Skip to main content

@Eugene Miller why are you in a bad place? I found that thread really interesting, btw!

I was considering suicide.

One of my clients wants me to build an OCR system. In order to pull this off I need to be able to take a picture of a document and then run it through an OCR program.

I've been working on the challenge of taking a picture of a document. I was stuck on how to detect corners of a page. The example code on that page was kinda broken. The end result was just a black image.

Thankfully someone in #opencv on freenode irc pointed me to https://www.pyimagesearch.com/2014/09/01/build-kick-ass-mobile-document-scanner-just-5-minutes/#

Image processing is a very difficult task for me. Hopefully I can figure out the rest from here.
# #opencv

@Eugene Miller That's exactly how I feel trying to apply ML tools in a couple of projects I'm working on, haha. I know this feeling well. It always feels kick ass when you overcome the problem, though!


It's amazing how ahead of its time that book was. Orwell was a genius!

...and welcome to social.linuxlusers.com! If you'd like you can add me get announcements and other info related to this node, like scheduled maintenance or updates.

Thanks for joining, and feel free to ask questions!

Oh, we also have a forum (that isn't that active currently, but hopefully that will change, haha) - @The Linux Luser Group




This website uses cookies. That said, social.linuxlusers.com only uses cookies/tracking within the constraints of what the Friendica project requires to function! No additional JavaScript or tracking code of any kind has been added to this site! We will never sell/release data without valid court order, and we only log what is needed to securely & legally operate the site.