Software for scanning of Indian rarities?

Mon Mar 16 20:45:27 UTC 1998

OCR for Indic scripts can be developed in stages.
As a first step, OCR can be developed for Tamil.
Because Tamil has a minimal set of consonants,
- no separate symbols for voiced consonants -
it will be far more easier to develop OCR for Tamil.
In addition, tamil has no cluster letters in orthography.

So, with the least number of symbols to be recognized
by OCR, Tamil will be a natural choice to develop and test
OCR software. For the same reason, Printing and Typewriting
came to Tamil far earlier than other Indian languages.

Once Tamil  OCR is field-tested, Nagari OCR will be easier
to develop.

N. Ganesan

More information about the INDOLOGY mailing list