Sub-stack Training & Uniform Background Indicator

This release includes two major developments:

Training PCA models requires large numbers of input images. In cases where the number of input images exceeds the memory contraints of your computer, this update allows you to generate a set of training images from approximately equal depth sub-stacks of median co-additions. This increases the number of images available for training and thus increases the accuracy of your PCA model.

The Uniform Background Indicator (or UBI) can be calculated for each image using the calculate_UBI function in fringez.model. The UBI is calculated by performing aperture photometry at random locations on the background of an image and dividing the standard deviation of the measured background fluxes with the median error in these flux measurements. For an image of uniform Gaussian noise, Ψ ≈ 1 as the distribution in background fluxes will be equal to the average flux error across all measurements. In this case, there exists no global variance which is not captured by the local error term. For an image with correlated background noise, Ψ > 1 because the different background values sampled by the apertures will introduce additional variance into the numerator of Equation 1 not captured by the local error terms in the denominator.

Correct Atmospheric Fringes in ZTF i-Band Images

v1.0.2

direct to github page for model dates