Sure
On Tue, Mar 18, 2025, 03:12 Tom Morris wrote:
> Mixotricha wrote on a separate thread:
>
> I had a thought that the vectors will probably be reasonable sized units.
> 5,10,15 and so on. If I was writing this Fortran that is probably what I
> would do. And then if I came back I might add sm
Mixotricha wrote on a separate thread:
I had a thought that the vectors will probably be reasonable sized units.
5,10,15 and so on. If I was writing this Fortran that is probably what I
would do. And then if I came back I might add smaller units between.
Context helps.
Yes, this was standar
Howdy Mixotricha,
I just happened upon your post and thought that I would share this
playlist, as it is a deep dive into a lot of the complexities of OCR.
Preprocessing is a major thing for getting optimal OCR results, that is why
I put the video title in Bold for it below.
OCR in Python
htt
On Tue, Feb 11, 2025 at 5:52 PM Mixotricha
wrote:
> Thanks that is a really helpful link. Unfortunately I do not have much
> chance of getting better documents. The second scan came from a helpful
> archivist at an installation that requires a classification to enter.
> Otherwise I would literall
Thanks that is a really helpful link. Unfortunately I do not have much
chance of getting better documents. The second scan came from a helpful
archivist at an installation that requires a classification to enter.
Otherwise I would literally get on a plane and go and look myself. I was
gratified
I can't help with tesseract advice - when I wanted to do the same thing I
found it easier to write a custom OCR for this specific problem from
scratch. It's very much an experiment and a work-in-progress (although
I've not worked on it for about a year I'm afraid) but you might find
something help
6 matches
Mail list logo
