I'm looking for the best piece, or combination or pieces of software where I take a scanned image of a table, apply some OCR and are able to convert it to a .csv format or similar.

I've been playing around with Tesseract, but it doesn't seem to preserve the whitespace for constructing tables. The only way I could think to do it, is by perhaps using Hough transforms to detect the lines of the tables, then segmenting the image into pieces and doing OCR on the patches of the table before reconstructing.

The above seems very involved and of course some tables don't have an outline. I've seen software out there like ABBYY/FineReader that do a fantastic job. I wondered if anyone knows of a fairly good open source equivalent?


There is a very promising project Layout Parser that aims to simplify table digitization. It is based on tesseract, but has many enhancements specifically for tables.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.