Anonymous ID: ff887b Feb. 12, 2022, 8:19 p.m. No.15615067   🗄️.is 🔗kun   >>5072 >>5100

>>15614990

>>15615020

TY anon and B!

 

To explain why I saidfinding images with textin my heading - that was my goal; to find the images I had, which had text in them, without looking at them all.

Tesseract helps me with exactly that.

Apologies for any confusion.

You are also, absolutely correct; the job tesseract-ocr does is to extract the text from the images.

My need was e.g., "where's the image that says 'Back to the Future' in it, maybe with some other text" etc.

So now, once I've written the script (I plan to do it in Python, getting ready to start it this evening I think) I will run it on all the images, so for "Screenshot.png" it'll write "Screenshot.txt" with any text that Tesseract finds in the image.

Then I can put that script in crontab, and have it run every hour or once a day; I can run it on-demand if I need to.

Will add the script here once done.

Thanks again!

Anonymous ID: ff887b Feb. 12, 2022, 8:21 p.m. No.15615077   🗄️.is 🔗kun   >>5087

>>15615072

>is tesseract linux only?

 

No see:

>>15614990

>^Should read; extracting text FROM images

>FYI Tesseract-OCR is for WindowsFags through Cygwin or MSYS2 and macOSFags through MacPorts or Homebrew too.

>https://tesseract-ocr.github.io/tessdoc/Installation.html