Skip to content

Latest commit

 

History

History
43 lines (29 loc) · 1.5 KB

README.md

File metadata and controls

43 lines (29 loc) · 1.5 KB

covert-scope-jobs

Run tifconvert.sh.
tiffix.py extracts metadata and performs shading correction on images.

When you collect a new reference for shading correction, push it to data/.

Data Flow

The scope compter uploads new tif files to ResearchNAS every 2 hours. The list of uploaded files is appended to robocopy.log.

This program first reads robocopy.log, extract a list of tif files as tifffiles.txt and remove the robocopy.log.

Each image in tifffiles.txt goes through tiffix.py for the following process.

  • shading correction if there is a proper reference for the corresponding channel, magnification and binning.
  • tif compression.

To change schedule for tifconvert.sh (not uploading)

crontab -e
0 */2 * * * cd ~/covert-scope-jobs/ && git pull && sh tifconvert.sh

Considerations

mount command seems to be a lot faster than using Connect to Server.
However, when you use mount, ResearchNAS does not allow you to write files without sudo privilage.
Use sudo visudo to exclude sh script from asking password.

TODO:

  • parallel processing
  • fix sudo issue
  • connect to researchNAS with a new account that has modify permission in instruments folder
  • truncate the odd size image to even size.
  • make metadata compatible with imageJ Show Info
  • logs for tifconvert.sh
  • collect more shading correction references.
  • write a function for estimating the reference from 1x1 binning image if 2-4 binning does not exist (imresize?)
  • (extract time points?)