FROM my_ocrd_workflow-core ENV TESSDATA_BEST_VERSION 4.0.0 ENV TESSDATA_PREFIX /usr/local/share/tessdata # Install Leptonica and Tesseract. RUN add-apt-repository ppa:alex-p/tesseract-ocr && \ apt-get update && \ apt-get install -y \ tesseract-ocr \ libtesseract-dev \ && \ apt-get clean && rm -rf /var/lib/apt/lists/* # Copy OCR models RUN mkdir -p $TESSDATA_PREFIX ADD data/mirror/github.com/tesseract-ocr/tessdata_best/archive/${TESSDATA_BEST_VERSION}-repacked.tar.gz $TESSDATA_PREFIX/ COPY data/tesseract-models/GT4HistOCR/GT4HistOCR_2000000.traineddata $TESSDATA_PREFIX/ # Build pip installable stuff RUN pip3 install --no-cache-dir \ # Now the real stuff: 'ocrd_tesserocr >= 0.9.0' # Check pip dependencies RUN pip3 check # Default command CMD ["ocrd-tesserocr-recognize"]