pyamiimage is a set of tools to extract semantic information from scientific diagrams.
The current goal is to extract terpene synthase pathway diagrams.
'Extraction' means that we will go from pixel values in an image to a 'smart diagram'. The output of pyamiimage is an image with annotations of substrate, products and enzymes.
We are working to add more support for open formats that encode chemical/pathway information such as CML and GPML.
To run pyamiimage on your local system you need to have Tesseract installed. If you don't have Tesseract installed, install it from here.
pip install pyamiimage
pyamiimage is a command-line tool and can be accessed via the terminal or command prompt. To bring up the help run:
pyamiimage --help
You can also include pyamiimage in your program using the provided classes.
AmiImage class provides methods for image manipulation.
from pyamiimage.ami_image import AmiImage
gray = AmiImage.create_grayscale_from_file(image_file_path)
AmiGraph class generate a graph from arrows in a diagram.
AmiOCR class provides methods to extract words from the iamge. Uses Tesseract.
merged main into nodes_and_pixels and re-branched