Merge pull request #120 from bertsky/fix-segment-line-hull

segment-line: validate intersection with parent
OCR-D · Apr 8, 2020 · f11c111 · f11c111
2 parents 7026429 + a6659f0
commit f11c111
Show file tree

Hide file tree

Showing 2 changed files with 15 additions and 2 deletions.
diff --git a/ocrd_tesserocr/segment_line.py b/ocrd_tesserocr/segment_line.py
@@ -111,7 +111,19 @@ def process(self):
                         line_poly = Polygon(line_polygon)
                         if not line_poly.within(region_poly):
                             # this could happen due to rotation
-                            line_poly = line_poly.intersection(region_poly).convex_hull
+                            interline = line_poly.intersection(region_poly)
+                            if interline.is_empty:
+                                continue # ignore this line
+                            if hasattr(interline, 'geoms'):
+                                # is (heterogeneous) GeometryCollection
+                                area = 0
+                                for geom in interline.geoms:
+                                    if geom.area > area:
+                                        area = geom.area
+                                        interline = geom
+                                if not area:
+                                    continue
+                            line_poly = interline.convex_hull
                             line_polygon = line_poly.exterior.coords
                         line_polygon = coordinates_for_segment(line_polygon, region_image, region_coords)
                         line_points = points_from_polygon(line_polygon)

diff --git a/ocrd_tesserocr/segment_table.py b/ocrd_tesserocr/segment_table.py
@@ -141,7 +141,8 @@ def process(self):
                     LOG.info("Detecting table cells in region '%s'", region.id)
                     #
                     # detect the region segments:
-                    tessapi.SetPageSegMode(PSM.AUTO) # treat table like page
+                    tessapi.SetPageSegMode(PSM.SPARSE_TEXT) # retrieve "cells"
+                    # TODO: we should XY-cut the sparse cells in regroup them into consistent cells
                     layout = tessapi.AnalyseLayout()
                     roelem = reading_order.get(region.id)
                     if not roelem: