Mauricio Tavares <raubvo...@gmail.com> writes: > On Wed, Dec 11, 2019 at 2:05 PM Giovanni Bechis <giova...@paclan.it> wrote: >> >> On 12/11/19 8:00 PM, Mauricio Tavares wrote: >> >> > I asked the project owner if I could put fuzzyocr on github. He said >> > go for it, so it is now at https://github.com/raubvogel/FuzzyOcr. >> > >> Cool, >> you can grab my patches (if they are needed) here: >> http://cvsweb.openbsd.org/ports/mail/p5-FuzzyOcr/patches/ >> > Will do! But, I do want to ensure your work is acknowledged; if > I wanted to claim someone else's work as mine I would be into > politics. ;)
That sounds like a very good news that you are taking back the maintenance of FuzzyOcr! Among the future development, I have been playing with the idea of FuzzyOcr pushing back the decoded text to SA for further analysis rather than using a local list of words. That is the way PdfAssassin is working, text is pushed as a text block and image as an image attachment (so they can be further processed by any image plugin like FuzzyOcr). If that could work, that would be a very great improvement to FuzzyOcr. Best regards, Olivier --
