Details
-
Type:
Story
-
Status: Invalid
-
Resolution: Done
-
Fix Version/s: None
-
Component/s: ip_diffim
-
Labels:
-
Story Points:1
-
Epic Link:
-
Sprint:AP F19-2
-
Team:Alert Production
Description
The option convolveTemplate=False in image differencing is broken: it convolves the science image instead of the template as intended, but it then subtracts the convolved science image from the science image instead of from the template. See lines 647-659 of imagePsfMatch.py. Note that in the below code, results.matchedExposure is the convolved template image if convolveTemplate=True and is the convolved science image if convolveTemplate=False.
subtractedExposure = afwImage.ExposureF(scienceExposure, True) |
if convolveTemplate: |
subtractedMaskedImage = subtractedExposure.getMaskedImage() |
subtractedMaskedImage -= results.matchedExposure.getMaskedImage() |
subtractedMaskedImage -= results.backgroundModel |
else: |
subtractedExposure.setMaskedImage(results.warpedExposure.getMaskedImage())
|
subtractedMaskedImage = subtractedExposure.getMaskedImage() |
subtractedMaskedImage -= results.matchedExposure.getMaskedImage() |
Attachments
Issue Links
- blocks
-
DM-21217 Add option to select image to convolve based on seeing
- To Do
Further investigation has revealed this issue to be more complicated than the above suggests. Gabor Kovacs [X] will address it on a separate ticket. (Gabor: please link your ticket from here!)