xtiff 0.4.1

A tiny Python 3 library for writing multi-channel TIFF stacks

xtiff

A tiny Python 3 library for writing multi-channel TIFF stacks.

The aim of this library is to provide an easy way to write multi-channel image stacks for external visualization and analysis. It acts as an interface to the popular tifffile package and supports xarray DataArrays as well as numpy-compatible data structures.

To maximize compatibility with third-party software, the images are written in standard-compliant fashion, with minimal metadata and in TZCYX channel order. In particular, a minimal (but extensible) subset of the OME-TIFF standard is supported, enabling the naming of channels.

Installation

Install from pypi:

pip install xtiff

Usage

The package provides the following main function for writing TIFF files:

to_tiff(img, file, image_name=None, image_date=None, channel_names=None, description=None,
        profile=TiffProfile.OME_TIFF, big_endian=None, big_tiff=None, big_tiff_threshold=4261412864, 
        compression_type=None, compression_level=0, pixel_size=None, pixel_depth=None, 
        ome_xml=get_ome_xml, ome_xml_template=_get_default_ome_xml_template(), **ome_xml_kwargs)

Documentation of the function parameters is available via Python's internal help system: help(xtiff.to_tiff)

In addition, the function get_ome_xml() is provided as the default OME-XML-generating function.

FAQ

What metadata is included in the written images?

In general, written metadata is kept at a minimum and only information that can be inferred from the raw image data is included (image dimensions, data type, number of channels, channel names). Additional metadata natively supported by the tifffile package can be specified using function parameters. For OME-TIFF files, the OME-XML "Description" tag contents can be further refined by customizing the OME-XML template or by providing a custom OME-XML-generating function.

Why should I care about TIFF? I use Zarr/NetCDF/whatever.

That's good! TIFF is an old and complex file format, has many disadvantages and is impractical for storing large images. However, it also remains one of the most widely used scientific image formats and is (at least partially) supported by many popular tools, such as ImageJ. With xtiff, you can continue to store your images in your favorite file format, while having the opportunity to easily convert them to a format that can be read by (almost) any tool.

Why can't I use the tifffile package directly?

Of course you can! Christoph Gohlke's tifffile package provides a very powerful and feature-complete interface for writing TIFF files and is the backend for xtiff. Essentially, the xtiff package is just a wrapper for tifffile. While you can in principle write any image directly with tifffile, in many cases, the flexibility of the TIFF format can be daunting. The xtiff package reduces the configuration burden and metadata to an essential minimum.

Change log

2019-12-12 v0.1.2 - Initial release
2019-12-12 v0.2.1 - Expose OME-XML to user
2019-12-12 v0.2.2 - Support for ome_xml_kwargs
2019-12-13 v0.3.0 - Simplified to_tiff interface
2019-12-15 v0.4.0 - Added description parameter
2020-01-13 v0.4.1 - Fixed XML encoding and XSD compliance

License

This project is licensed under the MIT license.