CLI

CLI

Mockup usage

$> ocrd --help

Global options:
  --model-repository=URL	   Set the model repository endpoint
  --processing-repository=URL  Set the endpoint for the repository serving intermediary results

Subcommands
  characterize		Extract intrinsic properties from an image
  binarize  		Binarise image(s)
  despeckle 		Despeckle image(s)
  dewarp    		Dewarp image(s)
  deskew    		Deskew image(s)
  split      		Split double pages
  detect-regions    Detect regions
  classify-regions  Tag regions with their typographic functionality
  detect-reading-order Detect the reading order of regions within the page(s)
  detect-lines      Detect lines within regions
  merge             Improve OCR accuracy by combining multiple results

Help for a specific subcommand:

$ ocrd help binarize

Implementations:
  binarize/sauvola		Binarize with Sauvola (2000)
  binarize/wolf         Binarize with Wolf (2002)

Help for specific implementation:

$> ocrd help binarize/sauvola

Options:
  --threshold=THRESHOLD Threshold, default: 0.04

Examples

Characterize the image(s) defined in the fileGrp of /path/to/mets.xml with the exiftool implementatio

ocrd characterize/exiftool /path/to/mets.xml

Binarize the image(s) defined in /path/to/mets.xml with the default implementation for binarization.

ocrd binarize /path/to/mets.xml

Binarize the image(s) defined in /path/to/mets.xml with the ocropy-nlbin implementation for binarization.

ocrd binarize/ocropus-nlbin /path/to/mets.xml

Run the actual OCR on mets.xml

ocrd recognize /path/to/mets.xml