Python script that extracts texts from images in specified directory using openCV library. In current stage it's just proof of concept, more updates will be done.
Lecturers at my girlfriend's school decided to paste images with text into the class documents instead of actual text, so if you want to find some information quickly in certain document, you cannot, because all Word or any Word/PDF viewer sees, is just images, making it impossible to use search in said document. This script extracts those texts from those images and then saves them in an output file, in which you can search.
Is wellcomed. Just open Issue/PR with stuff you want to change.
- Instead of feeding the script screenshots/images of the document pages, it should parse the word/pdf itself.
- Mutlifile/single file "upload" (you can parse multiple files and the script will output multiple files, 1:1)
- Maybe some web based uploader/desktop executable, for ease of use