Hey I am having the same question I have recently attempted the very same thing just wanted to know is it really possible?? n if yes how far have u reached
On Sunday, September 25, 2011 at 5:54:54 AM UTC+5:30, pottmi wrote: > > We are attempting to use Tesseract to do ICR on handwritten block > characters. > > We have been training using our own training data. > We are using a box around the characters to encourage users to write > in uniform manner. > > For our very first tests with our small training sample we are getting > 50% recognition. > We will expand our training samples and press on. > > My questions for group are: > 1) Is anyone successfully doing this in production? > > 2) How big of a training set should I expect to need; how may > different people contributing? > > 3) Are there any good techniques to programmatically remove the box so > that it does not interfere with the recognition? For experimentation > purposes we have been removing it by hand. > > 4) Does anyone have hand written block character training data that > they can share? > > 5) Is attempting to do ICR with tesseract like forcing a square peg in > a round hole and I should just find another solution? -- 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]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/7e6883b3-8ca7-4c6b-978b-94b1c32d64b6%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
