No description
Find a file
2020-08-03 12:46:42 +02:00
qurator 🔧 ocrd-tool.json: Update description, steps and categories 2020-05-29 17:30:53 +02:00
.gitignore Revert "Merge branch 'master' of https://github.com/qurator-spk/sbb_textline_detector" 2019-12-09 15:11:25 +01:00
.gitkeep 🧹 sbb_textline_docker: Rename to sbb_textline_detector 2019-10-10 16:13:07 +02:00
Dockerfile 🧹 sbb_textline_detector: Use same structure as the other projects 2019-10-10 16:24:28 +02:00
LICENSE Revert "Merge branch 'master' of https://github.com/qurator-spk/sbb_textline_detector" 2019-12-09 15:11:25 +01:00
ocrd-tool.json sbb_textline_detector: Add a OCR-D interface 2019-10-10 17:54:42 +02:00
README.md Update README.md 2020-08-03 12:46:42 +02:00
requirements.txt Enforce older version of Keras which does not use TensorFlow 2 2020-06-26 21:58:14 +02:00
setup.py Revert "Merge branch 'master' of https://github.com/qurator-spk/sbb_textline_detector" 2019-12-09 15:11:25 +01:00

Textline Detection

Detect textlines in document images

Introduction

This tool performs printspace, region and textline detection from document image data and returns the results as PAGE-XML. The goal of this project is to extract textlines of a document to feed an ocr model. This is achieved by four successive stages as follows:

  • Printspace or border extraction
  • Layout analysis
  • Textline detection
  • Heuristic methods

Installation

pip install .

Models

In order to run this tool you also need trained models. You can download our pretrained models from here:
https://qurator-data.de/sbb_textline_detector/

Usage

sbb_textline_detector -i <image file name> -o <directory to write output xml> -m <directory of models>

Usage with OCR-D

ocrd-example-binarize -I OCR-D-IMG -O OCR-D-IMG-BIN
ocrd-sbb-textline-detector -I OCR-D-IMG-BIN -O OCR-D-SEG-LINE-SBB \
        -p '{ "model": "/path/to/the/models/textline_detection" }'

Segmentation works on raw RGB images, but retains AlternativeImages from binarization steps, so it's OK to do binarization first, then perform the textline detection. The used binarization processor must produce an AlternativeImage for the binarized image, not replace the original raw RGB image.