This is a pretty insightful piece by OCR/data capture consultant Chris Riley of Living Analytics. One additional comment I’ll make is that the scanning hardware vendors continue to introduce more versatile machines that address some of the demands that Chris mentions in his article. For example, Kodak has come out with a whole new series of scanners that capture in 300 dpi at the same speed as 200 dpi, and color speeds have almost caught up bi-tonal speeds.
We’ve also had many interesting conversations with vendors that specialize in the separation technology that Chris mentions and claim it can introduce OCR results on certain types of documents. As we’ve said before, we are entering a brave new world for recognition technologies.
Fujitsu Computer Products of America’s Kevin Neal contributed these two helpful links on the same topic: