segment: add param 'overwrite_order'

Author: Robert Sachunsky

ocrd_cis/ocrd-tool.json

@@ -365,6 +365,11 @@
 					"default": 1.5,
 					"description": "(when operating on the page/table level) smallest width in multiples of scale/capheight of a valley in the horizontal or vertical profiles (across the binarized image) to still be regarded as a gap during recursive X-Y cut from lines to regions; needs to be smaller when more foreground noise is present, increase to avoid mistaking inter-line as paragraph gaps and inter-word as inter-column gaps"
 				},
+				"overwrite_order": {
+					"type": "boolean",
+					"default": true,
+					"description": "(when operating on the page/table level) remove any references for existing TextRegion elements within the top (page/table) reading order; otherwise append"
+				},
 				"overwrite_separators": {
 					"type": "boolean",
 					"default": true,

ocrd_cis/ocropy/common.py

@@ -1300,7 +1300,7 @@ def finalize():
                                                                  lpartitions[label2])
                                 lpartitions[label2] = [0]
                     # re-label and re-order surviving partitions
-                    #lpartitions = np.setdiff1d(np.unique(partitions), [0]) # without bg/sepm
+                    lpartitions = np.setdiff1d(np.unique(partitions), [0]) # without bg/sepm
                     npartitions = len(lpartitions)
                     if debug: LOG.debug('  %d sepmask partitions after filtering and merging', npartitions)
                     if npartitions > 1:

ocrd_cis/ocropy/segment.py

@@ -144,14 +144,17 @@ def process(self):
         then iterate over the element hierarchy down to the requested level.
         
         Depending on ``level-of-operation``, consider existing segments:
-        - if ``overwrite_separators=True`` on ``page`` level, then
-          delete any SeparatorRegions,
-        - if ``overwrite_regions=True`` on ``page`` level, then
-          delete any top-level TextRegions (along with ReadingOrder),
-        - if ``overwrite_regions=True`` on ``table`` level, then
-          delete any TextRegions in TableRegions (along with their OrderGroup),
-        - if ``overwrite_lines=True`` on ``region`` level, then
+        - If ``overwrite_separators=True`` on ``page`` level, then
+          delete any SeparatorRegions.
+        - If ``overwrite_regions=True`` on ``page`` level, then
+          delete any top-level TextRegions (along with ReadingOrder).
+        - If ``overwrite_regions=True`` on ``table`` level, then
+          delete any TextRegions in TableRegions (along with their OrderGroup).
+        - If ``overwrite_lines=True`` on ``region`` level, then
           delete any TextLines in TextRegions.
+        - If ``overwrite_order=True`` on ``page`` or ``table`` level, then
+          delete the reading order OrderedGroup entry corresponding
+          to the (page/table) segment.
         
         Next, get each element image according to the layout annotation (from
         the alternative image of the page/region, or by cropping via coordinates
@@ -206,6 +209,7 @@ def process(self):
         overwrite_lines = self.parameter['overwrite_lines']
         overwrite_regions = self.parameter['overwrite_regions']
         overwrite_separators = self.parameter['overwrite_separators']
+        overwrite_order = self.parameter['overwrite_order']
         oplevel = self.parameter['level-of-operation']
 
         for (n, input_file) in enumerate(self.input_files):
@@ -289,7 +293,7 @@ def process(self):
                         LOG.warning('keeping existing TextRegions in page "%s"', page_id)
                         ignore.extend(regions)
                 # create reading order if necessary
-                if not ro:
+                if not ro or overwrite_order:
                     ro = ReadingOrderType()
                     page.set_ReadingOrder(ro)
                 rogroup = ro.get_OrderedGroup() or ro.get_UnorderedGroup()
@@ -330,6 +334,11 @@ def process(self):
                     if not roelem:
                         LOG.warning("Page '%s' table region '%s' is not referenced in reading order (%s)",
                                     page_id, region.id, "no target to add cells to")
+                    elif overwrite_order:
+                        # replace by empty ordered group with same (index and) ref
+                        # (which can then take the cells as subregions)
+                        roelem = page_subgroup_in_reading_order(roelem)
+                        reading_order[region.id] = roelem
                     elif isinstance(roelem, (OrderedGroupType, OrderedGroupIndexedType)):
                         LOG.warning("Page '%s' table region '%s' already has an ordered group (%s)",
                                     page_id, region.id, "cells will be appended")