After image pre-processing, the recognition areas have to be defined.

Developers can do this manually or choose from 3 different modes for automatic document analysis (DA):

• Full text DA
  recognises all text on documents, including text embedded in pictures, charts and diagrams
• DA with layout retention
  brings auto-detection of blocks, tables, barcodes, and pictures
• Invoice pre-processing DA
  with a focus on numbers and tables
• Manual block creation
  is mostly used in Field Level/Zonal Recognition scenarios

… more to come

Back to: OCR Processing Steps