisms will demand much more precision.
I think I have been unrealistic in my expectations when I now consider the limitations of my current equipment. You guys have set the bar pretty high. At some point I think I need to be satisfied that I am getting the best possible results for what I have. I am learning while having a lot of fun. I suppose that is what it is really all about. And what better excuse than to continue collecting bigger and better 'scopes. :)
George

Locked