Known OCR, ML, NLP Issues

From iDigBio
Revision as of 22:39, 10 January 2013 by Dpaul (Talk | contribs)

Jump to: navigation, search

Specific Issues Needing Work

  1. how to get OCR to ignore a map (reduce OCR confusion)
  2. ... and ___ present a challenge and confuse OCR and parsing.
  3. figure out an algorithm that would separate images into sets with no handwriting, little handwriting (mostly text typed or printed), lots of handwriting