ocrd-segment 0.1.2

Page segmentation and segmentation evaluation

ocrd_segment

This repository aims to provide a number of OCR-D-compliant processors for layout analysis and evaluation.

Installation

In your virtual environment, run:

pip install .

Usage

• exporting page images (including results from preprocessing like cropping/masking, deskewing, dewarping or binarization) along with region polygon coordinates and metadata, also MS-COCO:
  • ocrd-segment-extract-pages
• exporting region images (including results from preprocessing like cropping/masking, deskewing, dewarping or binarization) along with region polygon coordinates and metadata:
  • ocrd-segment-extract-regions
• exporting line images (including results from preprocessing like cropping/masking, deskewing, dewarping or binarization) along with line polygon coordinates and metadata:
  • ocrd-segment-extract-lines
• importing layout segmentations from other formats (mask images, MS-COCO JSON annotation):
  • ocrd-segment-from-masks
  • ocrd-segment-from-coco
• repairing layout segmentations (input file groups N >= 1, based on heuristics implemented using Shapely):
  • ocrd-segment-repair :construction: (much to be done)
• comparing different layout segmentations (input file groups N = 2, compute the distance between two segmentations, e.g. automatic vs. manual):
  • ocrd-segment-evaluate :construction: (very early stage)
• pattern-based segmentation (input file groups N=1, based on a PAGE template, e.g. from Aletheia, and some XSLT or Python to apply it to the input file group)
  • ocrd-segment-via-template :construction: (unpublished)
• data-driven segmentation (input file groups N=1, based on a statistical model, e.g. Neural Network)
  • ocrd-segment-via-model :construction: (unpublished)

For detailed description on input/output and parameters, see ocrd-tool.json

Testing

None yet.