From 9bd3853c78fc8870d51b4d84a9c0218e9d0f186c Mon Sep 17 00:00:00 2001
From: "Gerber, Mike" <mike.gerber@sbb.spk-berlin.de>
Date: Wed, 19 Jun 2019 13:02:54 +0200
Subject: [PATCH] Add OCR step

---
 my_ocrd_workflow | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/my_ocrd_workflow b/my_ocrd_workflow
index 3e4484f..498446f 100755
--- a/my_ocrd_workflow
+++ b/my_ocrd_workflow
@@ -38,9 +38,18 @@ do_linesegmentation() {
   #ocrd workspace validate mets.xml
 }
 
+do_ocr() {
+  ocrd_tesserocr_recognize_parameters='{ "model": "eng" }'  # TODO mods:language + fontident → model
+  remove_filegrp OCR-D-OCR-TESS mets.xml
+  ocrd-tesserocr-recognize -l DEBUG \
+    -m mets.xml -I OCR-D-SEG-LINE -O OCR-D-OCR-TESS \
+    -p <(echo $ocrd_tesserocr_recognize_parameters)
+  #ocrd workspace validate mets.xml
+}
 
 do_fontident
 do_linesegmentation
+do_ocr
 
 
 # XXX Multiple calls create multiple identical mets:agent elements