These two scripts can be integrated into a pipeline to work in succession.
The slicer divides image into tiles of the same size and saves metadata file slicer_info.json that can be used to
stitch tiles back into the big image by stitcher. Both scripts works only with TIFF files.
This script can divide image into tiles by size or by number of tiles.
The -s parameter defines the size of the tile. For example, -s 1000 will produce tiles with the shape 1000 x 1000.
-i path to image file
-o path to output directory
-s size of a tile, default 1000, if set to 0, then -n parameter used instead
-v size of overlap on one side of image, default 0 (no overlap).
e.g. if -v 50, then each tile has 50px overlap on each side: top, bottom, left, right.
--nzplanes number of z-planes, default 1
--nchannels number of channels, default 1
--selected_channels space separated ids of channels you want to slice, e.g. 0 1 3, default all
python slicer_runner.py -i /path/to/img.tif -o /path/to/out/dir -s 1000 -v 50
This script allows to stitch tiles with constant size and overlap (provided by slicer).
It can pick up stitching parameters (overlap and padding) from slicer metadata or they can be provided manually.
By default it will try to load slicer metadata from slicer_info.json inside directory with tiles.
Otherwise, you can provide overlap and padding using -v and -p parameters.
It will ensure that border values in tiles are correctly stitched.
-i path to directory with images
-o path to output file
-v overlap size in pixels, default 0
-p image padding that should be removed, 4 comma separated numbers: left, right, top, bottom. Default: 0,0,0,0.
--slicer_info path to metadata from slicer that contains padding and overlap.
By default, will check for it is a file slicer_info.json in the directory with tiles.
python stitcher_runner.py -i /path/to/dir/with/tiles -o /path/to/out/stitched_img.tif -v 50 -p 50,100,30,0
numpy pandas scikit-image tifffile dask