2.80:OCR Synthesis (Concept)

From Grooper Wiki
Revision as of 13:51, 30 December 2019 by Configadmin (talk | contribs) (Created page with "The Synthesis functionality is Grooper's unique method of reprocessing results from the OCR engine. Raw character data is converted into a text flow. Words ar...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

The Synthesis functionality is Grooper's unique method of reprocessing results from the OCR engine. Raw character data is converted into a text flow. Words are simply groups of characters separated by white space as far as text data is concerned. While the individual character confidence is important, if the average confidence of a group of characters is low, it's likely that group of characters didn't get OCR'd as what is on the page. Synthesis allows users to group characters into segments separated by whitespace and reprocess them if their average character confidence falls below a certain threshold to ensure higher accuracy.