Global stitching Perkin Elmer Phenix Images

This is a data science pipeline to allow for the reconstruction of stitched images from the Perkin Elmer Phenix instrument. Currently, images collected at 5x objective, and 1% overlap are mosaic'ed together using this python script.

This project is organised as a data-science pipeline to allow for faithful reconstruction of test-data for ease of understanding how these function works. At least, I hope for you!

Happy to be tweeted @ajay_bhargava, and please give credit where credit is due.

Project Organization

.
├── AUTHORS.md
├── LICENSE
├── README.md
└── data
    ├── raw (not included raw data, but keep it here)
├── reports
    ├── figures
└── src
    ├── tools
    └── visualization (tools to do things with images once stitched)

Dependencies

This package relies on the following dependencies: tqdm,PIL, numpy, and xml.etree

Calling the function

Start by navigating to ./src/data/

The function PE_phenix_stack_stitcher.py takes two arguments: 1) Input Path to Images Folder 2) Output Path to where stitched images go

The function is called from terminal/bash as such:

foo@bar:~$ python3 PE_phenix_stack_stitcher.py "Input Path" "Output Path"

Getting the test data

Unfortunately because my test data is unpublished, it still cannot be released for you on AWS at the moment. Ask me later.

Future Work (or your contribution)

1) Trying to accommodate different kinds of images at different magnifications. 2) Doing time-lapse, multi-position (in the z-axis)