To use (with caution), simply do:
$ pip install mementor $ memento <dirname>
Make sure you have a OCR tool like tesseract and certain data value for comparing OCR, eg tesseract-data-eng along with Pillow and Wand for image conversion and loading to draw along with numpy and cv2 for all Image processing required to crop the image with border values defined which will be fetched during pip install.