PyOCR - get_availables_tools() returns an empty list / Can access tesseract from the command line

Question

I am attempting to install PyOCR on my computer running Windows 7. I have installed Tesseract-OCR 3.05 for Windows and have added the directory containing Tesseract (C:\Program Files (x86)\Tesseract-OCR) to the PATH user variable, the Path System Variable, and created the new system variable TESSDATA_PREFIX with the Tesseract directory.

I am able to use Tesseract directly from the command line to process images, so I am confident that Tesseract was correctly installed. I also made sure to installed Tesseract with the C/C++ library files.

I know this question has been posed before, but since I have added the directories to the environmental variables, I am unsure what to try next.

Below is the output of the "get_available_tools()" method.

Python 3.6.3 |Anaconda custom (64-bit)| (default, Oct 15 2017, 03:27:45) 
[MSC v.1900 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import pyocr
>>> import pyocr.builders
>>> pyocr.get_available_tools()
[]
>>>

Pawel Kam · Accepted Answer · 2019-03-16 16:11:29Z

3

get_available_tools() returns a list of ‘OCR tools available on the local system’ (see source code). As I understand, pyocr checks in PATH for available tools, which you have to install yourself.

For now, Pyocr supports the following OCR tools:

Libtesseract (Python bindings for the C API)
Tesseract (wrapper: fork + exec)
Cuneiform (wrapper: fork + exec)

If you have installed such tools, but for some reason you don’t have them in your local path (like me), you can always overwrite it like this:

pyocr.tesseract.TESSERACT_CMD = r'<full_path_to_your_tesseract_executable>'
pyocr.cuneiform.CUNEIFORM_CMD = r'<full_path_to_your_cuneiform_executable>'

answered Mar 16, 2019 at 16:11

Pawel Kam

2,1823 gold badges18 silver badges34 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

sekrett Over a year ago

Thanks. TESSERACT_CMD solved the issue. Just curious, what is Libtesseract and how to install it?

Pawel Kam Over a year ago

Libtesseract is an OCR engine written in C. I don't have any experience with this one, but for PyOCR I guess you could start with Python wrapper like this one github.com/virtuald/python-tesseract-sip.

knobi · Accepted Answer · 2018-12-06 10:11:25Z

1

Had a similar problem using libtesseract-4.dll: Tools list was empty.

As i discovered pyocr is using get_version which returned 0.0.0 . As a Workaround i patched get_version on my installation to give back the appropriate value.

This worked because the parameter passing especially the -psm / --psm parameter depends on the result of get_version().

The issue was accepted as bug and seems to be fixed by the author soon. https://gitlab.gnome.org/World/OpenPaperwork/pyocr/issues/106

answered Dec 6, 2018 at 10:11

knobi

9521 gold badge7 silver badges7 bronze badges

Collectives™ on Stack Overflow

PyOCR - get_availables_tools() returns an empty list / Can access tesseract from the command line

2 Answers 2

2 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related