I would hazard to guess this is a pretty Tough Problem and will need a
little bit more in depth approach than Tesseract's current scope.
Something like a separate text-detector well generalized to finding text in
local area that would be robust against the angle variation finding the
text,
I have added it as an issue. Please see
https://github.com/tesseract-ocr/tesseract/issues/754
You may want to create a pull request, if you have a solution.
ShreeDevi
भजन - कीर्तन - आरती @ http://bhajans.ramparivar.com
On Sun, Mar 5,
2 matches
Mail list logo
