Improve documentation of surface_hausdorff_distance.

Python/34_Segmentation_Evaluation.ipynb

@@ -490,11 +490,15 @@
     "    \"\"\"\n",
     "    Compute symmetric surface distances and take the maximum.\n",
     "    \"\"\"\n",
-    "    # When the segmentation touches the image border we have corner case for the computation.\n",
-    "    # The surface/contour computation yields a surface that is outside the image bounds and\n",
-    "    # is thus not included in the computation. This is not correct for our use case. By padding\n",
-    "    # the segmentations we ensure that we obtain the surface as expected and the distance computations\n",
-    "    # are correct. Padding has no effect on results if the original segmentations did not touch the image\n",
+    "    # When the segmentation reaches the image border, we have a corner case and we need to make an\n",
+    "    # assumption with respect to the values outside the image's physical extent. We can assume\n",
+    "    # (a) the region outside the segmentation image has the same values as inside, segmented region\n",
+    "    # continues outside the image so no surface at the image border (b) the region outside the\n",
+    "    # segmentation image is empty.\n",
+    "    # The implementation of the LabelContourFilter appears to use assumption 'a' but we would like to\n",
+    "    # use assumption 'b'. By padding the image with zeros we mimic the conditions of assumption\n",
+    "    # 'b' which will yield a surface at the image border.\n",
+    "    # Padding has no effect on results if the original segmentations did not touch the image\n",
     "    # border.\n",
     "    dim = reference_segmentation.GetDimension()\n",
     "    reference_segmentation = sitk.ConstantPad(\n",

tests/additional_dictionary.txt

@@ -160,6 +160,7 @@ Kamath
 LBFGS
 LSMImageIO
 LaTeX
+LabelContourFilter
 LabelContourImageFilter
 LabelMapContourOverlayImageFilter
 LabelOverlayImageFilter