shooting through water? (was: image stacking and stitching?)

Drbluethumb · Post by **Drbluethumb** » Thu Feb 16, 2012 3:13 pm

Hi

I have some question's about this photo that i recently took:

It's a image stacked recordia yuma coral from my aquarium, the size of a ping pong ball.

It was taken with a EOS Rebel T3, and a 50mm 1.8 len's, with a 12mm extension tube.

first question is.

Can I get a better/more detailed pic, if I were to use a slightly larger extension tube, then image stack the top left, then do the same to the top right, then went to the bottom left of the Yuma, to do the same, then finally to the bottom right side.

For a total of 4 final pictures that are imaged stacked. from there I would stitch the 4 images together

Would this work? would the images blend together right, if done like this?

magus424 · Post by **magus424** » Thu Feb 16, 2012 3:18 pm

I think you'll run into parallax issues if you try to do that unless you figure out how to rotate the lens just right

Drbluethumb · Post by **Drbluethumb** » Thu Feb 16, 2012 3:38 pm

magus424 wrote:I think you'll run into parallax issues if you try to do that unless you figure out how to rotate the lens just right

that's what I was kind of thinking.

I can get the subject close to the glass, it has lot's of light in the aquarium so lighting won't be an issue, but I want to get some more detail, and potentially fill the image with the subject.

How would I go about doing this?

can I do this with out buying a new len's?

any suggestion's on macro rail's, special head/ attachment's for the tripod, diy?

Cheaper preferred, would love to hear a way, if any, with out spending $1000.

I also have a playthoas coral, I would like to photograph, it is the size of a penny.

rjlittlefield · Post by **rjlittlefield** » Thu Feb 16, 2012 5:44 pm

First, let me mention one of our standard references, What is "stack-and-stitch", and how can I do it?.

I'm not sure the FAQ makes clear that stitching is a method for getting more sharp pixels than your camera sensor can provide in one shot. Stitching stacked images is not easy, so it should be considered as a last resort only after you've exhausted all other methods of getting enough sharpness in a single stack.

So, let's go back to the beginning and talk about your current stack.

The image that's posted looks to me to be reasonably but not super sharp. That makes me wonder just how much detail there is down at 100% pixels.

Shooting through thick plastic/glass/water is notoriously difficult because the extra material introduces "spherical aberration" that tends to soften the image. This problem becomes more severe with increasing magnification. It's no big deal if you're shooting a whole school of herring, but microscopists sometimes have to worry about an extra 0.01 mm of water degrading their image. You're working in a size range where shooting through plastic/glass/water is a concern, and until that concern is resolved I think there's no point in struggling with stitching.

Can you post out a 100% pixels crop so we can see what's down there to work with?

You mention "potentially fill the image with the subject". There are several ways to do that. Since you're using extension tubes already, why not just add a little more extension?

--Rik

Peter De Smidt · Post by **Peter De Smidt** » Thu Feb 16, 2012 5:53 pm

How long of an exposure are you using? The currents in a reef tank can be pretty intense. If you're not using flash from above or the side, you may have to switch off all of your pumps, and that means all power heads, return pumps, closed loops pumps...

Make sure to shoot perpendicular through the glass if possible.

Another possibility for stitching is to keep the lens in place and shift the body of your camera. For instance, you can attach your camera body to the back of a view camera and use the view cameras movements to move the camera body from side to side and up and down. Lenses gain coverage as you do closeups, and you should probably be ok.

See for instance: http://www.luminous-landscape.com/revie ... n_LD.shtml

Chris S. · Post by **Chris S.** » Thu Feb 16, 2012 10:32 pm

rjlittlefield wrote:You're working in a size range where shooting through plastic/glass/water is a concern, and until that concern is resolved I think there's no point in struggling with stitching.

I have some personal experience very much in line with Rik's statement. A few years ago I had a project to document freshwater pond weeds, some with structures of interest in the size range discussed here. Since they can assume unnatural positions when out of water, I wanted to photograph them floating, and so brought the subjects inside to an aquarium. This provided an opportunity to experiment with varied approaches; I ended up accumulating a couple of thousand test shots with different angles, lenses, filters, lighting regimes, etc.

One takeaway was that without exception, images shot through aquarium glass had a notable lack of sharpness and detail. Only by shooting down through the open top of the aquarium (through water only--no glass) did I get crisp, detailed pictures. Shooting through an air-water interface does bring its own set of challenges, but in my situation, these could be dealt with, whereas the degradation from aquarium glass could not be.

My gut feel, without knowing additional particulars for your posted image, is that you are probably at or near the resolution limit achievable through aquarium glass. Your image is lovely, in terms of lighting, color, just the right depth of field, and subject interest--it's just lacking in ultimate detail. Your limiting factor at this point is almost surely the aquarium glass. Can you find a way to avoid having to shoot through it?

--Chris

Harold Gough · Post by **Harold Gough** » Fri Feb 17, 2012 1:35 am

Not an answer for this application but, thin, non-reflective glass used for framed prints has some potential here. For close-ups and moderate macro an aquariun with at least one side of such glass could be constructed. You might want more than one, such that creatures of various shapes and sizes might be confined close to the glass.

Harold

Cactusdave · Post by **Cactusdave** » Fri Feb 17, 2012 2:22 am

I don't disagree with all the limiting technical issues raised here. However my first instinct was that this image didn't look too bad. I don't know how much you had done on the image, but I think a few Photoshop tweaks including Unsharp Mask would effect considerable improvement. I wonder if something could be done with the lighting and colour balance too. There's a lot of blue there which looks too strong. I don't know whether that's aquarium lights or something you added. To me the main subject needs to be spot lit like a performer on stage and I don't get that feel. I have seen some very nice shots through aquarium glass, and I'm sure you could get good results provided attention is given to the obvious, like distance of subject to the glass, cleanliness of the glass, reflection at the air/glass interface and to directional lighting.

Peter De Smidt · Post by **Peter De Smidt** » Fri Feb 17, 2012 3:58 am

The blue light is the aquarium lighting. Many use strong "actinic" lights to bring out the colors of the corals. Recently, this involves a lot of super blue LEDs.

Cactusdave · Post by **Cactusdave** » Fri Feb 17, 2012 7:35 am

I contacted a semi-pro photographer friend who I knew had done some aquarium photography for a local museum to publicise their new reef exhibit. He said he had them transfer subjects to a 'quarantine' aquarium behind the scenes where he had more control over lighting and access than the big exhibition tank. He said for most shots he used a view box type arrangement similar to the one about half way down on the link page, with a remote release operated by an assistant as they weren't keen on him putting his hands in the aquarium water. He said he got the idea from here http://www.advancedaquarist.com/2008/5/aafeature2 He said used top lighting with ring flash and a couple of LED array lamps for some shots. He reckoned it might be possible to stack using something like this with a vertical rail and a bit of ingenuity. He said he tried photography through the side glass and the results weren't bad if top lighting was used, the subject was near the glass and didn't move and great care was taken with reflections. He did all his photography in a darkened room apart from the lights he had control over to reduce reflection issues. I hope this helps.

ChrisR · Post by **ChrisR** » Fri Feb 17, 2012 8:09 am

?? surface silvered mirror, submerged, and camera above the tank?

You'd want a fairly long-focal length lens, depending on the size of your tank and subjects of course. Perhaps an enlarger lens, around 135-180mm? They're pretty good and not the price they once were.

Perhaps you could try a tube with a UV filter on the front, dunked into the water, in front of the lens, to give you a good quality clean glass-water interface. Like holding a snorkel mask as a "window" into the depths.

Cactusdave · Post by **Cactusdave** » Fri Feb 17, 2012 8:53 am

Chris , that last suggestion is pretty much one of the solutions in the link I posted above, and similar to the method I gather my friend used successfully.

Chris S. · Post by **Chris S.** » Sat Feb 18, 2012 12:36 am

Cactusdave wrote:I'm sure you could get good results provided attention is given to the obvious, like distance of subject to the glass, cleanliness of the glass, reflection at the air/glass interface and to directional lighting.

Hey, Dave, maybe we should think on how photographers differ in their sense of the word "good"?

Images of the quality in the thread you linked to would not have fulfilled my client obligations in the freshwater pond-weed study, nor my own standards. But they might well fill the needs of another photographer.

In this thread, the OP seems to have taken a "good" shot, and asked how to make it into a "better" shot. It would appear that his or her standards are pretty high. My sense is that these standards will not be met until spherical aberration is sharply reduced. As far as my testing has shown, this can not be done without the avoidance of shooting through aquarium glass.

I'll mention that in my tests, the aquarium and water were both very clean; angles were carefully considered and tested; lighting, distances, and polarization were likewise considered and tested with great care. Room light was controlled.

My conclusion: If you are picky, don't shoot through aquarium glass at these magnifications.

Very best regards, mate,

--Chris

Peter De Smidt · Post by **Peter De Smidt** » Sat Feb 18, 2012 3:31 pm

See: http://i955.photobucket.com/albums/ae37 ... cordia.jpg

rjlittlefield · Post by **rjlittlefield** » Sat Feb 18, 2012 3:42 pm

Peter, I admin-edited your posting to change your version of Drbluethumb's image from inline image to a link. This is in keeping with photomacrography.net's rules about reposting other people's images without prior permission. (Things stay calmer that way.)

I gather that your mods to the image were intended to illustrate what could be done in post-processing. Could you explain in more detail?

--Rik