Hi all, I am using Tesseract 3.02 and have created a lang.unicharambigs file for my data. It seems to work OK when I have more than one character on either side of the "equation", but otherwise doesn't have any affect on single-character substitutions (t -> r, or vice versa). I've been seeing a lot of references in the discussions in this group to DangAmbigs, but I'm not sure what that is or how it's different from unicharambigs. I have seen some examples of the syntax, but no clear indication of whether this is a stand-alone file, what it should be named, and where it goes. I'm assuming that it's a standalone file that would be named something like lang.DangAmbigs and go in the same folder as my DAWG files so that it can be included when it run combine_tessdata to create my language. Can anyone confirm that?
What I'd really like to know is how these two files are different? Is DanAmbigs just a unicharambigs that's used during dictionary lookup somehow, or is it an older version of the unicharambigs from earlier tesseracts? Maybe this will answer another question I have about unicharambigs: is unicharambigs consulted/processed before the dictionary lookups or after? I'm going to try an experiment to create a unicharmbigs file for my data that will turn all ligatures into their modern two-character equivalents since my dictionary doesn't contain any ligatures or long-Ses. Maybe that will answer my question. But I'd appreciate any comments on this from anyone who knows as well. I'll post my results to this thread. Thanks, Matt -- -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/tesseract-ocr?hl=en --- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
