-
Notifications
You must be signed in to change notification settings - Fork 0
AIA Image Calibration (aia_prep)
aia_prep is a SSWIDL routine that does the level 1 to level 1.5 conversion of aia fits files. In english this means it aligins and centres the Sun and updates the header. The SSWIDL code can be found here: From that code aia_prep has the following purpose:
"Perform image registration (rotation, translation, scaling) of Level 1 AIA images, and update the header information."
The goal of the conversion to SunPy is to produce identical results as the SSW version so there is no problem using code or images from either calibration routine.
The aia_prep routine consits of an affline transform and header parameter updates. The affline transform centres the Sun to the origin in the image, rotates the image and rescales it so the pixels are all identical sizes for all wavelengths and times to compensate for any spacecraft movement.
The affline transform used in the SSWIDL routine is in the IDL ROT routine. This uses a bi-cubic convolution interpolation method (by default in aia_prep.pro [with a interpolation parameter a = -0.5]). To maintain compatibility with the SSWIDL routine this method of interpolation for the affline transform was coded into a C-API extension and pushed into SunPy under the map.rotate() method in PR #288, along with support for the bilinear method supported with IDL's ROT.
The affline transform method in scipy is also supported, as a fall back in the case of no C-API extention, however this will not generate exactly the same results as the default SSWIDL aia_prep code.