hi all,
i compiled tesseract from git yesterday and played with it a little
bit.
pretty impressive what happened since around 2 years.
not only has tesseract a lower filesize but it seems its also faster
and more accurate.
But to the topic of this message:
I used the following command to read from a tif file and convert it
into a searchable pdf:
$ tesseract image.tif -l eng image pdf
which resulted into a text file with the very accurate ocr result and
a pdf.
when i opened the pdf with adobe reader i got an error that something
is wrong with the pdf file i just created.
the pdf was also large around 1 mb. a lot for just a little bit of
text.
am i doing something wrong there?
and i have a feature request as well.
more and more copiers/scanners output in multipage pdf which contain
images of the pages scanned.
is that planned to be included as well to be used as input file?
greetings,
simon
--
Simon Eigeldinger
simon.ei...@vol.at