Hi everyone,
I have some how managed to get a working copy going (actually works in
windows after building in Linux) and am now at a point where i am
scanning things and they are producing some form of result. However I
do have a couple of issues I don't know how to resolve.
Basically my aim is to produce a way of allowing people to take a
picture of a number plate on their phone and it returning the
characters. Below are my problems;
1) Even though I have created a white list, I still get 'blacklisted'
characters in some of the feedback. I create the white list as
follows;
TessBaseAPI baseApi = new TessBaseAPI();
baseApi.setVariable(TessBaseAPI.VAR_CHAR_WHITELIST,
"1234567890ABCDEFGHJKLMNPRSTVWXYZ");
baseApi.init(myDir.toString(), "eng");
baseApi.setImage(img);
recognizedText = baseApi.getUTF8Text();
baseApi.end();
Are there any reasons you can think of why non white list characters
may be getting returned? If so how could i fix this?
2) I had a little read through the wiki blog for training the engine,
however I really was not sure on how I would apply this training for
number plates. The majority of UK number plates are of similar font
(although the front plate is white/black text and the back yellow/
black text) and I can constrain the user input to be in a certain
camera view port (so as to only get the majority of the plate and
nothing else) however I could see no way to train via images. Would it
be possible to get some assistance/advice on the best way to proceed
with this.
Thanks for your time.
Ricki
--
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
