Improve tika for scan documents in PDF formats - Githubissues

RichJackson / cogstack

Database - Elasticsearch realtime mapping. With NLP goodiness.

Apache License 2.0

7 stars 2 forks source link

Improve tika for scan documents in PDF formats #18

Closed hkkenneth closed 7 years ago

hkkenneth commented 8 years ago

Use tika-config.xml explicitly
Store original text parsed from PDF (before OCR) in metadata
Change image magick command (which increase successful parsing rate)