FITS header behavior and new features
Posted: 22 Jan 2022, 17:17
I wanted to share some observations on the data in the FITS header created with AstroArt (not necessarily version-specific).
1. Binning:
binned images from RGB Demosaic (binning 2x2) still show the same PIXSZ as the original setting. This should be recalculated with the binning factor! e.g. 3.8µm then it should read 7.6µm. This works in other situations with binning (e.g. camera binning or software binning).
In addition, I would find it helpful to have a keyword on BINNING, (e.g. XBINNING and YBINNING like in SharpCap).
2. After deleting annotations in the "annotations" window, they are not removed from the FITS header and will be saved and reappear after reopening the file. I have to remove the corresponding COMMENT lines in the header manually.
I am using the "annotate saturated stars" for image capture and defining exposure time. After preprocessing and aligning the images, the circles have lost their original position and i want to remove them in the final image anyway.
3. New feature: With the modern CMOS cameras, there are more parameters to control the cameras than in the CCD times. Specifically, gain and offset/ black level (or similar terms) or the readout mode can be chosen and are impacting the resulting image and its quality. I am trying to add them to the COMMENT section manually, but may forget them sometimes. Ideally, if they could be read from the camera driver and (optionally) inserted into the header, that would help.
Other acquisition software like SharpCap are using these keywords already, although they are not standard FITS keywords, AFAIK.
4. New feature: When reading about the standard FITS keywords, i found 2 additional ones that could be helpful:
FILTER (optional): for filters used
MOONANGL (Distance to Moon): when the image is plate-solved, that could be helpful sometimes.
5. The camera "TEMPERAT" data seem to be truncated to 1 decimal now. Very good.
6. Also, to be able to save the processing history into the header is very good!
What do other users think about these topics?
Thanks a lot
Bernd
1. Binning:
binned images from RGB Demosaic (binning 2x2) still show the same PIXSZ as the original setting. This should be recalculated with the binning factor! e.g. 3.8µm then it should read 7.6µm. This works in other situations with binning (e.g. camera binning or software binning).
In addition, I would find it helpful to have a keyword on BINNING, (e.g. XBINNING and YBINNING like in SharpCap).
2. After deleting annotations in the "annotations" window, they are not removed from the FITS header and will be saved and reappear after reopening the file. I have to remove the corresponding COMMENT lines in the header manually.
I am using the "annotate saturated stars" for image capture and defining exposure time. After preprocessing and aligning the images, the circles have lost their original position and i want to remove them in the final image anyway.
3. New feature: With the modern CMOS cameras, there are more parameters to control the cameras than in the CCD times. Specifically, gain and offset/ black level (or similar terms) or the readout mode can be chosen and are impacting the resulting image and its quality. I am trying to add them to the COMMENT section manually, but may forget them sometimes. Ideally, if they could be read from the camera driver and (optionally) inserted into the header, that would help.
Other acquisition software like SharpCap are using these keywords already, although they are not standard FITS keywords, AFAIK.
4. New feature: When reading about the standard FITS keywords, i found 2 additional ones that could be helpful:
FILTER (optional): for filters used
MOONANGL (Distance to Moon): when the image is plate-solved, that could be helpful sometimes.
5. The camera "TEMPERAT" data seem to be truncated to 1 decimal now. Very good.
6. Also, to be able to save the processing history into the header is very good!
What do other users think about these topics?
Thanks a lot
Bernd