Dear all,
I am trying to do some improvement for the text recognized by Tesseract OCR.
For some low quality picture,some letters may be mis-recognized with some similar letters,
So I want to get the candidate letter for them to do some improvement, I have look through
the source ode built in Visual studio 2008, It is difficult to bring out the candidate results .
Is there other way to get the candidate?
Any help would be appreciated !