Batch preprocessing script & cosmetic correction

[lucchett]

Hello,
Does someone knows when cosmetic correction is applied?
I cannot get rid of smearing columns with the script.

In my opinion it should be applied at the end, after calibration.
Calibration does attenuate these issue, that also affect darks and flats.

many thanks,
Andrea

[Carlos Milovic]

Yes, it is performed after calibration, before alignment. If you need to fine tune the parameters, you may use an ImageContainer, and work with either CosmeticCorrection or DefectMap.

[lucchett]

Thank you Carlos.
I loaded the same defect correction map that I use in my manual process, I don't understand why it doesn't work if it is applied at the end.

Anyway, If I got your suggestion, I should create an I.container that has two steps (calibration through the batch preprocessing and application of defect map).

The thing that I am missing is probably tied to my limited knowledge of I. container: the output of step 1 will be a list of image files in a folder. How can I create a defect map icon that point to this list of file if they are still not created?

Thank you,
Andrea

[Carlos Milovic]

No, the defect map icon encapsulates the instance; this means, its parameters, and the identifier of the image used as map (do not close it, and keep the image identifier). The image container, on the other hand, stores the filepath (or image ids) and where to store the results. So, if you drop any process instance (either an icon or just the blue triangle) over that icon, it will call the filepath list and act on them.

[lucchett]

ok, got it.
I think I've understood also the issue.
Images open reversed in PI 1.8 RC5, comparing to the previous version.
my old map is therefore wrong.(reversed)
do you know if PI will stick to this standard from now on?

Many thanks,
Andrea

[Carlos Milovic]

Go to the FormatExplorer, select the FITS format, and change the orientation to top-bottom (top left corner). PI uses by default the standard used in professional astronomy, not the photographic standard.

[lucchett]

Hi Carlos,
I am sorry, I was wrong.
It seems that the image is opened as usual, but then it is rotated and flipped: both the calibrated and "cosmetized" image are flipped.
the cosmetic correction is therefore applied on a "wrong image"
the image is then restored to the initial orientation, but the defects are there becase the map is applied on the wrong pixels.

Thank you again,
Andrea