19 June 2021: Our upload server https://upload.astropixelprocessor.com/ has been migrated successfully to our new office with higher upload and download speeds (nearly 10MByte/sec up/down ) ! We now have 1 general upload user called: upload with password: upload. The users upload1 - upload5 have been disabled.
31 May 2021: APP 1.083-beta2 has been released ! APP 1.083 stable will follow soon afterwards. It includes a completely new Star Reducer Tool, New File Saver Module, Improved Comet registration and much more, check the release notes here!
Master Flat Warning Message
Today I am attempting a quick integration of a 4 Pane mosaic shot with an ASI1600 camera and Ha filter. I also shot flats with the same camera. After Calibration, as Star Analysis begins, I get the following warning message:
Doubting my sanity, I opened the master flat in Pixinsight and it is definitely a single channel greyscale image. The warning persists through the remainder of the integration process.
In Tab 0 I have tried both the Ha algorithm and Adaptive Airy Disc, but to no avail. I have integrated plenty of Ha images before, but have never seen this - am I overlooking something obvious?
I am running version 1.082
Thanks for any advice!
@jonesdee Can you open the master flat in APP and take a screenshot of the APP window including the histogram to the upper right? Also, for our reference, I assume that you have an ASI1600MM camera and not an ASI1600MC camera? I am sure you know but ZWO MM cameras are mono and MC cameras are color. Just summarizing all the obvious bits here in case we need them for future reference.
Here is the screenshot with the histogram, confirming in the filename that it is indeed an MM-Pro camera.
I am surprised at how much clipping has taken place in the histogram - is this normal?
@jonesdee Thanks. It indeed looks like the master flat is mono. Would you mind uploading a few lights and the master flat to
Using upload for username and upload for password? Create a directory called jonesdee-master-flat and place the files there. Let me know please when the files have been uploaded and I will have a look.
Also, clipping to the left is not necessarily an issue. There are always dark pixels (the small peak all the way to the left) which have values at or very close to zero. Clipping to the right would have been an issue since that would indicate saturated pixels.
Hi Wouter, all done. I included a couple of flats for good measure.
I have never checked the histogram for flats/masters, as they have always worked fine in the past, barring operator error which I have always managed to fix.
@jonesdee Hi Dave, thanks for the quick upload. I checked your files and the flats and master flat are color images! See these screenshots:
Note how both screenshots show that the ASI1600MM has been registered as imaging camera in the FITS header too! And that both FITS headers contain a CFA pattern or Bayer pattern keyword. The lights are B/W images. So that explains the popup in APP. The question is how those files came to be color and not mono.
Wow, that is crazy! They were taken with the same imaging setup as the lights, literally a few hours after the lights were shot. The difference is that the lights were captured using Voyager Automation software and the flats indoors with Astrophotography Tools (APT). I have always worked in this manner, but I suspect there is something in the APT setup that has saved the flats as RGB, although that has never occurred in the past!
Thanks for your help Wouter, I'll investigate and report back.
APT registers my camera as an ASI1600MM Pro Mono, as would be expected. The CFA keyword in the FITS header appears to relate to a function in APT which shows colour previews when OSC cameras are used:
Disabling this removes the keyword from the header, but APP still displays the file as a colour image when 'Force CFA' is enabled in Tab 0. This does not happen in Pixinsight with any of the flat files or the Master.
I have uploaded another flat without the header keyword - it is easily spotted as no cooling was running and the temperature shows 28* in the filename :).
@jonesdee Hi Dave, it is crazy indeed! What really amazes me is that the histogram indicates a color image as well. But this may indeed be an issue with the FITS header writing of APT. Can you use Voyager for taking flats and, if yes, what is the result of that?
I edited the header of several flats and removed the CFA reference. A quick integration with those flats, a few lights and darks worked fine - no warnings and nicely calibrated lights 🙂
Since last November I have been using my ASI2600 OSC, so there was no issue as all files were RGB - this is why I have only just noticed the problem. I expect that before last November, when I did not have an OSC, I didn't enable the debayered view in APT as there was no need.
I will drop a note to Ivo (APT) and see what he thinks.
Of course, I should be consistent and use Voyager for flats as well, but when you reach my age, old habits die hard :).
Thanks for helping with this one, I'll have another crack at my mosaic tomorrow.
@jonesdee Glad to see that that solved it. I still wonder why APP displays the histogram with a green left top part but that may be coincidence.