Tool for image diffs.
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
docs
imagediffer
tests
.gitignore
.travis.yml
LICENSE
MANIFEST.in
README.rst
requirements.readthedocs.txt
requirements.txt
setup.cfg
setup.py

README.rst

Image Differ

https://travis-ci.org/janslifka/imagediffer.svg?branch=master

https://readthedocs.org/projects/imagediffer/badge/?version=latest

Image Differ is a tool for image comparison. It loads 2 images (either local files or from the internet) and generates the diff image that shows where the images are different. It also calculates:

  • Percentage of pixels where the images differ (with adjustable threshold for difference)
  • Mean Square Error for the images
  • Structural Similarity Index

It can also calculate comparison stats for individual color channels or convert images to grayscale.

Image Differ can be used as a library for other projects or it provides GUI application.

Installation

The easiest way to install imagediffer is using pip

python -m pip install imagediffer

You can also clone this repository, install requirements using pip and then install imagediffer using setup.py.

python setup.py install

How to use the app

python -m imagediffer

Once the application window is opened you can use Image > Load first/second image from file/URL to open images. If the first and the second image have the same dimensions, the comparison image is generated.

You can choose if you want to use euclidean or chebyshev distance for color comparison. Toleracne defines the threshold where the colors are considered the same.

You can choose what color channels you want to compare from full color comparison, grayscale or individual color channels.

Stats show Mismatch - how many pixels exceeded the tolerance threshold, MSE (mean squared error) and SSIM (structural similarity index).

Documentation

If you want to build documentation yourself, you need to have sphinx doc installed and then run make html in the docs folder.

Development

Simply clone the project and install dependencies from requirements.txt file using pip.

python -m pip install -r requirements.txt

Used libraries