v0.1.5

Fixed:

• clip: keep alpha channel if already present

Changed:

• clip: enlarge non-text masks by coords instead of maximum filter (faster)

v0.1.4

Fixed:

• ocropy common: correct getLogger context
• ocropy clip: Pillow workaround for certain image modes

Changed:

• ocrolib: logging instead of plain print/write
• ocropy all: move processing-specific model/logger initialization out of constructor
• ocropy all: iterate regions recursively if possible, introduce table level where appropriate

v0.1.3

Fixed:

• make matplotlib/pylab dependency conditional/optional

Changed:

• aio: replace deprecated ocrd workspace CLI constructs
• logging according to OCR-D/core#599

v0.1.2

Fixed:

• segment: more robust intersection with parent

Changed:

• h/v-lines: increase final dilation along major axis

v0.1.1

Fixed:

• ocrolib: ensure closing FDs after mkstemp and PIL.Image.open
• remove old entry point for ocropy recognition
• set pcGtsId to METS file ID everywhere

v0.1.0

Fixed:

• binarize: zoom wrt. DPI to avoid inverse edges (and gain speed) at high DPI

Changed:

• put derived images under output fileGrp
• use make_file_id and assert_file_grp_cardinality

v0.0.10

Changed:

• segment: new param overwrite_order

v0.0.9

Fixed:

• segment: when overwrite_regions=True, ensure new ReadingOrder
• segment: when simplifying region polygons, avoid overlapping hulls
• segment: intersect new region/line polygons with their parent polygon
• segment: always avoid invalid polygon paths
• segment: ensure new region for lines grouped with ignored regions
• postcorrect, align: only override log level if actually set

Changed:

• postcorrect, align: update to latest JAR

v0.0.8

Fixed:

• ocrd-tool.json (minor formatting)

Changed:

• deskew: add image feature deskewed unconditionally
• test: add test for segment

v0.0.7

Fixed:

• recognize: regression from changed network initialization
• recognize: also load uncompressed models (Python 3 port)
• re/segment: avoid creating invalid polygon coordinates
• ocrolib scale estimation: make DPI-dependent, add fallback for empty/noise pages
• ocrolib morphology: avoid rounding artifacts and asymmetry

Changed:

• ocrolib / all ocropy processors: boost performance via OpenCV/PIL intead of SciPy
• binarize: expose threshold parameter
• re/segment: require images to be binarized already instead of ad-hoc binarization
• re/segment: much faster line segmentation, better separation of neighbouring lines
• segment: much more robust fg h/v-line and bg column detection, new image detection
• segment: add AlternativeImage with h/v-line or image non-text clipped to background
• segment: rewrite of region aggregation via hybrid recursive X-Y cut
• segment: also annotate detected lines (at detected regions) after page segmentation
• segment: expose many new parameters
• segment: add all new lines/regions in proper (but only top-down left-right) reading order
• segment: add table level (like page segmentation, but horizontal-first split strategy)
• segment: also recurse into table cells for region level
• segment: incremental annotation (ignore and re-order existing text/image/separator regions)
• profile: remove this CLI, but keep integrated via API in new pure OCR-D CLI postcorrect
• wer, training: remove these CLI
• postcorrect, align: update to latest JAR
• remove non-OCR-D scripts from installation
• add uninstall target
• update documentation (esp. training, testing and postcorrection)
• improve/extend automatic tests