I am sorry but I do not have have a crystal ball ;-)
Please provide necessary details (what version of OS you use, exact version
of tesseract, your compilation steps...)
Zdenko
On Wed, Jun 8, 2011 at 9:02 PM, ビ wrote:
> make all-recursive
> Making all in ccstruct
> /bin/sh ../libtool --tag=CXX
Hello!
I'm interested in system with minimum computational possibilities (the most
'slowest' CPU and minimum RAM) on which tesseract could still work.
The goal is to use it in minimized pc-platform device.
Maybe someone know another way to make minimized device with OCR capability.
BTW only digi
I'm trying to re-create ita traineddata with downloadable images. 1st
command, everything works fine:
unicharset_extractor -D ./train/452213 ./train/ita/ita.arial.box ./
train/ita/ita.arialbd.box ./train/ita/ita.arialbi.box ./train/ita/
ita.ariali.box ./train/ita/ita.b018012l.box ./train/ita
The GetTextDirection method does not work correctly. On a standard
page, it fails and returns false. However, if the line:
if (rows->length() != 1) {
return false;
}
to:
if (rows->length() < 1) {
return false;
}
in baseapi.cpp, then the method works as expected. Of course, this
method has
4 matches
Mail list logo