Add bounding box features to the previous version

varennes_single_processes/remove_nrcan_duplicates_bbox.py

@@ -0,0 +1,157 @@
+"""
+handle_varennes_ds_workflow module
+NRCan datalayer has two polygons for each buildings' footprint.
+The below workflow has been designed to remove the extra polygons.
+Project Developer: Alireza Adli alireza.adli@concordia.ca
+"""
+# You need to clone mtl_gis_oo project and
+# add it as a dependency of this new project
+from scrub_layer_class import *
+import pandas as pd
+
+# Change the paths by the location of your QGIS installation and datalayers
+qgis_path = 'C:/Program Files/QGIS 3.34.1/apps/qgis'
+
+varennes_nrcan_extra_polygons = \
+  'C:/Users/a_adli/PycharmProjects/varennes_gis_oo/' \
+  'data/initial_data/endeavor/nrcan_without_centroids/auto_building_2.shp'
+
+# First we duplicate the layer to preserve the main data layer.
+duplicated = \
+  'C:/Users/a_adli/PycharmProjects/varennes_gis_oo/' \
+  'data/initial_data/endeavor/nrcan_tolerance_7_removed_dups_pro/' \
+  'nrcan_tolerance_7_removed_dups.shp'
+
+varennes_nrcan = ScrubLayer(
+  qgis_path, varennes_nrcan_extra_polygons, 'NRCan Varennes')
+varennes_nrcan.duplicate_layer(duplicated)
+
+varennes_nrcan_centroids = ScrubLayer(
+  qgis_path, duplicated, 'NRCan Varennes with Coordinates')
+
+# Then we add coordinates to each polygon so we can remove the
+# very similar polygons based on coordinates.
+varennes_nrcan_centroids.layer.startEditing()
+
+# Add new fields for the centroid coordinates
+varennes_nrcan_centroids.layer.dataProvider().\
+  addAttributes([QgsField("centroid_x", QVariant.Double),
+                 QgsField("centroid_y", QVariant.Double),
+                 QgsField("min_x", QVariant.Double),
+                 QgsField("min_y", QVariant.Double),
+                 QgsField("max_x", QVariant.Double),
+                 QgsField("max_y", QVariant.Double)]
+                )
+varennes_nrcan_centroids.layer.updateFields()
+
+
+centroid_x_index = varennes_nrcan_centroids.\
+  layer.fields().indexFromName("centroid_x")
+centroid_y_index = varennes_nrcan_centroids.\
+  layer.fields().indexFromName("centroid_y")
+min_x_index = varennes_nrcan_centroids.\
+  layer.fields().indexFromName("min_x")
+min_y_index = varennes_nrcan_centroids.\
+  layer.fields().indexFromName("min_y")
+max_x_index = varennes_nrcan_centroids.\
+  layer.fields().indexFromName("max_x")
+max_y_index = varennes_nrcan_centroids.\
+  layer.fields().indexFromName("max_y")
+
+for feature in varennes_nrcan_centroids.layer.getFeatures():
+    centroid = feature.geometry().centroid().asPoint()
+    feature.setAttribute(centroid_x_index, centroid.x())
+    feature.setAttribute(centroid_y_index, centroid.y())
+
+    # Calculate bounding box coordinates
+    bbox = feature.geometry().boundingBox()
+    feature.setAttribute(min_x_index, bbox.xMinimum())
+    feature.setAttribute(min_y_index, bbox.yMinimum())
+    feature.setAttribute(max_x_index, bbox.xMaximum())
+    feature.setAttribute(max_y_index, bbox.yMaximum())
+    varennes_nrcan_centroids.layer.updateFeature(feature)
+
+# Commit the changes for adding centroids fields
+varennes_nrcan_centroids.layer.commitChanges()
+
+
+# Pandas is a better option to compare polygons and remove the duplicates
+# so we make a dataframe. We just transfer the necessary fields
+# to the dataframe.
+
+field_names = \
+  ['feature_id', 'centroid_x', 'centroid_y',
+   'min_x', 'min_y', 'max_x', 'max_y']
+
+# Get the indices of the specified fields
+field_indices = [varennes_nrcan_centroids.layer.fields().indexOf(field)
+                 for field in field_names]
+
+# Extract the attribute values and store them in a list of dictionaries
+data = []
+for feature in varennes_nrcan_centroids.layer.getFeatures():
+    attributes = [feature.attributes()[index] for index in field_indices]
+    data.append(dict(zip(field_names, attributes)))
+
+# Create a DataFrame from the list of dictionaries
+varennes_nrcan_centroids_df = pd.DataFrame(data)
+varennes_nrcan_centroids_df['ID'] = range(len(varennes_nrcan_centroids_df))
+
+# Removing polygones based on a diifference (tolerance variable)
+# between centroid_x of polygons and centroid_y of the polygons
+# The tolerance can be changed to one or five
+tolerance = 7
+counter = 0
+centroid_x = varennes_nrcan_centroids_df['centroid_x'].tolist()
+centroid_y = varennes_nrcan_centroids_df['centroid_y'].tolist()
+min_x = varennes_nrcan_centroids_df['min_x'].tolist()
+min_y = varennes_nrcan_centroids_df['min_y'].tolist()
+max_x = varennes_nrcan_centroids_df['max_x'].tolist()
+max_y = varennes_nrcan_centroids_df['max_y'].tolist()
+feature_ids_all = varennes_nrcan_centroids_df['feature_id'].tolist()
+duplicated_feature_ids = []
+
+for feature_index in range(len(centroid_x)):
+  for next_feature_index in range(feature_index + 1, len(centroid_x)):
+    a_x = centroid_x[feature_index]
+    b_x = centroid_x[next_feature_index]
+    subtract_centroid_x = a_x - b_x
+    a_y = centroid_y[feature_index]
+    b_y = centroid_y[next_feature_index]
+    subtract_centroid_y = a_y - b_y
+
+    a_min_x = min_x[feature_index]
+    b_min_x = min_x[next_feature_index]
+    subtract_min_x = a_min_x - b_min_x
+    a_min_y = min_y[feature_index]
+    b_min_y = min_y[next_feature_index]
+    subtract_min_y = a_min_y - b_min_y
+
+    a_max_x = max_x[feature_index]
+    b_max_x = max_x[next_feature_index]
+    subtract_max_x = a_max_x - b_max_x
+    a_max_y = max_y[feature_index]
+    b_max_y = max_y[next_feature_index]
+    subtract_max_y = a_max_y - b_max_y
+
+    if abs(subtract_centroid_x) < tolerance and \
+            abs(subtract_centroid_y) < tolerance and \
+            abs(subtract_min_x) < tolerance and \
+            abs(subtract_min_y) < tolerance and \
+            abs(subtract_max_x) < tolerance and \
+            abs(subtract_max_y) < tolerance:
+      duplicated_feature_ids.append(feature_ids_all[next_feature_index])
+
+# Removing records based on the duplicated_feature_ids list
+varennes_nrcan_centroids.layer.startEditing()
+
+
+features = varennes_nrcan_centroids.layer.getFeatures()
+# Iterate through features in the layer
+for feature in features:
+    if feature['feature_id'] in duplicated_feature_ids:
+        # Delete the feature
+        varennes_nrcan_centroids.layer.deleteFeature(feature.id())
+
+# Save changes and stop editing
+varennes_nrcan_centroids.layer.commitChanges()