SPIE FTW

November 23, 2009 1:15 PM

Fourier Optics Filter: Imaging the iphone camera point spread function- possible?

I'm working on a presentation for my optics class on diffraction effects in photography, and one of the things that I've become interested in is seeing if I can characterize my lenses through their psf (because I'm too cheap and lazy to set up a s-h system). It seems to me that it should be possible under optimal conditions to image the airy disk of a point source over a few pixels, and my quick calcs suggest this too, but so far have been unable to do so. If anyone has done this I'd love some suggestions.

BTW- I am not an optics student, so I might be making some simple mistakes- I have a laser with a pinhole set up, I can add ND filters if I need to (maybe I am getting exposure problems) and I can put my camera up to 30 feet away from the source- but I don't think I need to be that far away to get this to work.

posted by Large Marge to Science & Nature (3 answers total)

Update- I think it may be impossible because I cannot control focus and exposure
posted by Large Marge at 2:50 PM on November 23, 2009

I'm going to bet that the iphone doesn't have good enough optics for an Airy pattern to show up. It'll probably get washed out by a bunch of other aberrations. The other question is whether the pixels are sized appropriately to image at the diffraction limit. My guess is no. You can still measure the PSF by imaging the pinhole (just make it subtend as small of an area as possible), but you're going to see lots of detector noise and jpeg artifacts in the process too.

The more common method for characterizing photographic lenses is to use image a test card with various sorts of simple features on them, then compute the modulation transfer function. It's a more Fourier-based way of doing things. Have a look at dpreview.com; their lens reviews are pretty technical. I think the jpeg compression might present a big problem though.
posted by kiltedtaco at 4:08 PM on November 23, 2009

Correct me if I'm wrong, but if you want the PSF of the system, shouldn't you be putting a plane wave into it? Try recollimating the laser after you focus it through the pinhole, rather than putting a diverging beam into it. (Airy disk of a point source works for astronomical observations because the stars are far enough away that the spherical waves are essentially planar at the telescope aperture. Over the distances you're talking about, that won't be a good assumption.)

Are you saturating your detector? If not, I'd doubt exposure is a problem.
posted by Upton O'Good at 8:07 PM on November 26, 2009