masks in MSCSKYSUB
bomond wrote on Aug 26, 2007
Hi
I'm using MSCRED to reduce INT mosaic data and now I want to fit and subtract the sky with MSCSKYSUB. I've tryed to set "mask=!objmask" to use the output of OBJMASKS task or "mask=!BPM" to use the mask created by MSCIMAGE or not to use any masks at all and the result is the same. I've tried also to refer directly to the bpm file with no success. It looks like the task cannot use these masks. It tries to fit the empty regions with a pixel value of 0, outside the chips.
And one second question: How can I use both masks simultaneously?
many thanks,
bomond
I'm using MSCRED to reduce INT mosaic data and now I want to fit and subtract the sky with MSCSKYSUB. I've tryed to set "mask=!objmask" to use the output of OBJMASKS task or "mask=!BPM" to use the mask created by MSCIMAGE or not to use any masks at all and the result is the same. I've tried also to refer directly to the bpm file with no success. It looks like the task cannot use these masks. It tries to fit the empty regions with a pixel value of 0, outside the chips.
And one second question: How can I use both masks simultaneously?
many thanks,
bomond
Francisco Valdes wrote on Aug 26, 2007
Hi,
To use a mask you need to set regions to "mask" and set the mask parameter to the mask you want or a keyword pointer with the !keyword syntax. Good pixels have a value of 0. If this is what you are doing then you are deciding it is not using the mask because the result is the same? It may be that the bad pixels do not significantly alter the surface fit? If you want to discuss this further please include a parameter listing.
As for combining masks that has to be done externally. The most direct tool to use is imexpr with something like:
Yours,
Frank Valdes
To use a mask you need to set regions to "mask" and set the mask parameter to the mask you want or a keyword pointer with the !keyword syntax. Good pixels have a value of 0. If this is what you are doing then you are deciding it is not using the mask because the result is the same? It may be that the bad pixels do not significantly alter the surface fit? If you want to discuss this further please include a parameter listing.
As for combining masks that has to be done externally. The most direct tool to use is imexpr with something like:
cl> imexpr "(a==0&&b==0)?0 : 1" out.pl bpm objmask
Yours,
Frank Valdes
Last post on Aug 26, 2007