Tesseract OCR Pipeline

This software implements a heavily parallelized pipeline to recognize text in PDF files. It is used for nopaque's OCR service but you can also use it standalone, for that purpose a convenient wrapper script is provided.

Software used in this pipeline implementation

• Official Debian Docker image (buster-slim): https://hub.docker.com/_/debian
  • Software from Debian Buster's free repositories
• Leptonica Library (1.83.1): https://github.com/DanBloomberg/leptonica/releases/tag/1.83.1
• ocropy (1.3.3): https://github.com/ocropus-archive/DUP-ocropy/releases/tag/v1.3.3
• pyFlow (1.1.20): https://github.com/Illumina/pyflow/releases/tag/v1.1.20
• Tesseract OCR (5.3.3): https://github.com/tesseract-ocr/tesseract/releases/tag/5.3.3

Installation

1. Install Docker and Python 3.
2. Clone this repository: git clone https://gitlab.ub.uni-bielefeld.de/sfb1288inf/tesseract-ocr-pipeline.git
3. Build the Docker image: docker build -t tesseract-ocr-pipeline:latest tesseract-ocr-pipeline
4. Add the wrapper script (wrapper/tesseract-ocr-pipeline relative to this README file) to your ${PATH}.
5. Create working directories for the pipeline: mkdir -p /<my_data_location>/{input,models,output}.
6. Place your Tesseract OCR model(s) inside /<my_data_location>/models.

Use the Pipeline

1. Place your PDF files inside /<my_data_location>/input. Files should all contain text of the same language.
2. Clear your /<my_data_location>/output directory.
3. Start the pipeline process. Check the pipeline help (tesseract-ocr-pipeline --help) for more details.

cd /<my_data_location>
# <model_code> is the model filename without the ".traineddata" suffix
tesseract-ocr-pipeline \
  --input-dir input \
  --output-dir output \
  --model-file models/<model>
  -m <model_code> <optional_pipeline_arguments>
# More then one model
tesseract-ocr-pipeline \
  --input-dir input \
  --output-dir output \
  --model-file models/<model1>
  --model-file models/<model2>
  -m <model1_code>+<model2_code> <optional_pipeline_arguments>
# Instead of multiple --model-file statements, you can also use
tesseract-ocr-pipeline \
  --input-dir input \
  --output-dir output \
  --model-file models/*
  -m <model1_code>+<model2_code> <optional_pipeline_arguments>

4. Check your results in the /<my_data_location>/output directory.