Affiliation
American Association of Variable Star Observers (AAVSO)
Sun, 12/05/2021 - 06:30
I recently started imaging with a CMOS camera. My file size is over 100 mb. If I zip 2 files it is over 100 mb. I can zip 1 file and get it just under 100 mb. So I can upload 1 zipped file at a time. For mediocre internet speeds this can take a while with multiple files. Are there any plans to increase the 100 mb file limit?
Barbara
Barbara,
If you can bin 2x2 your files should be reduced to one quarter of their unbinned size.
Roy
Thanks Roy. That will leave me significantly undersampled. I can defocus to compensate a little but wonder how much the undersampling will effect my photometry. I guess I will test it and run photometry on a 1x1 then a 2x2 binned image and compare results.
Barbara
This is one of the restrictions on a web-based photometry system. File upload time is long; file storage gets expensive. Many of the nice aspects of VPHOT are also available in LesvePhot, which is something you might try.
So this camera is not for your 16 inch?
Arne
Arne,
That is correct. It is not for the 16”. It is for the 80mm refractor on top of the 16”. I replaced a DSLR with an ASI2600MM Pro+LRGB filters. The set up was mainly for pretty pics but I decided to calculate transformation coefficients for the RGB filters to do photometry since my Proline camera has been at FLI for repairs for a month now.
Barbara
Hi Barbara,
The ASI2600mm-pro is a 26MPIx camera. If you are generating 100MB files, then you are saving your images in floating point. Have you tried to limit the pixel type to 16-bit unsigned integer?
Arne
That is exactly it. I normally save in fits 16 bit unsigned. I guess setting up a new configuration for this camera in Maximdl it somehow got changed. Changing the format to 16 bit unsigned brings the file size down to 50 mb.
Barbara
Barbara,
CMOS cameras like mine bin using software. For 2x2 binning the signals from 4 pixels are averaged. I've not thought about it before because all my images are oversampled, but if you did this, your original image sampling would not be affected. The downside is that a pixel could be saturated but the average value might not show this. Therefore checking for saturated pixels must be done on an unbinned image.
Roy
Hy, i use a cmos camera with the a resolution of 8266 x 5644 and Bining 1x1. Each image is near 100 Mb. I take the images and then i use macro function in astroimagej to cut the image with a the resolution 4000 x 4000. After this the image has a size of 31 Mb. Then i Zip this file and have a size of 7 Mb. Prhaps this is a way for you?
Volker
Volker,
That is a reasonable option since most of the time I don’t require the full resolution of 6248x4176.
Barbara
I have a 2600mm too. I…
Hi,
I have a 2600mm too. I also bin (in ASTAP software) to be able to send to VPHOT. Even that is not always a fluent operation.
But, I am using the Python Astropy and Photutils modules to write a python program myself. I do use also the formulas of AstroImageJ.
My program works very well and super accurate, also there are a lot less misses as in VPHOT or other photometry software I have been looking at. To be able to be way more accurate I did write some code to inspect the annulus, to position super accurately the annulus and so on. Also it does analyse all the variables in the field in one go, no need to run several times the software (as in VPHOT)
But, my program is not yet distributable to other users. Need some weeks/months to make it more userfriendly to avoid other users running away screaming like hell. I do have a GUI already (using pyside6 and QT), but there is still some work to do. I do have too much ideas for extra modules also...
It all works on the computer itself, no internet or whatever necessay. Also no pinpoint software necessary.
More importantly;
If I do NOT bin, my SNR is way way better than when I bin. I can go much lower in magnitude and the error bars are better
Small rectification (after I've got a message via "contact form", which I do not find or know what it is, sorry):
Indeed, with VPHOT you only have to run once the cycle to analyse multiple variables, with LesvePhotography it is not the case. AstroImageJ has its own problems. I've played with all of them and try to synthesise their strong points and avoid their a bit weaker points.
Attention : all three are GREAT and WONDERFULL programs, but ...
For me there are some rather less ok points with VPHOT, besides being very great software, don't let me be misunderstood, I do not want to offend anybody.
1. The upload is very cumbersome. I do have a very powerfull internet connection (Belgium), but I've been eating shoes when waiting and waiting and then realising it is just the connection that fell away
2. I do have to bin the photo's to be able to upload them, with as consequence a much less good SNR. Binning is done for the moment via ASTAP, no idea if that it the reason of the much lower SNR.
3. When analysing a timeseries there are almost always several mishits,caused by the software not exacly positioning the annulus on the star or comparison stars. I've been studiying it, but did not find a solution.
4. Those not catch/excepted error messages.
5. Sometimes an uploaded photo is not accepted without getting a reason why.
Another reason that I am making my own software, is that it is just a lot of fun.
Also, I want to integrate - later - plate solving and calibration, and period analysis and visualisation while observing, and...
So that all what I need is in one software and not in a multitude of other things. But it's just a project. No idea if I ever will do an 'official' release.