Bug head shots

November 13, 2011 8:32 PM

Paging macro photographers. I need help taking great photographs of tiny bugs (sub-questions on eye-fi, focus stacking, and lighting).

I’m about to begin a fairly intensive entomology experiment. Here is the relevant information regarding my setup:

For the next several weeks, I will be spending alternate days photographing leaves on tiny bean plants (plants are 20 cm high), each with two leaves (clipped to 5 x 4 cm to keep surface area constant). Each leaf will have tiny plant mites (both on the top and underside) in various combinations of eggs, larvae, and adults and since they reproduce fairly quickly (2-3 days per generation), I need to photograph them every other day (so I can later count the number in each category) for two weeks.

Since I can’t count directly, I have to photograph them (front and back of each leaf), really close, and then count later.


My setup
Hardware:

• I have a Nikon D5000.
• 60 mm macro lens
• A really nice, heavy Slik tripod
• Two reflectors with these LCD lights. (Initially I had Tungsten lights which although brighter were too hot).
• A remote
• A eye-fi card
• A macbook pro
• And a wireless router (not connected to the internet).

Software:

• Photoshop CS 5.1
• Eye-fi proprietary software
• Have a budget for other software if you have recommendations

Questions:
1. I need to get focused pictures taken up close so I can count all of my bugs later on. From previous experience, it takes at least 4 pictures to cover one leaf surface (so for one plant that would be 16 pictures minimum). I have 40 plants to do so time adds up. I’d like to be efficient and accurate at the same time. A Mefite told me about Focus Stacking. Ideally I’d like to take more than 4 (overlapping pictures but it would be faster since I can quickly take many), and then have them merged using CS5.

Can I move the camera when attempting focus stacking? I mean, if I focus on quadrant 1, and then want to focus on quadrant 4, I can’t do it without moving the camera. My trials seem to work sometimes but different parts of the leaf have different exposures.

Focus stacking blogs suggest keeping exposure the same. I’m using a smallish aperture (5.6 to 8 because smaller makes pictures blurry) to get greater depth of focus (my bugs spin a web so they can be at different heights on the leaf) but the exposure time is not always the same. Also, it gets difficult to maintain the same focal distance for the leaf underside (because the leaf stalk is in the middle). How do I avoid that?

2. I bought an eye-fi card with the express idea that I as I take pictures they would appear on my Macbook Pro’s large screen so my assistant can determine if they are in focus or not. That way we can reshoot right away (if we fuck up a plant and don't reshoot same day, we're screwed). The point behind this is that the picture often looks good on the camera LCD but awful at full size.
The proprietary eye-fi software is quite awful (only shows a thumbnail on the screen). Is there something better that will show a full screen picture on my mac?

3. Worst case should I just revert to full auto-focus and close off the flash? On the pro side, it would save time but not sure if it would allow me to assemble a full leaf picture from multiple images with different settings.

posted by babbyʼ); Drop table users; -- to Media & Arts (3 answers total)

I use the Aperture watch folder script and Aperture to show the full screen pictures as they come in to the computer from the eye-fi. I wish there was a better solution because it can be slow at times.
posted by Infernarl at 10:13 PM on November 13, 2011

I don't see how you could successfully focus-stack and move the camera between shots - at a very basic level, it works by "slicing" out the areas fore & aft of the focus zone & stitching them back in from the other shots done with different focus points. To do that and move the camera would require some non-linear image convolution, even if you can keep the lens the same distance from your object plane.

This confuses me: "I’m using a smallish aperture (5.6 to 8 because smaller makes pictures blurry)"

With that lens you'll need a much narrower aperture - I'm guessing F11, or probably F16 or F22 - to get sufficient depth of field for counting insects that build 3D structures without focus stacking. That, of course, necessitates either longer exposures or more light. If you're getting blurring beyond about F8, I'm guessing the camera, leaf, or insects are moving too much to be 'frozen' by the long-ish exposure.

• You'll need to clamp the camera solidly - most tripods won't be good enough; you'll need a pedestal or bench-clamp; preferably with a focussing rack for manual focus.
• You'll have to clamp the leaf properly - ideally, using a jig to carefully hold the leaf flat, immobile, and parallel to the camera's focal plane.
• You'll need an exposure short enough to account for insect movement - which may mean upping the amount of light.

I'm curious though - is this out in the field, or in the lab? Normally, for this sort of thing you'd use a camera mounted on a low-power microscope. They've generally got better depth of field too…

I suspect in the end it'll come down to you having to hold everything solidly, flat, and parallel; using a narrower aperture to get enough DoF; and throwing as much light as you can on the leaf while shooting.

(I won't comment on your experimental design, as I don't know what you're trying to achieve. But, as someone who himself built an experiment that's ending up requiring a lot more work than expected, I'd be looking to redesign it as much as possible to minimise the work involved. The first thing I'd be looking at is sampling rather than censusing each 5 x 4cm leaf area.

Edge effects from sampling are easier to deal with than boundary effects from a strictly-defined area, unless you can argue a good reason that the boundary effects are realistic ;-)
posted by Pinback at 10:43 PM on November 13, 2011

Best answer: For something like this I wouldn't use a DSLR. I often shoot with my DSLR and 105mm macro lense, but in this case I don't think it's the best tool. You can get far, far, far greater depth of field with a superzoom point and shoot and some Raynox lenses. I use an old Panasonic FZ8 camera and The Raynox DCR-150 and DCR-250 to take my macros. Results here.

You can probably just use the DCR-150 for what you want to do. The awesome depth of field will likely cut down on the number of photos you have to take dramatically.


You could probably duplicate my setup using ebay for $100-200.
posted by sanka at 6:10 AM on November 14, 2011 [1 favorite]