Scripts/notebooks to segment and track platelets in a growing thrombus

Current files:

• fl.py contains some utilities used in the notebooks, including objects for
  opening the ND2 file and keeping track of various processed data products
  (smoothing, thresholding, segmentation, etc.)
• The "process images" notebook is the first step, goes from images to
  segmentation. It creates output files in a timestamped subfolder of the
  current working directory.
• The "define variables" notebook then uses that segmentation to create tracks.

Goals:

• turn this from a set of loosely tied notebooks to a command-line utility that
  runs various parts of the pipeline
• use zarr to save the original data and intermediate products
• use dask for processing where possible
• get the tracks into the napari tracks layer from napari/napari#1361

Installation:

git clone https://github.com/jni/pia-tracking.git
cd pia-tracking
conda env create -f environment.yml
conda activate platelet-tracking