Merge pull request #184 from bacetiner/master

Addressed the bug resulting in BRAILS searching for nonexistent street-level imagery

brails/InventoryGenerator.py

@@ -40,7 +40,7 @@
 # Satish Rao
 #
 # Last updated:
-# 02-01-2024 
+# 02-07-2024 
 
 import random
 import sys
@@ -229,7 +229,7 @@ def parse_attribute_input(attrIn,attrEnabled):
 
             return attrOut
 
-        def write_inventory_output(inventorydf,outFile):          
+        def write_inventory_output(inventorydf,outFile,lengthUnit):          
             """
             Function that writes the data in inventorydf DataFrame into a CSV 
             or GeoJSON file based on the file name defined in the outFile.  
@@ -254,7 +254,7 @@ def write_inventory_output(inventorydf,outFile):
             dfout = inventorydf.copy(deep=True)
             dfout = dfout.drop(columns=['satellite_images', 'street_images'], 
                                errors='ignore')
-
+        
 
             for index, row in inventorydf.iterrows():            
                 dfout.loc[index, 'Footprint'] = ('{"type":"Feature","geometry":' + 
@@ -267,14 +267,14 @@ def write_inventory_output(inventorydf,outFile):
 
             # Rearrange the column order of dfout such that the Footprint field is 
             # the last:
-            cols = [col for col in dfout.columns if col!='Footprint'] 
+            cols = [col for col in dfout.columns if col not in ['Footprint','Latitude','Longitude']] 
             new_cols = ['Latitude','Longitude'] + cols + ['Footprint']
             dfout = dfout[new_cols]
 
             # If the inventory is desired in CSV format, write dfout to a CSV:
             if '.csv' in outFile.lower():
                 dfout.to_csv(outFile, index=True, index_label='id') 
-
+        
             # Else write the inventory into a GeoJSON file:
             else:
                 if '.geojson' not in outFile.lower():
@@ -287,7 +287,7 @@ def write_inventory_output(inventorydf,outFile):
                                     "properties": {"name": "urn:ogc:def:crs:OGC:1.3:CRS84" }},
                             'units': {"length": lengthUnit},
                             'features':[]}
-
+        
                 attrs = dfout.columns.values.tolist()
                 attrs.remove('Footprint')
                 for index, row in dfout.iterrows(): 
@@ -300,13 +300,18 @@ def write_inventory_output(inventorydf,outFile):
                     feature['geometry']['coordinates'] = json.loads(fp)
                     feature['properties']['id'] = index
                     for attr in attrs:
-                        feature['properties'][attr] = row[attr]
+                        res = row[attr]
+                        if pd.isnull(res):
+                            feature['properties'][attr] = 'NA'
+                        else:
+                            feature['properties'][attr] = res
                     geojson['features'].append(feature)
-
+        
                 with open(outFile, 'w') as output_file:
                     json.dump(geojson, output_file, indent=2)
 
             print(f'\nFinal inventory data available in {outFile} in {os.getcwd()}')
+
         # Parse/correct the list of user requested building attributes:
         self.attributes = parse_attribute_input(attributes, self.enabledAttributes)      
 
@@ -465,7 +470,7 @@ def write_inventory_output(inventorydf,outFile):
                 self.inventory['roofeaveheight'] = self.inventory['roofeaveheight'].apply(lambda x: x*0.3048)
 
         # Write the genereated inventory in outFile:
-        write_inventory_output(self.inventory,outFile)   
+        write_inventory_output(self.inventory,outFile,lengthUnit)   
 
         # Merge the DataFrame of predicted attributes with the DataFrame of
         # incomplete inventory and print the resulting table to the output file

brails/workflow/ImHandler.py

@@ -37,7 +37,7 @@
 # Barbaros Cetiner
 #
 # Last updated:
-# 01-09-2024   
+# 02-07-2024   
 
 
 import os
@@ -707,7 +707,7 @@ def download_streetlev_image(fp, fpcent, im_name, depthmap_name, apikey,
             try:
                 pano['id'] = get_pano_id(pano['queryLatLon'],apikey)
             except:
-                return
+                return None
 
             # Get the metdata for the pano:
             pano = get_pano_meta(pano, savedmap = True, dmapoutname = depthmap_name)
@@ -740,7 +740,7 @@ def download_streetlev_image(fp, fpcent, im_name, depthmap_name, apikey,
         # street-level imagery:
         self.footprints = footprints
         self.centroids = []
-        self.street_images = []
+        street_images = []
         inps = [] 
         for footprint in footprints:
             fp = np.fliplr(np.squeeze(np.array(footprint))).tolist()
@@ -750,7 +750,7 @@ def download_streetlev_image(fp, fpcent, im_name, depthmap_name, apikey,
             imName.replace(".","")
             im_name = f"tmp/images/street/imstreet_{imName}.jpg"
             depthmap_name = f"tmp/images/depthmap/dmstreet_{imName}.txt"            
-            self.street_images.append(im_name)
+            street_images.append(im_name)
             inps.append((fp,(fp_cent.y,fp_cent.x),im_name,depthmap_name))
 
         # Download building-wise street-level imagery and depthmap strings:
@@ -778,11 +778,12 @@ def download_streetlev_image(fp, fpcent, im_name, depthmap_name, apikey,
         if save_all_cam_metadata==False:
             self.cam_elevs = []
             self.depthmaps = [] 
-            for im in self.street_images:
+            for (ind,im) in enumerate(street_images):
                 if results[im] is not None:
                     self.cam_elevs.append(results[im][0])
                     self.depthmaps.append(results[im][1])
                 else:
+                    street_images[ind] = None
                     self.cam_elevs.append(None)
                     self.depthmaps.append(None)
         else:
@@ -793,7 +794,7 @@ def download_streetlev_image(fp, fpcent, im_name, depthmap_name, apikey,
             self.headings = []
             self.pitch = []
             self.zoom_levels = []
-            for im in self.street_images:
+            for (ind,im) in enumerate(street_images):
                 if results[im] is not None:
                     self.cam_elevs.append(results[im][0])
                     self.cam_latlons.append(results[im][1])
@@ -810,7 +811,9 @@ def download_streetlev_image(fp, fpcent, im_name, depthmap_name, apikey,
                     self.headings.append(None)
                     self.pitch.append(None)
                     self.zoom_levels.append(None)
-
+                    street_images[ind] = None
+        self.street_images = street_images.copy()
+
     def GetGoogleStreetImageAPI(self,footprints):
         self.footprints = footprints[:]
         # Function that downloads a file given its URL and the desired path to save it: