0

I am working on tesseract. I want create new training language for tesseract. Please can anyone tell me what are the specific steps for training new language. And also tell me how to run text2image.cpp program. Thanks in advance.

user3479979
  • 92
  • 3
  • 11

1 Answers1

2

To run text2image first compile and link text2image.cpp using an appropriate C++ toolchain then run the executable with the appropriate text file as an input. Alternatively you can download a windows installer which will give you an executable to use rather than the .cpp.

Instructions on building the tesseract tool chain are here and on how to train for currently unsupported languages is here.

Steve Barnes
  • 27,618
  • 6
  • 63
  • 73
  • I compile it in Turbo c++. It gives errors like unable to read library. – user3479979 Apr 01 '14 at 05:22
  • At what stage, what exact error message, which library? – Steve Barnes Apr 01 '14 at 06:04
  • Error messages like "Unable to read include stdlib" "Unable to read include sting" "Unable to read include iostream" – user3479979 Apr 01 '14 at 06:23
  • What is your include path set to? Those are all standard libraries - also are you compiling as C or C++ you need to look into the compiler flags. Instructions on how to build https://code.google.com/p/tesseract-ocr/wiki/Compiling – Steve Barnes Apr 01 '14 at 06:31
  • Include path is "C:\TC\include" and "C:\TC\lib" and I am compiling it as cpp program. I followed documentation but it can't help me so much. Even I can't run tesseract command from CLI. – user3479979 Apr 01 '14 at 06:41
  • Can you tell me simplified steps to train new language in tsseract? I am using winn xp. And I have text file which I want to convert in ttf. – user3479979 Apr 01 '14 at 06:47
  • When using the windows installer, where is text2image.exe placed? – Zaz Aug 01 '14 at 08:49
  • The installer should offer you the chance to change it but the default will probably be `C:\Program Files` or `C:\Program Files (x86)\` – Steve Barnes Aug 01 '14 at 17:08