Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

WIP: Compute image ipix and only update masked indices in all sky image #144

Open
wants to merge 1 commit into
base: master
Choose a base branch
from

Conversation

adl1995
Copy link
Member

@adl1995 adl1995 commented Aug 10, 2019

This is a followup of #79. Previously, the output image contained bright lines as the values were being summed twice at tile edges. The current change applies the resulting tile image only at certain indices, which it receives by applying a mask on the all sky image based on the ipix values.

Previous output:
result_unmasked

Current output:
result_masked

However, the updated output image now contains 'holes' where pixels values are not inserted. I believe this is an issue with the warping function we're currently using. I will trying using either skimage.transform.warp_coords or scipy.ndimage.map_coordiantes and see if it resolves the issue.

@adl1995 adl1995 changed the title WIP: Compute image ipix and only update masked indexes in all sky image WIP: Compute image ipix and only update masked indices in all sky image Aug 10, 2019
@cdeil
Copy link
Contributor

cdeil commented Aug 14, 2019

@adl1995 - Thanks for attempting to improve this!

To avoid issues at the edges, you have to call a function in scipy.ndimage or scikit-image to compute the pixel values at "target" coordinates that has an option or even by default extrapolates, i.e. if the target coordinate is just slightly outside the tile image, it will fill the tile edge pixel value and not black or white.

And to be efficient, you have to first compute the target coordinate array, only for the pixels for which a given tile should be used, and only then call the "reproject" or "interpolate" function to compute the pixel values for exactly those target positions.

Hope it's clear ...

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants