Hi,
Firstly ASTAP is a great piece of software - the tilt and sensor analysis, plate solving and annotation are all fantastic.
I'm having trouble trying to use it as a FITS viewer though. I have a live stacked FITS file created in ASI LIve from an OSC colour camera. However, when I view in ASTAP some weird stuff seems to happen with the star colours. I attach 4 screen clips below. To be clear these are all screenshots from the same FITS file viewed four different ways:
Output from ASI FITS viewer - who knows what automatic magic it applies behind the scenes but it looks good to me as live stacks go
Output from Cloudmakers FITS preview - this does nothing other than a set the black and white points at 0 and 1024.
Ouput from ASTAP - I manually set the same white and black points to compare with above - it's the whacky colours that surpirse me
Output from ASTAP using the Low setting for the histogram
(Not sure I can attach multiple files to a posting so I've mangled them all into one)
The shifted star colour is what you typically see if there is a colour failure in the optical system or if your work with seperate colour filters. Since your using a OSC camera this must be the optical system, The colour blue is on the top and red is at the bottom. I assume the other program do a colour smooth making all stars a similar white. ASTAP can also do if you select option colour smooth in tab "stack method" (do this before debayer) or as after care in tab pixel math 1, "smart colour smoothing" Note this colour smoothing takes some time.
Since the image is already stacked, the image must be already in colour. Then you can only apply "smart colour smoothing" in tab pixel math 1. See attached.
So ASTAP is showing the real image not a filtered version.
You can attach more images once you have created a post at this forum.
Hope this helps, f you want, attach the fits image and I can have a look.
Thanks. The original 20 files are pretty clean and if I stack them ASTAP and view the result it's fine. I think it's a legacy of how ZWO deals with colour in the stacking (i.e. not as good) and how different viewer programs handle clipping when the highlight slider or auto histogram does it's thing.
For the avoidance of doubt I have only just started using ASTAP so if I'm using it wrong sorry....
If I set the highlight slider to 65536 in all the programs the results looks the same (phew!). There is some feint colour in the edges presumably a legacy of the unsophisticated live stacker working with OSC data. However, in other viewers as I drag the highlight slider down the central white part grows completely white and the border expands as lower levels are pulled out of the background - which is what I would expect.
In ASTAP (this maybe a settings error on my part) but as I drag the maximum slider down instead of going white the star core goes saturated multicoloured as if it is preserving and exaggerating the feint colours that are there. I realise there may be situations where a pixel that is say 65534,65534,65535 (i.e. very very slightly blue) may wish to preserve some of the relative blueness but it seems to quite strong in this example and makes the auto stretching mildly unhelpful.
I've uploaded 5 files of screenshots of what happens as a drag the maximum slider down in ASTAP. This may be intentional but if so is there a way to turn it off for a more convential clipping approach?
I probably can't attach the FITS file as it's massive but I could create a shared folder on dropbox.
The star colours look pretty normal. The sensor Bayer matrix is not very suited for star colours. The star has a very sharp peak value and further aways from the star center the flux value drops fast. So you can't measure accurately the colour of the intensity slope of a star. One side of the Bayer matrix will be closer the the star center then the other side. The only thing you can do it to smooth the colours and maintain the intensity. That function is available as mentioned earlier.
Share some images by Dropbox and I will have a look how they stack.
Regards, Han
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi,
Firstly ASTAP is a great piece of software - the tilt and sensor analysis, plate solving and annotation are all fantastic.
I'm having trouble trying to use it as a FITS viewer though. I have a live stacked FITS file created in ASI LIve from an OSC colour camera. However, when I view in ASTAP some weird stuff seems to happen with the star colours. I attach 4 screen clips below. To be clear these are all screenshots from the same FITS file viewed four different ways:
(Not sure I can attach multiple files to a posting so I've mangled them all into one)
Any thoughts?
Ian
The shifted star colour is what you typically see if there is a colour failure in the optical system or if your work with seperate colour filters. Since your using a OSC camera this must be the optical system, The colour blue is on the top and red is at the bottom. I assume the other program do a colour smooth making all stars a similar white. ASTAP can also do if you select option colour smooth in tab "stack method" (do this before debayer) or as after care in tab pixel math 1, "smart colour smoothing" Note this colour smoothing takes some time.
Since the image is already stacked, the image must be already in colour. Then you can only apply "smart colour smoothing" in tab pixel math 1. See attached.
So ASTAP is showing the real image not a filtered version.
You can attach more images once you have created a post at this forum.
Hope this helps, f you want, attach the fits image and I can have a look.
Cheers, Han
Thanks. The original 20 files are pretty clean and if I stack them ASTAP and view the result it's fine. I think it's a legacy of how ZWO deals with colour in the stacking (i.e. not as good) and how different viewer programs handle clipping when the highlight slider or auto histogram does it's thing.
For the avoidance of doubt I have only just started using ASTAP so if I'm using it wrong sorry....
If I set the highlight slider to 65536 in all the programs the results looks the same (phew!). There is some feint colour in the edges presumably a legacy of the unsophisticated live stacker working with OSC data. However, in other viewers as I drag the highlight slider down the central white part grows completely white and the border expands as lower levels are pulled out of the background - which is what I would expect.
In ASTAP (this maybe a settings error on my part) but as I drag the maximum slider down instead of going white the star core goes saturated multicoloured as if it is preserving and exaggerating the feint colours that are there. I realise there may be situations where a pixel that is say 65534,65534,65535 (i.e. very very slightly blue) may wish to preserve some of the relative blueness but it seems to quite strong in this example and makes the auto stretching mildly unhelpful.
I've uploaded 5 files of screenshots of what happens as a drag the maximum slider down in ASTAP. This may be intentional but if so is there a way to turn it off for a more convential clipping approach?
I probably can't attach the FITS file as it's massive but I could create a shared folder on dropbox.
Regards,
Ian
Hello Ian,
The star colours look pretty normal. The sensor Bayer matrix is not very suited for star colours. The star has a very sharp peak value and further aways from the star center the flux value drops fast. So you can't measure accurately the colour of the intensity slope of a star. One side of the Bayer matrix will be closer the the star center then the other side. The only thing you can do it to smooth the colours and maintain the intensity. That function is available as mentioned earlier.
Share some images by Dropbox and I will have a look how they stack.
Regards, Han