Hello,
I am a new member of this mailing list. I am creating a small project to read
electronic screens through OCR. In other words, we set up some equipment that
capture
the VGA output of computers and other devices, and converts the signal to RCA
composite, so I can take pictures and videos of the machine. My idea is to
capture
BIOS images (Setup) and convert them to text that can be read by visually
impaired users, like me. The fact is that Tesseract does not seem to understand
my images,
but other commercial OCRs can read almost 99% of the text, with amazing
accuracy. Before purchasing any license these OCRs, I wonder if there is
anything I can
do
to make the tesseract is able to read my screens with some precision. I've
tried to make the tesseract tessinput.tif return the file, and the result is a
1KB file
with bad picture quality, with completely blurred and distorted letters. I
believe this happens because the tesseract tries to improve the image
internally, but
ends up destroying it. If there was any option for tesseract not modify the
image, certainly I could best results. Even in commercial OCRs, I can choose
whether
you want the image to be converted to black and white.
Does anyone have any idea how I can do this by tesseract? I thought even in
trying to recompile the tesseract in a way that it does not alter the original
image.
Or is there some binary for Windows or Linux you already have this capability?
Very grateful.
Sincerely,
Lucas Alexandre
--
You received this message because you are subscribed to the Google Groups
"tesseract-ocr" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/tesseract-ocr.
To view this discussion on the web visit
https://groups.google.com/d/msgid/tesseract-ocr/57b75ef6.9a35ed0a.27a7a.1bb5%40mx.google.com.
For more options, visit https://groups.google.com/d/optout.
