You need to execute and follow the istructions of the setup file your system can execute.
Thats all, we use tesseract ocr for reading text in images(Just that for now). If theres no text the image is renamed to a md5 of its data in binary :P
Idk, this project is licensed under CC-CN-BY-SA and tesseract is licensed under apache license 2
If im wrong tell me i guess. Love ya!