Sat, 11/21/2020 - 20:59
I'm posting here in response to my issue resulting from https://aavso.org/possibility-vsx-utilization-during-platesolving-analy…. When I upload an image to V-Phot and open it in the editor, it is entirely white. I can hover over it with the mouse and it does show coordinates, but the image itself shows nothing.
I've tried varying the display settings (by default were from 0 - about 65000) to no avail. What could be the issue here?
I'm hesitant to upload the .fits file directly as this is a public forum and the header contains very exact location information, but I can send over E-mail if there is no obvious solution from this problem description.
I've had this exact problem before, don't know why. It went away when I played around with the display tool long enough.
For starters, try Low 25, High 450. Don't forget to click Apply.
sadly that didn't change it, it's still entirely white. After I attempted to play around with the values more, VPHOT even crashed. Then after reloading it varying the values sadly still didn't change the white image.
The image loads perfectly in AstroImageJ, DS9, and Pixinsight (as well as on astrometry.net, the source).
Does this happen only for one particular image? Have you tried other images?
I tried several final images which have had bias, flat, and dark frames applied. These all turned out white.
However I now tried a raw image, and the image did load for the most part. There are still three white squares in the image when viewing it in VPHOT, but the vast majority of the image loads. If I platesolve this image using astrometry.net, the entire image loads.
I'm not sure what the issue with the calibrated image is or why three white squares appeared in the raw one, but fortunately after some playing around with images, I'm able to use VPHOT to access VSX.
I've seen a white display, but usually when the image has been scaled to very low values (like from 0 to 1). The fact that you can display a raw image goes along with that assumption. What program are you using to do the dark subtraction and flat fielding?
I am using Pixinsight, considering its very wide-spread use and price I don't think it would have an internal issue. AstroImageJ, the fits viewing software DS9, ASTAP, and astrometry.net can all open the image without problem after testing, but then in VPhot it is entirely white.
Pixinsight is a great tool for deep-sky imaging, but terrible for photometric use. It does rescale the image to between 0 and 1, which causes problems with VPHOT, which does its image display in integer units and not floating point. You will need to rescale to something more reasonable (0-65535). Most other image analysis programs like AIJ and DS9 handle image buffers in floating point, so they can still work with the 0-1 range.
You are not the only person who has run into this issue. We need a FAQ for VPHOT!
I'm not sure about this. I use Pixinsight for calibration frames, cropping, and star alignment. In these it is a state-of-the-art software (as these overlap with deep sky imaging), especially in star alignment, able to fully utilize all CPU cores and produce results I have not been able to match with other software.
Of course it cannot be used for the photometry itself, but it being a "modern" software with regular updates, it can handle my nightly 70 to 130GB file sizes with the CMOS camera which makes it (in my experience/from what I've read) my best choice for processing.
I will do some research on the 0-1 scaling issue. When saving Pixinsight offers an entire page full of format & file options, so I'd expect there is some way to save directly to integer units. If I find a solution I'll edit the original post here in hopes of it being useful to someone else - I agree a FAQ would be useful .
"Pixinsight is a great tool for deep-sky imaging, but terrible for photometric use."
I can confirm Arne's comment about Pixinsight. It has caused much Weeping and Gnashing of Teeth for students in the photometry courses.