A attempt to make a ticket (fiscal) reader using open cv & tesseract (OCR)
Not working on transparent images.
- find a way to detect blurry images and switch "preprocess" mode based on that
- deal with transparent images
- find a way to improve char detection