Geek Answers Handbook

Tesseract or any other OCR lib

I am looking for an explanation / API doc / examples of using (and learning?) Tesseract in C ++, nothing useful on the Google Tesseract page and still find something on the Internet.

Any useful sources, experience would be more than welcome, since I do not know how to start with it.

PS:

  • I am open to suggestions for other libraries.
  • FREE libraries only
+7
c ++ ocr image-recognition tesseract
source share
2 answers

I have experience with Tesseract ... a simple google "tesseract" shows this page: http://code.google.com/p/tesseract-ocr/wiki/TrainingTesseract where you have to choose the version of tesseract you want to train. While 3 is the latest version, it is completely new, and therefore people are still hiding any problems - I am still using version 2.4. In any case, you will see about 9 steps in learning tesseract for a particular “language” (or what should be called “fonts” or “character sets”). You can also use the existing 'eng' language, but it depends on your application. For example, in my application, I would need to analyze a document and take a specific region and want an OCR 13-character string of numbers - and I needed high precision - and I didn’t want it to read “5”, S 'and' 0 ' like 'O', etc., so it was logical to create a specific “language” for my particular font set consisting of only 0..9 characters, whereas it might not matter to you if you get extra “noise”

+4
source share

Tesseract Ocr is an open source library for optical character detection. You just need to include library files if you are using visual studio. If you are using qt creator, you need to build a library to work with QT. To create a library you need to use CMakelist or Cmake Gui. You can visit the Opencv Ocr build link for Qt 5.4 mingw

0
source share

More articles:


All Articles