python-tesseract not returning same results as command line tesseract
GoogleCodeExporter opened this issue · comments
What steps will reproduce the problem?
1. Perform OCR on the attached image with command line tesseract (tesseract
3.png 3 && more 3.txt)
2. Perform OCR on the attached image with python-tesseract
3. Notice the errors thrown from python-tesseract
What is the expected output? What do you see instead?
On certain images the results I'm seeing from python-tesseract are not exactly
same as the results from command line tesseract on the same image.
Additionally, some images (like the one installed) throw errors when
python-tesseract performs OCR on them. Even though the result is same for this
image from command line tesseract and python-tesseract.
In my script this option returns False "print
"HAVE_LIBLEPT=",tesseract.isLibLept()" I am assuming maybe this is why the
results are different?
What version of the product are you using? On what operating system?
$ tesseract --version
tesseract 3.02.02
leptonica-1.69
libjpeg 8d : libpng 1.5.14 : libtiff 4.0.3 : zlib 1.2.5
Please provide any additional information below.
I have searched google for this but have not found anything. My question is
simply that should tesseract.isLibLept() be returning True for python-tesseract
to return exactly the same results as the command line?
My script is like this:
api = tesseract.TessBaseAPI()
api.SetOutputName("outputName");
print "HAVE_LIBLEPT=",tesseract.isLibLept()
api.Init(".","eng",tesseract.OEM_DEFAULT)
api.SetPageSegMode(tesseract.PSM_AUTO)
Original issue reported on code.google.com by Bhaara...@gmail.com
on 14 Oct 2013 at 3:33
Attachments:
What OS are you using?
Original comment by FreeT...@gmail.com
on 10 Nov 2013 at 3:38
OSX
Original comment by Bhaara...@gmail.com
on 10 Nov 2013 at 3:52
Please confirm whether the problem have been fixed?
Original comment by FreeT...@gmail.com
on 28 Nov 2013 at 6:28
Original comment by FreeT...@gmail.com
on 11 Jun 2014 at 8:37
- Changed state: Fixed