On 12/15/2015 03:00 PM, Tom Horsley wrote:
If you have pdf files with actual characters, the
pdftotext tool works well for extracting the text
(though not necessarily the layout).
there is an option: -layout
It does a good job with preserving the layout.
David
As far as doing OCR from actual image files,
I always found tesseract to work better than most
(but it was still pretty feeble).