OME-TIFF
OME-TIFF Specification
We aim to fully comply with the official specification for OME-TIFF as defined by:
https://ome-model.readthedocs.io/en/stable/ome-tiff/index.html
ome-writers supports writing to OME-TIFF, a stable and widely used standard
for storing multi-dimensional biological image data. OME-TIFF is an extension
of the standard TIFF format that includes metadata defined by the OME-XML Data
Model to
describe the higher order dimensionality and experimental context of the image
data.
Expected Output Structure
When writing an acquisition to OME-TIFF, the expected output structure depends
on your AcquisitionSettings. OME-TIFF
supports no more than 5 dimensions per image (strictly: T, C, Z, Y,
X). Therefore, acquisitions with additional dimensions (e.g., stage position,
tiles, angles, etc.) are broken into series of ≤5D images. The OME-TIFF
specification supports arranging those series either in a single large file, or
as multiple files in a directory structure. ome-writers currently only
supports the latter approach: positions/series are always written as separate
files in the output directory
Any acquisition with a combination of no more than "TCZXY" dimensions is
stored as a single OME-TIFF file, directly at the
AcquisitionSettings.output_path.
Example Code
Acquisitions that contain multiple positions (e.g., stage positions, tiled
images, angles on a light-sheet microscope, etc.) or that exceed 5
dimensions in any way other will be structured as multiple OME-TIFF files, one per
position/series. For example, if AcquisitionSettings.root_path is
"dest/my_file.ome.tiff" and there are 4 positions in a
position dimension (type='position'), the output will be:
dest/my_file_position_p000.ome.tiff
dest/my_file_position_p001.ome.tiff
dest/my_file_position_p002.ome.tiff
dest/my_file_position_p003.ome.tiff
Example Code
If you declare AcquisitionSettings.plate along with a position dimension
(type='position') containing Positions that define
plate_well/plate_column information, the output follows the same pattern as above,
but HCS metadata is included in the OME-XML. Work in progress: full support for
HCS metadata in OME-TIFF is still being developed.
Example Code
Backends
tifffile(default): Uses the populartifffilePython library to write OME-TIFF files, andome-typesto generate the OME-XML metadata embedded in the TIFF header.