Tesseract - Forum

#1 2015-01-17 02:22:02

littlejoe5

Member

Tesseract seems to ignore very short word,or expressions.

In some documents that I am trying to recover, there are very short abreviations that include (for example) a letter, and a number. eg: m1, or z11.   These are esential to an understanding of the documents. Tesseract seems to completely iognore them if they are shorter than 3 or 4 charachters.  On those that are slightly longer, it ios near 100 percent accurate.  Is ther a setting that will allow tesseract to payt attention to the3se short abreviations?

littlejoe5

#2 2015-01-17 13:54:59

RaZgRiZ

Moderator

Re: Tesseract seems to ignore very short word,or expressions.

I have absolutely no idea what you're talking about. Also i hope this isn't about Tesseract-OCR because this is a forum for a game engine.

#3 2015-01-19 02:17:05

littlejoe5

Member

Re: Tesseract seems to ignore very short word,or expressions.

Oh! It IS about Tesseract-OCR. I didn't realise that there was a game by that name. Sorry. Any idea where I can get help with     Tesseract-OCR ?

Last edited by littlejoe5 (2015-01-19 02:17:39)