Welcome to imread’s documentation!

Imread is a very simple libray. It has three functions


Reads an image from disk


Reads multiple images from disk (only for file formats that support multiple images)


Save an image to disk

That’s it.

This needs to be used with a a computer vision & image processing packages:

This grew out of frustration at current image loading solutions in Python, in either my packages [mahotas] or packages from others [scikit-image, for example].

The relationship with numpy is very contained and this could be easily repurposed to load images in other frameworks, even other programming languages.


This package is an off-shoot of mahotas. As it, currently, does not have its own publication, so you are asked to cite the mother package: If you use imread-mahotas on a scientific publication, please cite:

Luis Pedro Coelho Mahotas: Open source software for scriptable computer vision in Journal of Open Research Software, vol 1, 2013. [DOI]

In Bibtex format:

    author = {Luis Pedro Coelho},
    title = {Mahotas: Open source software for scriptable computer vision},
    journal = {Journal of Open Research Software},
    year = {2013},
    doi = {https://dx.doi.org/10.5334/jors.ac},
    month = {July},
    volume = {1}


On Windows, you can also just download a pre-built package from C. Gohlke’s repository

To compile on debian/ubuntu:

sudo apt-get install libpng12-dev libtiff4-dev libwebp-dev
sudo apt-get install xcftools

To compile on Mac:

sudo port install libpng tiff webp

Either way, you can then install:

pip install imread


Bug Reports

Please report any bugs either on github or by email to luis@luispedro.org

If you have a test case where are not sure of whether imread is behaving correctly, you can discuss this on the pythonvision mailing list

If at all possible, include a small image as a test case.

Indices and tables