Construction
1
feat: use eccentricity instead of minor_length in detection (line 52, 117)
2
fix: object shape bug when using min_sep (line 71)
3
fix: changed to softer constraint of amt of sources in ComputeTransformXYShift (could be made variable?)
4
feat: added automatic metadata and header to stack (from last image used for stack) because of issue with querying the stack before
5
fix: minor fixes with slicing and shapes in fluxes (line 126, line 297)
6
feat: masking, as image attribute, used in DAOFindStars and AperturePhotometry Currently, the mask is not set anywhere. I would like to discuss this. At first, I had it set for every image in the CleanBadPixels block to be set to the computed(or imposed) bad pixel map. However, I feel like it might be advantageous to leave more flexibility, maybe by adding a block called 'MaskImage', that takes a mask as an argument and writes it to the mask attribute of every image that passes through in the sequence. This way, one can use all different kinds of masks, instead of just bad pixel masks. Furthermore, one could leave out the CleanBadPixels block altogether (which currently automatically interpolates images when applied) in this way and still be able to mask nonetheless.
What are your thoughts?
Thanks for this PR @schackey 🙏🏼 these are important changes. I am happy to merge once we are set on the few comments I've made.
Hi @schackey, should I take over this one?