Has anyone tried bright star photometry using images captured by an iPhone (or another smart phone)? These cameras are quite capable these days, and while not a first choice for science, it might be a good educational tool.
1 Like
Smartphone images are hardly postprocessed, so reliability of photometry needs testing. Calibration may be a problem too.
However, there are info on all-sky cameras using for the brightest stars photometry. Probably, modern smartphones may bring even better results.
I myself used smartphone just for recording of images got with planetary camera (and tried to get broad skyscapes with phone minicamera for fan).
I think, smartphones may be useful for meteors monitoring. Well as there are some articles about UFO monitoring with smartphones.
@bjs , welcome to the AAVSO forums!
Most smartphone cameras are either 12-bit or 14-bit sensors under the hood. The raw sensor data will have very similar characteristics to what one would get from DSLR cameras or small smart telescopes. I authored a paper with several AAVSO members about a decade ago which showed that 50 mm DSLR cameras can provide 0.010 - 0.100 magnitude photometry over the V~4-9 range with 10-second exposures. Given that smartphones have much smaller apertures, I’d guess they could deliver similar performance, but over a V~1-6 magnitude range.
With that said, it is very important that you get access to the raw sensor data. You’ll need a special application for this. I’ve heard good things about Open Camera for Android, but have never used it myself. Once you have the data, follow the reduction instructions in the DSLR Photometry Manual.
Please note that the JPEG images you get from smart phones are not very well suited to doing photometry. They are compressed to 8-bits which only provides 256 levels of gray in each color channel. While you can do photometry with this data, you have to be much more careful about comparison star selection.
Brian
1 Like
Anusha Mehta presented a nice poster about smartphone photometry at the most recent Annual Meeting. Here’s the corresponding abstract:
https://www.aavso.org/113/annual-meeting-abstracts#Mehta
Personally, my main concern is that there may be destructive post-processing which is applied even to the raw images, as in the infamous Sony “Star Eater” issue (where a noise-reduction routine within the camera’s firmware caused stars to disappear from RAW images).
Today’s smartphone cameras rely heavily on computational post-processing to produce images which look good despite their optical limitations, so it wouldn’t surprise me if some phones have similar firmware. But that’s all the more reason to start experimenting and testing cameras! You won’t know until you try.
Good luck! Smartphone photometry is a great idea, and I’d love to see someone come up with a method or app which makes it easier for beginners.
1 Like
Hi Brian,
I’ve seen your paper, and some of the others. I’ve just scratched the surface of the tech - I need to start playing and see what happens. There are a couple of iPhone apps as well, though I haven’t tried them yet.
Apple’s ProRAW images are 12-bit, which helps the dynamic range issue. I’ve seen a couple of cookbook articles on separating out the R,G, and B channels. Haven’t tried it yet, since it’s just a though experiment so far. I did figure it would be pretty similar to the DSLR stuff.
Thanks for the tips!
Interesting, and exactly where I wanted to go. I don’t suppose anyone has contact info for Anusha Mehta?
I’m certain your concerns about post-processing are real. I’ve experimented with a variety of Sony sensors connected through a CSI bus. This is an identical setup to what one would find in a modern smartphone. I’ve always had to explicitly set registers on the device to turn off a variety of post-processing functions. The Star Eater is probably the best known issue, but cameras also apply other functions to the pixel data. For example, most cameras have a built-in color correction matrix that is applied to the RGGB data to address differences in spectral response between the pixels and factory expectations. This is multiplicative, so it functions like a RGGB-pixel-dependent gain and can be calibrated out. Nevertheless, I think it best to shut this off.
@bjs I agree with Lauren: The only way we’ll find out is to get out there and do some experimentation!
Brian