Rewrite

.circleci/config.yml

@@ -2,43 +2,53 @@ version: 2
 
 jobs:
 
-  build-python35:
+  build-python36:
     docker:
-      - image: python:3.5.10-buster
+      - image: python:3.6.15-buster
     steps:
       - checkout
-      - run: make deps-ubuntu deps-test deps install assets
+      # workaround for OCR-D/core#923
+      - run: apt-get update && apt-get install -y libgeos-dev
+      - run: make deps-ubuntu deps-test deps install
+      - run: ocrd resmgr download ocrd-kraken-segment blla.mlmodel
+      - run: ocrd resmgr download ocrd-kraken-recognize en_best.mlmodel
       - run: make test
 
-  build-python36:
+  build-python37:
     docker:
-      - image: python:3.6.12-buster
+      - image: python:3.7.15-buster
     steps:
       - checkout
-      - run: make deps-ubuntu deps-test deps install assets
+      - run: make deps-ubuntu deps-test deps install
+      - run: ocrd resmgr download ocrd-kraken-segment blla.mlmodel
+      - run: ocrd resmgr download ocrd-kraken-recognize en_best.mlmodel
       - run: make test
 
-  build-python37:
+  build-python38:
     docker:
-      - image: python:3.7.9-buster
+      - image: python:3.8.14-buster
     steps:
       - checkout
-      - run: make deps-ubuntu deps-test deps install assets
+      - run: make deps-ubuntu deps-test deps install
+      - run: ocrd resmgr download ocrd-kraken-segment blla.mlmodel
+      - run: ocrd resmgr download ocrd-kraken-recognize en_best.mlmodel
       - run: make test
 
-  build-python38:
+  build-python39:
     docker:
-      - image: python:3.8.5-buster
+      - image: python:3.9.14-buster
     steps:
       - checkout
-      - run: make deps-ubuntu deps-test deps install assets
+      - run: make deps-ubuntu deps-test deps install
+      - run: ocrd resmgr download ocrd-kraken-segment blla.mlmodel
+      - run: ocrd resmgr download ocrd-kraken-recognize en_best.mlmodel
       - run: make test
 
 workflows:
   version: 2
   build:
     jobs:
-      - build-python35
       - build-python36
       - build-python37
       - build-python38
+      - build-python39

CHANGELOG.md

@@ -5,7 +5,14 @@ Versioned according to [Semantic Versioning](http://semver.org/).
 
 ## Unreleased
 
-## [0.1.2] - 2019-09-24
+## [0.2.0] - 2022-07-18
+
+Added:
+
+  * text recognition with `ocrd-kraken-recognize`, #33
+  * rewrite of `ocrd-kraken-segment`, #33
+
+## [0.1.2] - 2020-09-24
 
 Fixed:
 
@@ -43,6 +50,7 @@ Changed:
 Initial release
 
 <!-- link-labels -->
+[0.2.0]: v0.2.0...v0.1.2
 [0.1.2]: v0.1.2...v0.1.1
 [0.1.1]: v0.1.1...v0.1.0
 [0.1.0]: v0.1.0...v0.0.2

Dockerfile

@@ -1,20 +1,34 @@
 FROM ocrd/core
-MAINTAINER OCR-D
+ARG VCS_REF
+ARG BUILD_DATE
+LABEL \
+    maintainer="https://ocr-d.de/kontakt" \
+    org.label-schema.vcs-ref=$VCS_REF \
+    org.label-schema.vcs-url="https://github.com/OCR-D/ocrd_tesserocr" \
+    org.label-schema.build-date=$BUILD_DATE
+
 ENV DEBIAN_FRONTEND noninteractive
 ENV PYTHONIOENCODING utf8
 ENV LC_ALL C.UTF-8
 ENV LANG C.UTF-8
+ENV PIP pip3
+
+# avoid HOME/.local/share (hard to predict USER here)
+# so let XDG_DATA_HOME coincide with fixed system location
+# (can still be overridden by derived stages)
+ENV XDG_DATA_HOME /usr/local/share
 
 WORKDIR /build-ocrd
 COPY setup.py .
-COPY requirements.txt .
-RUN apt-get update && \
-    apt-get -y install --no-install-recommends \
-    ca-certificates \
-    make \
-    git
 COPY ocrd_kraken ./ocrd_kraken
-RUN pip3 install --upgrade pip
-RUN pip3 install .
+COPY ocrd_kraken/ocrd-tool.json .
+COPY README.md .
+COPY requirements.txt .
+COPY Makefile .
+RUN make deps-ubuntu \
+    && make deps install \
+    && rm -fr /build-ocrd
+
+WORKDIR /data
+VOLUME /data
 
-ENTRYPOINT ["/bin/sh", "-c"]

Makefile

@@ -18,14 +18,16 @@ help:
 	@echo "    deps         Install python deps via pip"
 	@echo "    deps-test    Install testing deps via pip"
 	@echo "    install      Install"
-	@echo "    docker       Build docker image"
+	@echo "    install-dev  Install in editable mode"
+	@echo "    docker       Build Docker image"
 	@echo "    test         Run test"
 	@echo "    repo/assets  Clone OCR-D/assets to ./repo/assets"
-	@echo "    assets       Setup test assets"
+	@echo "    tests/assets       Setup test assets"
 	@echo ""
 	@echo "  Variables"
 	@echo ""
 	@echo "    DOCKER_TAG  Docker container tag ("$(DOCKER_TAG)")"
+	@echo "    PYTEST_ARGS Additional runtime options for pytest ("$(PYTEST_ARGS)")"
 
 # END-EVAL
 
@@ -44,16 +46,21 @@ deps-test:
 
 # Install
 install:
+	$(PIP) install .
+
+install-dev:
 	$(PIP) install -e .
 
 # Build docker image
 docker:
-	docker build -t $(DOCKER_TAG) .
+	docker build \
+        --build-arg VCS_REF=$$(git rev-parse --short HEAD) \
+        --build-arg BUILD_DATE=$$(date -u +"%Y-%m-%dT%H:%M:%SZ") \
+	-t $(DOCKER_TAG) .
 
-.PHONY: test
 # Run test
-test:
-	$(PYTHON) -m pytest tests
+test: tests/assets
+	$(PYTHON) -m pytest tests $(PYTEST_ARGS)
 
 #
 # Assets
@@ -66,6 +73,8 @@ repo/assets:
 
 
 # Setup test assets
-assets: repo/assets
+tests/assets: repo/assets
 	mkdir -p tests/assets
 	cp -r -t tests/assets repo/assets/data/*
+
+.PHONY: docker install install-dev deps deps-ubuntu deps-test test help

README.md

@@ -1,7 +1,96 @@
 # ocrd_kraken
 
-> Wrapper for the kraken OCR engine
+> OCR-D wrapper for the Kraken OCR engine
 
 [![image](https://travis-ci.org/OCR-D/ocrd_kraken.svg?branch=master)](https://travis-ci.org/OCR-D/ocrd_kraken)
 [![Docker Automated build](https://img.shields.io/docker/automated/ocrd/kraken.svg)](https://hub.docker.com/r/ocrd/kraken/tags/)
 [![image](https://circleci.com/gh/OCR-D/ocrd_kraken.svg?style=svg)](https://circleci.com/gh/OCR-D/ocrd_kraken)
+
+## Introduction
+
+This package offers [OCR-D](https://ocr-d.de/en/spec) compliant [workspace processors](https://ocr-d.de/en/spec/cli)
+for (some of) the functionality of [Kraken](https://kraken.re).
+
+(Each processor is a parameterizable step in a configurable [workflow](https://ocr-d.de/en/workflows)
+of the [OCR-D functional model](https://ocr-d.de/en/about).
+There are usually various alternative processor implementations for each step.
+Data is represented with [METS](https://ocr-d.de/en/spec/mets) and [PAGE](https://ocr-d.de/en/spec/page).)
+
+It includes image preprocessing (binarization), layout analysis (region and line+baseline segmentation), and text recognition.
+
+## Installation
+
+### With Docker
+
+This is the best option if you want to run the software in a container.
+
+You need to have [Docker](https://docs.docker.com/install/linux/docker-ce/ubuntu/)
+
+
+    docker pull ocrd/kraken
+
+
+To run with Docker:
+
+
+    docker run -v path/to/workspaces:/data ocrd/kraken ocrd-kraken-recognize ...
+
+### Native, from PyPI
+
+This is the best option if you want to use the stable, released version.
+
+    pip install ocrd_kraken
+
+
+### Native, from git
+
+Use this option if you want to change the source code or install the latest, unpublished changes.
+
+We strongly recommend to use [venv](https://packaging.python.org/guides/installing-using-pip-and-virtual-environments/).
+
+    git clone https://github.com/OCR-D/ocrd_kraken
+    cd ocrd_kraken
+    sudo make deps-ubuntu # or manually from git or via ocrd_all
+    make deps        # or pip install -r requirements.txt
+    make install     # or pip install .
+
+## Models
+
+Kraken uses data-driven (neural) models for segmentation and recognition, but comes with no pretrained "official" models.
+There is a [public repository](https://zenodo.org/communities/ocr_models) of community-provided models, which can also
+be queried and downloaded from via `kraken` standalone CLI.
+(See [Kraken docs](https://kraken.re/master/advanced.html#repo) for details.)
+
+For the OCR-D wrapper, since all OCR-D processors must resolve file/data resources in a [standardized way](https://ocr-d.de/en/spec/cli#processor-resources), there is a general mechanism for managing models, i.e. installing and using them by name. We currently manage our own list of recommended models (without delegating to the above repo).
+
+Models always use the filename suffix `.mlmodel`, but are just loaded by their basename.
+
+See the [OCR-D model guide](https://ocr-d.de/en/models) and
+
+    ocrd resmgr --help
+
+## Usage
+
+For details, see docstrings in the individual processors and [ocrd-tool.json](ocrd_tesserocr/ocrd-tool.json) descriptions,
+or simply `--help`.
+
+Available [OCR-D processors](https://ocr-d.de/en/spec/cli) are:
+
+- [ocrd-kraken-binarize](ocrd_kraken/binarize.py) (nlbin – not recommended)  
+  - adds `AlternativeImage` files (per page, region or line) to the output fileGrp
+- [ocrd-kraken-segment](ocrd_kraken/segment.py) (all-in-one segmentation – recommended for handwriting and simply layouted prints)  
+  - adds `TextRegion`s, `TableRegion`s, `ImageRegion`s, `MathsRegion`s, `NoiseRegion`s, `ReadingOrder` and `AlternativeImage` to `Page` (depending on model training)
+  - adds `TextLine`s to `TextRegion`s, including their `Baseline`
+- [ocrd-kraken-recognize](ocrd_kraken/recognize.py)
+  - adds `Word`s to `TextLine`s
+  - adds `Glyph`s to `Word`s
+  - adds `TextEquiv`
+
+## Testing
+
+    make test
+
+
+This downloads test data from https://github.com/OCR-D/assets under `repo/assets`, and runs some basic tests of the Python API.
+
+Set `PYTEST_ARGS="-s --verbose"` to see log output (`-s`) and individual test results (`--verbose`).

ocrd_kraken/__init__.py

@@ -1,3 +0,0 @@
-from .binarize import KrakenBinarize
-from .ocr import KrakenOcr
-from .segment import KrakenSegment

ocrd_kraken/cli.py

@@ -2,14 +2,9 @@
 
 from ocrd.decorators import ocrd_cli_options, ocrd_cli_wrap_processor
 from ocrd_kraken.binarize import KrakenBinarize
-from ocrd_kraken.segment import KrakenSegment
 
 @click.command()
 @ocrd_cli_options
 def ocrd_kraken_binarize(*args, **kwargs):
     return ocrd_cli_wrap_processor(KrakenBinarize, *args, **kwargs)
 
-@click.command()
-@ocrd_cli_options
-def ocrd_kraken_segment(*args, **kwargs):
-    return ocrd_cli_wrap_processor(KrakenSegment, *args, **kwargs)

ocrd_kraken/cli/binarize.py

@@ -0,0 +1,9 @@
+import click
+
+from ocrd.decorators import ocrd_cli_options, ocrd_cli_wrap_processor
+from ocrd_kraken.binarize import KrakenBinarize
+
+@click.command()
+@ocrd_cli_options
+def cli(*args, **kwargs):
+    return ocrd_cli_wrap_processor(KrakenBinarize, *args, **kwargs)

ocrd_kraken/cli/recognize.py

@@ -0,0 +1,11 @@
+import click
+
+from ocrd.decorators import ocrd_cli_options, ocrd_cli_wrap_processor
+from ocrd_kraken.recognize import KrakenRecognize
+
+@click.command()
+@ocrd_cli_options
+def cli(*args, **kwargs):
+    return ocrd_cli_wrap_processor(KrakenRecognize, *args, **kwargs)
+
+

ocrd_kraken/cli/segment.py

@@ -0,0 +1,8 @@
+import click
+from ocrd.decorators import ocrd_cli_options, ocrd_cli_wrap_processor
+from ocrd_kraken.segment import KrakenSegment
+
+@click.command()
+@ocrd_cli_options
+def cli(*args, **kwargs):
+    return ocrd_cli_wrap_processor(KrakenSegment, *args, **kwargs)

ocrd_kraken/config.py

@@ -1,4 +1,5 @@
 import json
-from pkg_resources import resource_string
+from pkg_resources import resource_filename
 
-OCRD_TOOL = json.loads(resource_string(__name__, 'ocrd-tool.json').decode('utf8'))
+with open(resource_filename(__name__, 'ocrd-tool.json'), 'r', encoding='utf-8') as f:
+    OCRD_TOOL = json.load(f)