search

Nat1405 / Nifty4Gemini

A Gemini Data Reduction Pipeline Framework
MIT License
0 stars 0 forks source link

Cube Merging for Non-Sidereal Targets #6

Open Nat1405 opened 4 years ago

Nat1405 commented 4 years ago

When reducing several NIFS observations of Titan, the final cube merging creates a frame that looks like this:

image

Where an individual observation looks like this:

image

Maybe there's a way to look into the FITS headers to fix this, or maybe a new option to find the centroid of sources would be nice.

Nat1405 commented 4 years ago

Tentative work plan for this (can be done in resizeAndCentreCubes() and makeWavelengthOffsets() ):

Nat1405 commented 4 years ago

Pushed a sort-of fix in 7f1ad57c3b05f. As you can see the cube merging still isn't perfect (a fainter cube is slightly up and left of the bright cube) and I still need to check if the imperfection is due to the merging method or centering errors of the target. As well, header information still isn't properly propagated to the final merged cube. However, PIs do have the option of providing their own offsets to the pipeline if they think the cube merging is a little off centre.

The test final merged cube:

image

One of the input cubes:

image