We present a zero-configuration image analysis algorithm that can identify and extract monolayer graphene features from a single RGB image.

Install pipenv with python3 -m pip install pipenv. Then a simple pipenv shell and pipenv install should suffice.

The bulk of the algorithm is held in identify.py. The main function, monolayers, takes as input an RGB image of graphene and returns an integer mask representing each individual piece of graphene.

For a demo of the algorithm in action, one can call main.py, which takes a list of files and annotates each of them with a perimeter around each identified piece of graphene as well as its physical dimensions. For example, to run the algorithm on imgs/gr1.png and imgs/gr2.png, run python main.py imgs/gr1.png imgs/gr2.png. Also supports regex so python main.py imgs/*.png is valid. The output is in the annotated/ folder.

For a more detailed description of the algorithm, read here.