has anyone ever used a smart phone camera for photometry? In combination with a telescope or better even without?
My interest is not really scientific applications but rather education. I want to give a course at a local "junior university" (for kids aged 14+) and I assume most of them do not have DSLRs anymore but certainly all have 1-n smart phones...
I know there are a few apps out which allow setting relevant parameters manually incl. ISO, exposure time, raw format, focus to infinity etc. Therefore it should at least work to do some bright star photometry, maybe Betelgeuse or Mira near maximum. Errors of order 0.1 mag should be perfectly ok.
Our community needs more young enthusiasts I think.
I think that is…
I think that is an EXCELLENT idea! This could be a model for high schools and colleges everywhere.
The only issue I see, and maybe you've solved it, is getting the data out of the phone uncompressed. If you can get RAW images, there are FITS converters available.
If an eVscope can detect exoplanet milli-magnitude changes you ought to be able to detect variability in Mira and other brighter variable stars.
I hope that when you pursue this endeavour you keep us up to date on your results. And to me, it sound like an interesting paper in JAAVSO or Sky and Telescope!
Thanks to all.
I did a first "test": I took my Samsung A-52, ISO 3200, 0.4s and just hands-on shot on Altair.
Downloaded the jpeg image and seperated the green channel. I compared Altair to Tarazad: V difference is 1.95, the "measurement" is 1.5. Not good but not miles away.
Worth doing a bit more, in particular using raw-type format and longer exposure with a tripod.
And no, I would never report to AUD. Purely educational.
JPEGs are both lossy and nonlinear, so the fact that you did not get the correct ratio is not a surprise.
Find out how you phone saves raw images. If it's DNG = Adobe Digital Negative then you're home free. It could be 12-bit or even 14-bit data, You should be able to convert this to FITS for analysis in whatever software you use for FITS files.
A simple test would be to shoot a sequence of images of different stars using the same settings. Center each star in the frame, then do photometry on the stars. Plot your derived magnitudes against the catalog magnitudes. To do it right, you should correct for extinction. You should find a reasonably straight line fit.
You can fit unknow stars into the fit and get the magnitude out.
This is a good method for education so long as the students have reasonably advanced phones that support a "pro" mode and a raw format.
I too am very interested in this possibility. Here in a rural community of San Diego County some of us are trying to contain the advance of light pollution in the vicinity of the Palomar Observatory. Involving students in our local high school in the science of photometry using their smart phones might assist us in this effort. I just looked and my new Samsung Galaxy S22+ has a setting for JPEG and raw formats.
Please keep us updated on your progress.
I tested my Samsung Galaxy S9+ to see whether photometry was possible with it, and the answer was "Yes." This was a couple years ago, and it occurred to me that an aperture of a millimeter or so would be right for measuring Betelgeuse. I'm not sure where a saved the data and analysis on my computer, but I will look for it.
The Galaxy S9+ has a "Pro" mode that saves raw image in 12-bit DNG format. The DNG format is so close to TIFF that AIP4Win loads the images as if they were TIFF files, so there were no problems treating the images as 12-bit science images. AIP4Win was formerly bundled with The Handbook of Astronomical Image Processing but is now available for free from firstname.lastname@example.org.
Cellphone images are encoded with a Bayer array, but they have an IR-blocking filter. I found the overall color sensitivity could be described as "wide-V" or "similar to VIS." This probably varies from phone to phone, but if you're interested in an observing program such as Algol, this would not matter. Unfortunately, for measuring the magnitude of the very red star Betelgeuse in a field where the available comp stars are blue, it was possible to determine that Betelgeuse had varied, but not to get a trustworthy V magnitude. However, for observing anything but very red stars, it's well worth doing some experiments to see.
You can couple a cellphone to any optical system that has no obscuration of the aperture (i.e., refractors = yes, Newtonians = no.). The lens of the cellphone must be centered in the exit pupil (i.e, the Ramsden disk.) This shouldn't be hard to do, but you'd want a jig or some sort to hold the phone at the right location. I did not try photometry through any optics, but it should work. A 'scope like a 7x50 monocular would be pretty good.
Richard suggested low power optics such as a 7x50 monocular.
Another option would be one of the commercially available lens kits for smart phones.
I tried to make some photos with Mac 150/1800 and smartphone. Magnification 42x, ISO 3200 and 10 seconds gave about 11m. Here is Z UMa, for example: https://nova.astrometry.net/user_images/6451959#annotated
That's a good detection. I'd try and repeat that a few times and look for how consistent your results are. But a heck of a start.
Did you "debayer" the original image and only use the green pixels or just use them all. I can see reasons for both!
Dear colleagues, I am absolute beginner in variable star observations and I kindly ask you to check my first results. I tried to make some photos with smartphone (regretfully Aliexpress does not send ordered camera ASI432MM during two months). RAW shots have been stacked, darks substracted and green channel extracted. Pictures are not applicable for real photometry, that’s why I used calibration stars from AAVSO maps and called this strange method “smartphone assisted visual observations”.
Find please my pictures there and compare please evaluations with [AAVSO data]:
BG Cyg 11.6CV, 10.3CR [11.5-11.7V]
FL Tau 13.7CV [12.8CR-13.3V old data]
V815 Tau 13.2CV [12.3-12.7V old data]
RS UMa 13.5CR, 13.9CV [<12.6-13.2Vis 2022.09]
RY UMa too bright [7.3-7.7Vis], not used
X Cam 9.8CV [8.6-10.2Vis]
(Use plese full size pictures with stretched histogram, because previews are too dark).
I uploaded first results to database as “Vis” with needed comment, but if experienced observers will reject them – I do not insist :)
I will be glad to get critical notes and advises (and I hope to get real camera finally) :)))
The R Cyg seems to be a good estimate. It is brightening, and was 10.9 on 11/10. RS UMA looks good. T UMA also right in there.
I'd say your system is pretty good!