Merge pull request #61 from bertsky/segment-improve-sep-line-detection

segment: valid polygons, slightly improved h/v-line seps

Author: bertsky

ocrd_cis/ocrd-tool.json

@@ -1,6 +1,6 @@
 {
 	"git_url": "https://github.com/cisocrgroup/ocrd_cis",
-	"version": "0.1.1",
+	"version": "0.1.2",
 	"tools": {
 		"ocrd-cis-ocropy-binarize": {
 			"executable": "ocrd-cis-ocropy-binarize",

ocrd_cis/ocropy/common.py

@@ -561,7 +561,7 @@ def compute_hlines(binary, scale,
     DSAVE('hlines5_selected', horiz+0.6*binary)
     # 6- dilate vertically a little
     #    to get a smooth contour without gaps
-    horiz = morph.r_dilation(horiz, (d0,d1))
+    horiz = morph.r_dilation(horiz, (d0,odd(scale)))
     DSAVE('hlines6_v-dilated', horiz+0.6*binary)
     return horiz > 0
 
@@ -630,7 +630,7 @@ def compute_separators_morph(binary, scale,
     DSAVE('colseps5_selected', vert+0.6*binary)
     # 6- dilate horizontally a little
     #    to get a smooth contour without gaps
-    vert = morph.r_dilation(vert, (d0,d1))
+    vert = morph.r_dilation(vert, (odd(scale),d1))
     DSAVE('colseps6_h-dilated', vert+0.6*binary)
     return vert > 0
 

ocrd_cis/ocropy/segment.py

@@ -5,8 +5,9 @@
 from skimage import draw
 from skimage.morphology import convex_hull_image
 import cv2
-from shapely.geometry import Polygon
+from shapely.geometry import Polygon, asPolygon
 from shapely.prepared import prep
+from shapely.ops import unary_union
 
 from ocrd_modelfactory import page_from_file
 from ocrd_models.ocrd_page import (
@@ -666,16 +667,47 @@ def polygon_for_parent(polygon, parent):
                                [parent.get_imageWidth(),0]])
     else:
         parentp = Polygon(polygon_from_points(parent.get_Coords().points))
+    # check if clipping is necessary
     if childp.within(parentp):
         return polygon
+    # ensure input coords have valid paths (without self-intersection)
+    # (this can happen when shapes valid in floating point are rounded)
+    childp = make_valid(childp)
+    parentp = make_valid(parentp)
+    # clip to parent
     interp = childp.intersection(parentp)
-    if interp.is_empty:
+    # post-process
+    if interp.is_empty or interp.area == 0.0:
         # FIXME: we need a better strategy against this
         raise Exception("intersection of would-be segment with parent is empty")
+    if interp.type == 'GeometryCollection':
+        # heterogeneous result: filter zero-area shapes (LineString, Point)
+        interp = unary_union([geom for geom in interp.geoms if geom.area > 0])
     if interp.type == 'MultiPolygon':
+        # homogeneous result: construct convex hull to connect
+        # FIXME: construct concave hull / alpha shape
         interp = interp.convex_hull
+    if interp.minimum_clearance < 1.0:
+        # follow-up calculations will necessarily be integer;
+        # so anticipate rounding here and then ensure validity
+        interp = asPolygon(np.round(interp.exterior.coords))
+        interp = make_valid(interp)
     return interp.exterior.coords[:-1] # keep open
 
+def make_valid(polygon):
+    for split in range(1, len(polygon.exterior.coords)-1):
+        if polygon.is_valid or polygon.simplify(polygon.area).is_valid:
+            break
+        # simplification may not be possible (at all) due to ordering
+        # in that case, try another starting point
+        polygon = Polygon(polygon.exterior.coords[-split:]+polygon.exterior.coords[:-split])
+    for tolerance in range(1, int(polygon.area)):
+        if polygon.is_valid:
+            break
+        # simplification may require a larger tolerance
+        polygon = polygon.simplify(tolerance)
+    return polygon
+
 def page_get_reading_order(ro, rogroup):
     """Add all elements from the given reading order group to the given dictionary.
     

setup.py

@@ -43,7 +43,7 @@
         'scipy',
         'numpy>=1.17.0',
         'pillow>=7.1.2',
-        'shapely',
+        'shapely>=1.7.1',
         'scikit-image',
         'opencv-python-headless',
         'python-Levenshtein',