windows 64bit, 'tesseract.exe has stopped working'
Closed this issue · 5 comments
Environment
- Tesseract Version: tesseract-ocr-w32-setup-v5.0.0-alpha.20191010.exe
- Commit Number:
- Platform: windows 7 64bit, windows2008 64bit
Current Behavior:
Run the following command:
tesseract c:\t\wz123.jpg c:\t\wz.txt
All physical machines run successfully.
Successfully executed under Windows 10 (VMware), failed to execute in Windows 7 (VMware), windows 2008 (VMware, 2 virtual machines), and the error information is the same:
tesseract.exe has stopped working.
Problem detail:
Problem signature:
Problem Event Name: APPCRASH
Application Name: tesseract.exe
Application Version: 0.0.0.0
Application Timestamp: 5db94754
Fault Module Name: libtesseract-5.dll
Fault Module Version: 0.0.0.0
Fault Module Timestamp: 5db94750
Exception Code: c000001d
Exception Offset: 001ffa09
OS Version: 6.1.7600.2.0.0.256.4
Locale ID: 2052
Additional Information 1: 0a9e
Additional Information 2: 0a9e372d3b4ad19135b953a78882e789
Additional Information 3: 0a9e
Additional Information 4: 0a9e372d3b4ad19135b953a78882e789
Expected Behavior:
Hope to generate OCR results
Suggested Fix:
Is there a lack of dependency? Or how can I run correctly to get OCR results?
Thank you!
What is the output from tesseract -v
on the machines where it is not working?
The output is as follows:
C:\Program Files (x86)\Tesseract-OCR>tesseract -v
tesseract v5.0.0-alpha.20191010
leptonica-1.78.0
libgif 5.1.4 : libjpeg 8d (libjpeg-turbo 1.5.3) : libpng 1.6.34 : libtiff 4.0.
9 : zlib 1.2.11 : libwebp 0.6.1 : libopenjp2 2.3.0
Found AVX2
Found AVX
Found FMA
Found SSE
Found libarchive 3.3.2 zlib/1.2.11 liblzma/5.2.3 bz2lib/1.0.6 liblz4/1.7.5
Please try running Tesseract with parameter -c dotproduct=sse
.
Thank you very much for your reply. There will be no error in the running process.
However, OCR results still have some problems. The test process is as follows:
- The content in wz123.jpg is:
My Name is houji
My Name is houji
My Name is houji
My Name is houji
- Test 1:
tesseract c:\t\wz123.jpg c:\t\wz.txt -c dotproduct=sse
Results:
§Q$»&«Y«/»«&«&«Y/»M»§L_L&_&X°Qj‘\Q7“(“‘&°7«¥/
Z§7#&_&2?‘7?«0“Y$»/»p§$§/»}«p®2§«»J,¥»2§VA
Q2%&H?&«™2é4_«9Y©&x«2N§«§€»«§»§»§«&§
/«3°«»«»«2»#Swx°“‘V#»(’2%]2X/3¥/&/#=(2#=4»S
- Test 2(use
tessdata_best--eng.traineddata
):
tesseract c:\t\wz123.jpg c:\t\wz.txt -c dotproduct=sse
Results:
My Name is houji
My Name is houji
My Name is houji
My Name is houji
- Test3(use
tessdata_fast--chi_sim.traineddata
):
tesseract -l chi_sim c:\t\wz123.jpg c:\t\wz.txt -c dotproduct=sse
Results:
M^fUM0^MqM'?kw‰\'_〈q"※″_{〈@′※@_※E※E〈E‰(E〈E=〇〗″'M〇'′「E〗「“
M'〉?「w'K「GQ_'?!V、《〗=〇V「“〗'〗!〇,′〉V〇′″」「※〇w′」zQ″「Gw'
x〗〗〇」「'\「※i“M「dMi」〇」〇'「'《′‖′〉M′。″M〗'2'…^!^
w'p「《〔「〔※「'「]″〗〇〉″〗″w※w'w'?'〗?w〗〇『^y〇″y
- Test 4(use
tessdata_best--chi_sim.traineddata
):
tesseract -l chi_sim c:\t\wz123.jpg c:\t\wz.txt -c dotproduct=sse
Results:
My Name is houji
My Name is houji
My Name is houji
My Name is houji
Whether it is possible to further optimize by setting parameters to achieve appropriate results by using tessdata_fast--chi_sim.traineddata
, thank you!
I saw the '-c dotproduct=' instruction information from tesseract-ocr#2098 today, the problem has been solved, thank you!