OCRmyPDF VS invoice2data

Try https://github.com/ocrmypdf/OCRmyPDF - it uses Tesseract behind the scenes and it absolutely brilliant.

I recommend running any such PDFs through OCRmyPDF.

https://github.com/ocrmypdf/OCRmyPDF

If by "like a photocopy" you mean the file contains images of text rather than text, the MacOS viewer presumably does OCR on the images. I don't know if there's a Linux document viewer with that capability built-in, but a quick search turned up the standalone tool OCRmyPDF.

For anyone interested I solved the problem by first ocr files through the free and open source software ocrmypdf avaible here

ocrmypdf is nice for stuff like that.

Use imagemagick to convert them to PDF and ocrmypdf to straighten and OCR. See this explanation.

Fair enough, maybe this might work for you, it should seperate the text from image anyway and if you have Adobe acrobat it should be able delete the background too with the edit function. It may already be able to do that if you haven't tried it

Using regex: to match patterns in text after converting the PDF to plain text. Examples include invoice2data and traprange-invoice. However, this method requires knowledge of the format of the data fields.

It's not really working. Tried 2 English PDF invoices. Normal format. One came back empty, the other only had the amount right.

I'm assuming they only trained on some specific documents (passport of country X, etc) and all others don't work.

If someone processes the same document all the time, then my invoice2data project may work better and is open source. It's based on Regx, rather than machine learning: https://github.com/invoice-x/invoice2data