Too many clouds in the last days and only a small window of hope next week, so I spent the last days on re-processing old data and an interesting question (for me) popped up where I'd love to hear your thoughts:
I do narrowband imaging with an OSC Camera and over time I have now gathered data with the same scope/frame duration but with different filters. I am now thinking about the best process of combining the data.
One pre-requisite: I try to keep my image chain as dust free as possible and looking at my flats this seems to have worked out quite well over the past months. When framing I rotate the whole telescope so that the position between Camera and Scope stays the same.
For example, I now have a dataset for the Seagull nebula shot with the Antlia Ha (6h) /Oiii (3.5h)/Sii(7h) filter and when finally the ALP-T filter was available I also shot arround 4h for the same target.
So my goal is to get the best possible integration out of the dataset.
My current process is as folowing:
change the fits header for FILTER of ALP-T to match the Antlia Ha Filter
copy the ALP-T data to a new directory and change the FILTER Header to match my Antlia Oiii Filter
add all files+flats+darkflats+darks+bias to Weighted Batch Processing and let it do it's magic…
I am taking two shortcuts here, pictures from different sessions share the same flat and I ignore the flats of the ALP-T and only use the flats for the single band filters, but I hope to get away with that because of the rather dust free image chain.
The approach in AstroPixelProcessor with it's sessions looks like overall better suited as there I can have sessions with flats for each session and I can tell it that it should only process Ha (or Oiii). But to be honest, I am not a too big fan of APP and try to avoid it if I can as the Weighted Batch Processor Script is evolving with every PixInsight release and APP seems kind of stuck in development, although for this case it looks as a better fit…..
As another process I thought about running Weighted Batch processor several times and then using the files from the debayered folder (which should be propery calibrated) as a new input for a 2nd run of weighted batch processor that takes all the debayered images that were preprocessed in a 'Session' way.
But to be able to do so I there also have to tweak the fits data, likely I need to remove the bayer matrix entry and also here will have to tweak the FILTER keyword.
As it is not so easy for me to modify fits headers in xisf files this will first need some experimentation on how to best automate the process.
Any thoughts on how the whole process can be done more efficient (or better) in Weighted Batch Processor? I still plan to keep the process of shooting both ALP-T and the single band filters on the same target, in no-so bright-moon nights the ALP-T allows me to get Oiii Integration time for free and when the moon is rather bright I want to optimize by using the smaller bands of my Ha Filter.
Michael
I do narrowband imaging with an OSC Camera and over time I have now gathered data with the same scope/frame duration but with different filters. I am now thinking about the best process of combining the data.
One pre-requisite: I try to keep my image chain as dust free as possible and looking at my flats this seems to have worked out quite well over the past months. When framing I rotate the whole telescope so that the position between Camera and Scope stays the same.
For example, I now have a dataset for the Seagull nebula shot with the Antlia Ha (6h) /Oiii (3.5h)/Sii(7h) filter and when finally the ALP-T filter was available I also shot arround 4h for the same target.
So my goal is to get the best possible integration out of the dataset.
My current process is as folowing:
change the fits header for FILTER of ALP-T to match the Antlia Ha Filter
copy the ALP-T data to a new directory and change the FILTER Header to match my Antlia Oiii Filter
add all files+flats+darkflats+darks+bias to Weighted Batch Processing and let it do it's magic…
I am taking two shortcuts here, pictures from different sessions share the same flat and I ignore the flats of the ALP-T and only use the flats for the single band filters, but I hope to get away with that because of the rather dust free image chain.
The approach in AstroPixelProcessor with it's sessions looks like overall better suited as there I can have sessions with flats for each session and I can tell it that it should only process Ha (or Oiii). But to be honest, I am not a too big fan of APP and try to avoid it if I can as the Weighted Batch Processor Script is evolving with every PixInsight release and APP seems kind of stuck in development, although for this case it looks as a better fit…..
As another process I thought about running Weighted Batch processor several times and then using the files from the debayered folder (which should be propery calibrated) as a new input for a 2nd run of weighted batch processor that takes all the debayered images that were preprocessed in a 'Session' way.
But to be able to do so I there also have to tweak the fits data, likely I need to remove the bayer matrix entry and also here will have to tweak the FILTER keyword.
As it is not so easy for me to modify fits headers in xisf files this will first need some experimentation on how to best automate the process.
Any thoughts on how the whole process can be done more efficient (or better) in Weighted Batch Processor? I still plan to keep the process of shooting both ALP-T and the single band filters on the same target, in no-so bright-moon nights the ALP-T allows me to get Oiii Integration time for free and when the moon is rather bright I want to optimize by using the smaller bands of my Ha Filter.
Michael
