[tesseract-ocr] Training OCR

[MedCo]

Hello,

I need to get text out of small bitmap files. I used tesseract for it and
it works ok. The problem is that the text we have is not generic, it's
comprised of different upper case, lower case characters and some special
symbols.
This will need lot of OCR training.

When I run it to recognize images, that will need to be on through a dll or
command prompt so my automated script can feed in the image and get text
out of it. Tesseract functionality works good through command prompt, but
training part seems very challenging.

Is there any OCR available which is easy to train, may be training with
some GUI interface?

thanks,

[Quan Nguyen]

There are several training tools available:

https://code.google.com/p/tesseract-ocr/wiki/AddOns

Post by MedCo
Hello,
I need to get text out of small bitmap files. I used tesseract for it and
it works ok. The problem is that the text we have is not generic, it's
comprised of different upper case, lower case characters and some special
symbols.
This will need lot of OCR training.
When I run it to recognize images, that will need to be on through a dll
or command prompt so my automated script can feed in the image and get text
out of it. Tesseract functionality works good through command prompt, but
training part seems very challenging.
Is there any OCR available which is easy to train, may be training with
some GUI interface?
thanks,