Another enjoyable morning today working with Dustin Lang (Toronto) on combining images. We looked at how the Zackay paper I talked about yesterday relates to ideas in speckle imaging. And Lang showed me his very nice argument that a detection map (like an image but really a statement about maximum-likelihood flux estimates for stars) that is sufficient for detecting point sources in the field is also sufficient for detecting galaxies. That is intuitively surprising. But it is also a claim in the Zackay paper. We have generalized the result, I think, to include variations in pixel scale, noise, and bandpass across the collection of images used to make the detection map.
No comments:
Post a Comment