Rewrite

.circleci/config.yml

@@ -2,20 +2,14 @@ version: 2
 
 jobs:
 
-  build-python35:
-    docker:
-      - image: python:3.5.10-buster
-    steps:
-      - checkout
-      - run: make deps-ubuntu deps-test deps install assets
-      - run: make test
-
   build-python36:
     docker:
       - image: python:3.6.12-buster
     steps:
       - checkout
       - run: make deps-ubuntu deps-test deps install assets
+      - run: ocrd resmgr download ocrd-kraken-segment blla.mlmodel
+      - run: ocrd resmgr download ocrd-kraken-recognize en_best.mlmodel
       - run: make test
 
   build-python37:
@@ -24,6 +18,8 @@ jobs:
     steps:
       - checkout
       - run: make deps-ubuntu deps-test deps install assets
+      - run: ocrd resmgr download ocrd-kraken-segment blla.mlmodel
+      - run: ocrd resmgr download ocrd-kraken-recognize en_best.mlmodel
       - run: make test
 
   build-python38:
@@ -32,13 +28,14 @@ jobs:
     steps:
       - checkout
       - run: make deps-ubuntu deps-test deps install assets
+      - 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

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

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)

ocrd_kraken/ocrd-tool.json

@@ -1,6 +1,6 @@
 {
   "git_url": "https://github.com/OCR-D/ocrd_kraken",
-  "version": "0.0.2",
+  "version": "0.2.0",
   "tools": {
     "ocrd-kraken-binarize": {
       "executable": "ocrd-kraken-binarize",
@@ -15,6 +15,7 @@
       "description": "Binarize images with kraken",
       "parameters": {
         "level-of-operation": {
+          "description": "level-of-operation",
           "type": "string",
           "default": "page",
           "enum": ["page", "block", "line"]
@@ -32,43 +33,100 @@
       "description": "Block segmentation with kraken",
       "parameters": {
         "text_direction": {
-          "type": "string",
-          "description": "Sets principal text direction",
-          "enum": ["horizontal-lr", "horizontal-rl", "vertical-lr", "vertical-rl"],
+          "type": "string", 
+          "description": "Sets principal text direction", 
+          "enum": ["horizontal-lr", "horizontal-rl", "vertical-lr", "vertical-rl"], 
           "default": "horizontal-lr"
         },
-        "script_detect": {
-          "type": "boolean",
-          "description": "Enable script detection on segmenter output",
+        "maxcolseps": {
+          "description": "Maximum number of column separators. Set to 0 for single-column text to avoid unnecessary computation.", 
+          "type": "number", 
+          "format": "integer", 
+          "default": 2
+        },
+        "scale": {
+          "description": "mean xheight size of glyphs (guessed if zero)", 
+          "type": "number", 
+          "format": "float", 
+          "default": 0
+        },
+        "black_colseps": {
+          "description": "Whether column separators are assumed to be vertical black lines or not", 
+          "type": "boolean", 
           "default": false
         },
-        "maxcolseps": {"type": "number", "format": "integer", "default": 2},
-        "scale": {"type": "number", "format": "float", "default": 0},
-        "black_colseps": {"type": "boolean", "default": false},
-        "white_colseps": {"type": "boolean", "default": false},
-        "level-of-operation": {
+        "remove_hlines": {
+          "description": "Remove horizontal colseps before segmentation", 
+          "type": "boolean", 
+          "default": true
+        },
+        "blla_model": {
+          "description": "Model used for baseline detection and page segmentation. Ignored if use_legacy.",
           "type": "string",
-          "default": "page",
-          "enum": ["page", "block", "line"]
+          "format": "uri",
+          "content-type": "application/python-cpickle",
+          "cacheable": true,
+          "default": "blla.mlmodel"
+        },
+        "blla_classes": {
+          "description": "Class mapping for the region types trained into blla_model.",
+          "type": "object",
+          "minProperties": 2,
+          "additionalProperties": { "type": "string",
+                                    "enum": ["TextRegion", "ImageRegion", "LineDrawingRegion",
+                                             "GraphicRegion", "TableRegion", "ChartRegion",
+                                             "MapRegion", "SeparatorRegion", "MathsRegion",
+                                             "ChemRegion", "MusicRegion", "AdvertRegion",
+                                             "NoiseRegion", "UnknownRegion", "CustomRegion"] },
+          "default": {"text": "TextRegion", "image": "ImageRegion", "line drawing": "LineDrawingRegion",
+                      "graphic": "GraphicRegion", "table": "TableRegion", "chart": "ChartRegion",
+                      "map": "MapRegion", "separator": "SeparatorRegion", "maths": "MathsRegion",
+                      "chem": "ChemRegion", "music": "MusicRegion", "advert": "AdvertRegion",
+                      "noise": "NoiseRegion", "unknown": "UnknownRegion", "custom": "CustomRegion"}
+        },
+        "device": {
+          "description": "GPU ID or 'cpu' to run on CPU only", 
+          "type": "string", 
+          "default": "cpu"
+        },
+        "use_legacy": {
+          "description": "Use legacy box segmenter as opposed to neural net baseline segmenter", 
+          "type": "boolean", 
+          "default": false
         }
       }
     },
-    "ocrd-kraken-ocr": {
-      "executable": "ocrd-kraken-ocr",
+    "ocrd-kraken-recognize": {
+      "executable": "ocrd-kraken-recognize",
       "categories": ["Text recognition and optimization"],
-      "steps": [
-        "recognition/text-recognition"
-      ],
+      "steps": ["recognition/text-recognition"],
       "description": "OCR with kraken",
       "parameters": {
-        "lines-json": {
+        "model": {
+          "description": "OCR model to recognize with",
+          "type": "string",
+          "format": "uri",
+          "content-type": "application/python-cpickle",
+          "cacheable": true,
+          "default": "en_best.mlmodel"
+        },
+        "pad": {
+          "description": "Extra blank padding to the left and right of text line.",
+          "type": "number",
+          "format": "integer",
+          "default": 16
+        },
+        "bidi_reordering": {
+          "description": "Reorder classes in the ocr_record according to  the Unicode bidirectional algorithm for correct display.",
+          "type": "boolean",
+          "default": true
+        },
+        "device": {
+          "description": "CUDA ID (e.g. 'cuda:0') for computation on GPU, or 'cpu' to run on CPU only",
           "type": "string",
-          "format": "url",
-          "required": "true",
-          "description": "URL to line segmentation in JSON"
+          "default": "cpu"
         }
       }
     }
-
   }
 }