Some scientists here want to measure the iridescence of a bunch of specimens. They might want me to build some for them, and I want to see if anyone has any cool designs.
Attached is a paper about the spectroscope
Here’s a paper on measuring iridescence https://royalsocietypublishing.org/doi/10.1098/rsfs.2018.0049
It seems to be that the phenomenon is spectrally quite pure, so I’m not sure how much pertinent info a spectrogram would contain.
Edit: I have misunderstood, ignore me
A while ago, I got fascinated with taking light-field imagery of iridescent things, and some of those experiments might be relevant to your project. Photographing light fields of a still subject is pretty easy – you put the camera on a 1-axis or 2-axis stage and move it around, taking photographs from different angles as you move it, like this camera I built to film light fields videos of a blooming flower. Depending on the angle between the light source, subject and the camera, the colors change. Take a bunch of photos, align the subject across all the photos, and you have a good representation of how the subject looks at different angles.
@Andy, in the paper you shared, they’re not really looking at angular dependence – the spectrophotometer is just pointing straight down at the butterfly wing, and I assume the light comes from the spectrophotometer. I feel like that data is a little shaky – we know that iridescence has an angular dependence, but they’re not controlling the angle of the butterfly’s wing. If you sample at a differently-tiled part of the wing, you’ll get a different reflectance curve.
In Harold’s paper, they call this movable stage a ‘goniometer’, which sounds much worse than it is – think of a spherical stage that makes a camera orbit its subject, always pointing to the center of the sphere. A super-fancy goniometer has two spherical stages, to independently change the position of the light source and the camera.
I was in it purely for the aesthetics and the weird optical challenges, and didn’t try to take any quantitative data. I feel like the data you want to get is a set of normalized reflectance/wavelength curves at many different angles. Here are a couple easy ways that you could get data out:
- Put the subject on a black background. Light the scene from a point source far away (like a bright halogen bulb several meters away). Move the camera around with the stage and take photos at every location. The camera exposure settings need to be fixed, so you can normalize the reflectance data across all the images. Write an openCV script to dump out a spectrogram for each photo, and calculate the angle between the light source and the camera for each photo. Boom! Data!
- Same thing as (1), but use a spectrophotometer instead of a camera.
- Instead of moving the camera/spectrophotometer, use a fixed camera and a grid of light sources so that you can adjust the angle by turning on a different light.
- Paul Debevec’s team made something they call a ‘light stage’, which is a super high-speed goniometer with a ton of cameras and lights but no moving parts. They use it to take ultra-realistic textures for 3D representation from every angle – used in fancy hollywood productions and cool research. It’s wayyyy more expensive and complicated to build than the moving stage, but it looks extremely cool. See attached photo.
*thinking about Andy’s paper, I think it also matters a lot what your subject is. If you’re trying to measure the iridescence of a wing, and you want actual, repeatable data, I think you need to control the mounting of the wing – like, use spray adhesive and hold it flat on a piece of paper, in a repeatable orientation. It’s tricky, because the phenomenon you’re measuring is angle dependent, but if you have, say, a curved surface like beetle elytra, there are lots and lots of different angles there – which one are you measuring? Butterfly wings are uniquely nice to measure here because they’re planar.
–my $.02
RIGHT!!?! okay, That’s what I felt like I was confused about this thing. Like if somebody asked me to measure iridescence, I would try to do something like what you described or maybe have a fixed camera with an array of lights.
It turns out the butterfly scientists already have that spectrophotometer thing and they really just want me to build this little rig. But again I kind of feel like the science is sort of weird. I look more into what you wrote in a second!