News for hackers. Stuff that matters.

A comment on What's so hard about PDF text extraction?.

Re:What's so hard about PDF text extraction? by mkjmkumar on Tuesday March 03, 2020 @09:28AM (#22474678)
Around couple of years ago I am working on a home project and utilised Tesseract and Laptonica for OCR. Storage and search HDFS, HBase and SolrCloud on extracted text. You can find the details here on my website. I was very impressed with conversion of hand written pdf docs with 90% readable accuracy. I have named it as Content Data Store(CDS) http://ammozon.co.in/headtohead/?p=153 . Source code is open and you may find steps on installation and how to run here. http://ammozon.co.in/headtohead/?p=129 http://ammozon.co.in/headtohead/?p=126 A short demo http://ammozon.co.in/gif/ocr.gif

I didnot get time to enhance it further but planning to containerize the whole application. See if you find it useful in its current form.

[ Parent | on HN ]
Re:What's so hard about PDF text extraction? by hylian on Tuesday March 03, 2020 @09:33AM (#22474720)
I had a similar problem and ended up using AWS' Textract tool to return the text as well as bounding box data for each letter, then overlayed that on a UI with an SVG of the original page, allowing the user to highlight handwritten and typed text. I plan to open source it so if anyone's interested let me know.

Not a fan of the potential vendor lock in though, so it's only really suitable for those in an already AWS environment not worried about them harvesting your data.

[ Parent | on HN ]
Re:What's so hard about PDF text extraction? by rwojo on Tuesday March 03, 2020 @10:32AM (#22475231)
Very interested to see this as I was about to work on the same thing!
[ Parent | on HN ]