Flat frame gradient removal and optical axis

[froesccn]

It appears the AstroImageJ data processor plugin runs a gradient removal when generating the master flat. I noticed this when processing a data set that had been cropped from a much larger field of view. The center of vignetting was actually off axis, but forcibly shifted to the image center by the data processor. It would seem that this makes sense only when the optical axis is actually in the center of the image, otherwise it can introduce almost arbitrarily large errors.

I wondered whether the assumption is hard-coded or whether there is a FITS header keyword that should be set to indicate the optical axis. If not it might be useful to make the gradient removal optional, and/or to have a warning when the estimated gradient is much larger than what could be expected from a reasonable flat frame. The cropping use case is probably exotic, but I suppose there could also be some cameras with multi-chip configurations where this might be a problem.

BTW I don't need an urgent fix, just curious. I'm processing data using my own code for my study and used AstroImageJ results as a baseline for comparison.

[karenacollins]

Gradient removal can be disabled by deselecting Data Processor->Preferences->"Remove gradient from calibrated raw flats when creating master flat". This should solve the problem.
On 10/20/2013 12:15 PM, froesccn [via AstroImageJ] wrote:

It appears the AstroImageJ data processor plugin runs a gradient removal when generating the master flat. I noticed this when processing a data set that had been cropped from a much larger field of view. The center of vignetting was actually off axis, but forcibly shifted to the image center by the data processor. It would seem that this makes sense only when the optical axis is actually in the center of the image, otherwise it can introduce almost arbitrarily large errors.

I wondered whether the assumption is hard-coded or whether there is a FITS header keyword that should be set to indicate the optical axis. If not it might be useful to make the gradient removal optional, and/or to have a warning when the estimated gradient is much larger than what could be expected from a reasonable flat frame. The cropping use case is probably exotic, but I suppose there could also be some cameras with multi-chip configurations where this might be a problem.

BTW I don't need an urgent fix, just curious. I'm processing data using my own code for my study and used AstroImageJ results as a baseline for comparison.

---

If you reply to this email, your message will be added to the discussion below:

http://astroimagej.1065399.n5.nabble.com/Flat-frame-gradient-removal-and-optical-axis-tp116.html

To start a new topic under AstroImageJ, email [hidden email]
To unsubscribe from AstroImageJ, click here.
NAML

[froesccn]

karenacollins wrote

Gradient removal can be disabled by deselecting Data
Processor->Preferences->"Remove gradient from calibrated raw flats when
creating master flat". This should solve the problem.

Oops ... I completely overlooked that this dialog has a menu bar of its own ;)

Thanks!

  Christian

[karenacollins]

No problem. I hope that takes care of the issue.
On 10/20/2013 1:03 PM, froesccn [via AstroImageJ] wrote:

karenacollins wrote

Gradient removal can be disabled by deselecting Data
Processor->Preferences->"Remove gradient from calibrated raw flats when
creating master flat". This should solve the problem.

Oops ... I completely overlooked that this dialog has a menu bar of its own ;)

Thanks!

  Christian

---

If you reply to this email, your message will be added to the discussion below:

http://astroimagej.1065399.n5.nabble.com/Flat-frame-gradient-removal-and-optical-axis-tp116p118.html

To start a new topic under AstroImageJ, email [hidden email]
To unsubscribe from AstroImageJ, click here.
NAML