fitting on the celestial sphere

Lang and I got the Tractor working today in celestial coordinates. Usually source detection, photometry, and morphological fitting is done in images in pixel space (x-y position on the CCD). But if you want to properly combine data from multiple images (and especially images in different bands or with different PSFs), you have to do your fitting simultaneously in both images, and use some world coordinate system to convert positions in one image to positions in another. We gave the Tractor this capability today, in a very one-sided pair-coding session (Lang worked, I chatted).

No comments:

Post a Comment