This tool is a simplified but powerful interface to Bio-Formats using jpype for direct access to the library. This tool is designed with scalable image analysis in mind, with a simple interface to treat any image like a memory mapped array.
Docker containers with all necessary components are available (see Docker Containers section).
Note: bfio can be used without Java, but only the python and zarr
backends will be usable. This means only files in tiled OME Tiff or OME Zarr format can be
read/written.
In order to use the bioformats backend, it is necessary to first install the JDK and Maven.
The bfio package is generally tested with
JDK 8,
but JDK 11 and later also appear to work.
Here are some info on installing Maven on various OS (Windows | Linux | Mac)
The bfio package and the core dependencies (numpy, tifffile, imagecodecs, scyjava) can
be installed using pip:
pip install bfio
Ubuntu based container with bfio and all dependencies (including Java).
Same as above, except comes with ImageJ and PyImageJ.
Tensorflow container with bfio installed.
Documentation and examples are available on Read the Docs.
We use SemVer for versioning. For the versions available, see the tags on this repository.
Nick Schaub (nick.schaub@nih.gov, nick.schaub@axleinfo.com) Sameeul B Samee (sameeul.samee@axleinfo.com)
This project is licensed under the MIT License Creative Commons License - see the LICENSE file for details.
- Parts of this code were written/modified from existing code found in
tifffile.