# ocrd_segment
This repository aims to provide a number of [OCR-D](https://ocr-d.de) [compliant](https://ocr-d.de/en/spec) [processors](https://ocr-d.de/en/spec/cli) for layout analysis and evaluation.
[![CircleCI](https://dl.circleci.com/status-badge/img/gh/OCR-D/ocrd_segment/tree/master.svg?style=svg)](https://dl.circleci.com/status-badge/redirect/gh/OCR-D/ocrd_segment/tree/master)
[![image](https://img.shields.io/pypi/v/ocrd_segment.svg)](https://pypi.org/project/ocrd_segment/)
[![Docker Automated build](https://img.shields.io/docker/automated/ocrd/segment.svg)](https://hub.docker.com/r/ocrd/segment/tags/)
## Installation
In your [Python virtual environment](https://packaging.python.org/guides/installing-using-pip-and-virtual-environments/), run:
pip install ocrd_segment
## Usage
Contains processors for various tasks:
- exporting segment images (including results from preprocessing like cropping/masking, deskewing, dewarping or binarization) along with polygon coordinates and metadata:
- [ocrd-segment-extract-pages](ocrd_segment/extract_pages.py) (for pages, also exports [MS-COCO](https://cocodataset.org/) format and pageview plots)
- [ocrd-segment-extract-regions](ocrd_segment/extract_regions.py) (for regions, so exports [MS-COCO](https://cocodataset.org/) format))
- [ocrd-segment-extract-lines](ocrd_segment/extract_lines.py) (for lines, also exports text and .xlsx)
- [ocrd-segment-extract-words](ocrd_segment/extract_words.py) (for words, also exports text)
- [ocrd-segment-extract-glyphs](ocrd_segment/extract_glyphs.py) (for glyphs, also exports text)
- importing layout segmentations from other formats:
- [ocrd-segment-from-masks](ocrd_segment/import_image_segmentation.py) (for mask/label images, i.e. semantic segmentation)
- [ocrd-segment-from-coco](ocrd_segment/import_coco_segmentation.py) (for [MS-COCO](https://cocodataset.org/) annotation)
- post-processing or repairing layout segmentations:
- [ocrd-segment-repair](ocrd_segment/repair.py) (validity and consistency of all coordinates; also, for regions, reduce overlaps/redundancy between neighbours, and/or simplify polygons, and/or shrink to the alpha shape of foreground contours)
- [ocrd-segment-project](ocrd_segment/project.py) (remake segment coordinates into the concave hull / alpha shape of their constituents)
- [ocrd-segment-replace-original](ocrd_segment/replace_original.py) (rebase all segments on cropped+deskewed border frame as new full page)
- [ocrd-segment-replace-page](ocrd_segment/replace_page.py) (2 input fileGrps; overwrite segmentation below page of first fileGrp by all segments of second fileGrp, rebasing all coordinates; "inverse" of `replace-original`)
- [ocrd-segment-replace-text](ocrd_segment/replace_text.py) (insert text below page from single-segment text files; "inverse" of `extract-*`)
- comparing different layout segmentations:
- [ocrd-segment-evaluate](ocrd_segment/evaluate.py) :construction: (2 input fileGrps; align, compare and evaluate page segmentations; early stage)
- [page-segment-evaluate](ocrd_segment/evaluate.py) (same with standalone CLI)
- 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 behaviour, see `--help` on each processor CLI.
For detailed description on input/output and parameters, see [ocrd-tool.json](ocrd_segment/ocrd-tool.json) or `--dump-json` on each processor CLI.
## Development
### Prerequisities
Requires `libgeos-dev` library for building `shapely` binary requirement, see [Shapely Installation from source](https://shapely.readthedocs.io/en/stable/installation.html#installation-from-source-with-custom-geos-libary). Please ensure it's available before trying to install local requirements.
## Testing
None yet.
Raw data
{
"_id": null,
"home_page": "https://github.com/OCR-D/ocrd_segment",
"name": "ocrd-segment",
"maintainer": null,
"docs_url": null,
"requires_python": null,
"maintainer_email": null,
"keywords": null,
"author": "Konstantin Baierer, Kay-Michael W\u00fcrzner, Robert Sachunsky",
"author_email": "unixprog@gmail.com, wuerzner@gmail.com, sachunsky@informatik.uni-leipzig.de",
"download_url": "https://files.pythonhosted.org/packages/4b/01/bd598397db82e075a9fcc7591d098160d1facfb3c7be55d7cfeea46e3392/ocrd_segment-0.1.24.tar.gz",
"platform": null,
"description": "# ocrd_segment\n\nThis repository aims to provide a number of [OCR-D](https://ocr-d.de) [compliant](https://ocr-d.de/en/spec) [processors](https://ocr-d.de/en/spec/cli) for layout analysis and evaluation.\n\n[![CircleCI](https://dl.circleci.com/status-badge/img/gh/OCR-D/ocrd_segment/tree/master.svg?style=svg)](https://dl.circleci.com/status-badge/redirect/gh/OCR-D/ocrd_segment/tree/master)\n[![image](https://img.shields.io/pypi/v/ocrd_segment.svg)](https://pypi.org/project/ocrd_segment/)\n[![Docker Automated build](https://img.shields.io/docker/automated/ocrd/segment.svg)](https://hub.docker.com/r/ocrd/segment/tags/)\n\n## Installation\n\nIn your [Python virtual environment](https://packaging.python.org/guides/installing-using-pip-and-virtual-environments/), run:\n\n pip install ocrd_segment\n\n## Usage\n\nContains processors for various tasks:\n\n- exporting segment images (including results from preprocessing like cropping/masking, deskewing, dewarping or binarization) along with polygon coordinates and metadata:\n - [ocrd-segment-extract-pages](ocrd_segment/extract_pages.py) (for pages, also exports [MS-COCO](https://cocodataset.org/) format and pageview plots)\n - [ocrd-segment-extract-regions](ocrd_segment/extract_regions.py) (for regions, so exports [MS-COCO](https://cocodataset.org/) format))\n - [ocrd-segment-extract-lines](ocrd_segment/extract_lines.py) (for lines, also exports text and .xlsx)\n - [ocrd-segment-extract-words](ocrd_segment/extract_words.py) (for words, also exports text)\n - [ocrd-segment-extract-glyphs](ocrd_segment/extract_glyphs.py) (for glyphs, also exports text)\n- importing layout segmentations from other formats:\n - [ocrd-segment-from-masks](ocrd_segment/import_image_segmentation.py) (for mask/label images, i.e. semantic segmentation)\n - [ocrd-segment-from-coco](ocrd_segment/import_coco_segmentation.py) (for [MS-COCO](https://cocodataset.org/) annotation)\n- post-processing or repairing layout segmentations:\n - [ocrd-segment-repair](ocrd_segment/repair.py) (validity and consistency of all coordinates; also, for regions, reduce overlaps/redundancy between neighbours, and/or simplify polygons, and/or shrink to the alpha shape of foreground contours)\n - [ocrd-segment-project](ocrd_segment/project.py) (remake segment coordinates into the concave hull / alpha shape of their constituents)\n - [ocrd-segment-replace-original](ocrd_segment/replace_original.py) (rebase all segments on cropped+deskewed border frame as new full page)\n - [ocrd-segment-replace-page](ocrd_segment/replace_page.py) (2 input fileGrps; overwrite segmentation below page of first fileGrp by all segments of second fileGrp, rebasing all coordinates; \"inverse\" of `replace-original`)\n - [ocrd-segment-replace-text](ocrd_segment/replace_text.py) (insert text below page from single-segment text files; \"inverse\" of `extract-*`)\n- comparing different layout segmentations:\n - [ocrd-segment-evaluate](ocrd_segment/evaluate.py) :construction: (2 input fileGrps; align, compare and evaluate page segmentations; early stage)\n - [page-segment-evaluate](ocrd_segment/evaluate.py) (same with standalone CLI)\n- 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)\n - `ocrd-segment-via-template` :construction: (unpublished)\n- data-driven segmentation (input file groups N=1, based on a statistical model, e.g. Neural Network) \n - `ocrd-segment-via-model` :construction: (unpublished)\n\nFor detailed behaviour, see `--help` on each processor CLI.\nFor detailed description on input/output and parameters, see [ocrd-tool.json](ocrd_segment/ocrd-tool.json) or `--dump-json` on each processor CLI.\n\n## Development\n\n### Prerequisities\n\nRequires `libgeos-dev` library for building `shapely` binary requirement, see [Shapely Installation from source](https://shapely.readthedocs.io/en/stable/installation.html#installation-from-source-with-custom-geos-libary). Please ensure it's available before trying to install local requirements.\n\n## Testing\n\nNone yet.\n",
"bugtrack_url": null,
"license": "Apache License 2.0",
"summary": "Page segmentation and segmentation evaluation",
"version": "0.1.24",
"project_urls": {
"Homepage": "https://github.com/OCR-D/ocrd_segment"
},
"split_keywords": [],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "66622e7bb113ff69bfeaf4525a90e393b5bcebdf86fadc74b7c498cde69a23fd",
"md5": "8b20674715e28d6caa051a351f1212c9",
"sha256": "28fa5428c47761c8c602cd949859675afc007f095fc595cf6cf6f0f17d270b5d"
},
"downloads": -1,
"filename": "ocrd_segment-0.1.24-py3-none-any.whl",
"has_sig": false,
"md5_digest": "8b20674715e28d6caa051a351f1212c9",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": null,
"size": 64322,
"upload_time": "2024-04-25T15:40:53",
"upload_time_iso_8601": "2024-04-25T15:40:53.466258Z",
"url": "https://files.pythonhosted.org/packages/66/62/2e7bb113ff69bfeaf4525a90e393b5bcebdf86fadc74b7c498cde69a23fd/ocrd_segment-0.1.24-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "4b01bd598397db82e075a9fcc7591d098160d1facfb3c7be55d7cfeea46e3392",
"md5": "ea25a846502e5ce473833feaf5b982d4",
"sha256": "3de0a3d56f652044557d7de5272572f44cb16b183e55cf9c6e1be63fcd66ee2e"
},
"downloads": -1,
"filename": "ocrd_segment-0.1.24.tar.gz",
"has_sig": false,
"md5_digest": "ea25a846502e5ce473833feaf5b982d4",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 56277,
"upload_time": "2024-04-25T15:40:54",
"upload_time_iso_8601": "2024-04-25T15:40:54.900640Z",
"url": "https://files.pythonhosted.org/packages/4b/01/bd598397db82e075a9fcc7591d098160d1facfb3c7be55d7cfeea46e3392/ocrd_segment-0.1.24.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-04-25 15:40:54",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "OCR-D",
"github_project": "ocrd_segment",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"circle": true,
"requirements": [
{
"name": "ocrd",
"specs": [
[
">=",
"2.20.0"
]
]
},
{
"name": "shapely",
"specs": [
[
">=",
"2.0"
]
]
},
{
"name": "scikit-image",
"specs": []
},
{
"name": "numpy",
"specs": []
},
{
"name": "xlsxwriter",
"specs": []
},
{
"name": "rapidfuzz",
"specs": []
},
{
"name": "opencv-python-headless",
"specs": []
},
{
"name": "ocrd-fork-pycocotools",
"specs": []
}
],
"lcname": "ocrd-segment"
}