Regions (columns) on ocr recognition

Asked by pfister

Is there any timetable available when layout detection (regions, columns) will work?

Frankly spoken, I find the cuneiform linux port a great project, but I miss layout detection. As I offered in an earlier mail, if I can sponser the development I would be ready to do.

Urs

Question information

Language:
English Edit question
Status:
Solved
For:
Cuneiform for Linux Edit question
Assignee:
No assignee Edit question
Last query:
Last reply:
Revision history for this message
Jussi Pakkanen (jpakkane) said :
#1

I have done some very quick & dirty checks, and the layout code seems to work (where "work" means "it does not crash horribly"). The missing link is outputting the recognition result in a format that contains layout information. The easiest path is to make the HTML output more hOCR compatible.

Currently I'm working on porting the code to 64 bit. Once this is done I'll release version 0.3. The layout thing has to wait until then. If someone else wants to do it, I'll be glad to accept patches.