The latest push is working fine but when image is blury or have some noise it can't able to pass the image. it shows Detected 12 diacritics . The previous version was working fine with my images.
intallation
!apt-get install python-dev libxml2-dev libxslt1-dev antiword unrtf poppler-utils pstotext tesseract-ocr
!sudo apt-get install libenchant1c2a
!sudo add-apt-repository -y ppa:alex-p/tesseract-ocr-devel
!sudo apt-get update
# !sudo apt-get install tesseract-ocr
!sudo apt install -y tesseract-ocr tesseract-ocr-eng tesseract-ocr-osd
!fc-cache -vf
!apt-get install libtesseract-dev libleptonica-dev pkg-config
!pip install pytesseract pdf2image
!tesseract -v
download image
Tesseract cmd
!tesseract "
e182036ad97b14eb4b9db05d43484b0e.jpg " "out"
Tesseract Open Source OCR Engine v5.0.0-alpha-20201231-417-g2d805 with Leptonica
Estimating resolution as 845
Detected 12 diacritics
Result
out.txt file is empty. i also checked with other images. i think the latest push have some issues.
can you tell me how can i downgrade to last tesseract 5 version tesseract 5.0.0-alpha-20201231-246-gfe61.