A library to read Andor SIF file

SIF, Andor, Image, reader
pip install sifreader==0.2.4


SIF Reader

This package is designed to read Andor SIF image file.

In addition to raw image data, it provides access to the wavelength axis for spectra, and metadata such as exposure time, gain, recording date etc. Image data can be read as a numpy array or as an xarray DataArray, which contains the image data as well as the coordinate axes and labels in a single object. For spectra, the spectral axis can be returned as nanometers, wavenumbers or electron volts.

To the best of our knowledge, this is currently the only Python package that can extract wavelength information from .sif files using only pure Python and numpy (i.e. without relying on any Andor dll libraries).


numpy (optionally xarray, pytest to run the unit tests)


Install with pip:

>pip install sifreader


Open a file and print the metadata:

>file = SIFFile('my_image.sif')
<sifreader.sifreader.sifreader.SIFFile object at 0x30f9eecc0>
Original Filename:             E:\test_sequence.sif
Date:                          Thu May 10 12:01:48 2018

Read a single or all frames contained in the file as numpy arrays:

>first_frame = file.read_block(0)
>all_frames = file.read_all()

The horizontal axis is contained in the x_axis member variable. If the file contains a spectrum, the axis will be the wavelength in namometers, otherwise it will contain the pixel numbers:

>wavelengths = file.x_axis

With the optional xarray package

Read all frames in the file as a DataArray:

>xarr = file.as_xarray()

For spectra: make a DataArray that contains the photon energy in eV rather than the default wavelength:

>spectrum_file = SIFFile('my_spectrum.sif')
>xarr = spectrum_file.as_xarray('photon_energy')

The options for this method are 'wavelength' (default), 'wavenumber' and 'photon_energy'. Note, that this only makes a difference if the file contains a spectrum. For images, both axes will always contain the pixel numbers.

One of the nice features of DataArray is the ability to easily select data and plot it in a single line:

spectrum_file.as_xarray().sel(frames=0, wavelength=slice(749.5, 768.2)).plot()

Version History

  • 0.2: Added support for wavelength information, xarray exporting and unit tests
  • 0.1: First release