Extract image text (OCR)

Asked

Viewed 766 times

2

I wonder if anyone knows any library that can do the image reading and extract what is in it (can be in any language), or give me a north of what I must study to understand and develop something similar.

Observing: Very simple image with black and white text.

Observation2: I know this must be a little complex, but I’m willing to go for it.

  • 1

    Are you looking for an OCR? Would that be it? Vague question...

  • Extract text? Colors? Objects? Faces?

  • @Tássioauad this msm OCR. It is q when I asked the question did not know where to start and reading the suggestions vi that it is OCR

  • @Pabloalmeida The intection is to extract text, for example, photo shooting of some black and white text and the system captures what is written in the image.

  • Define a language, OCR can be done in R, JS, C, JAVA, PHP

  • @Marcelobonifazio believe that java or js are interesting because I want to create an android app with the function of recognizing the content (text) of a photo/image. You think viable?

  • 1

    Yes, the @Mateus answer is pertinent, search on this Tesseract library, there are versions in Java, JS and other languages as well, there is also a library that makes use of Tesseract: http://antimatter15.com/ocrad.js/demo.html

Show 2 more comments

1 answer

1


There is a library I’ve used in mobility projects called Tesseract OCR to recognize characters in a bitmap. This is an API that does character recognition.

However, to calibrate the library, more important than knowing how to use it, you need to learn how to "train" the Tesseract to recognize what you want in an image.

Browser other questions tagged

You are not signed in. Login or sign up in order to post.