Use the model

You can use this model in the Detect Objects Using Deep Learning tool available in the Image Analyst toolbox for inference or the Extract Features Using AI Models tool available in the GeoAI toolbox in ArcGIS Pro for inference and postprocessing of the inferred results. Follow the steps below to use the model to extract building footprints from high resolution imagery.

Supported imagery

Orthorectified imagery (on-the-fly or persisted ortho products) with the following characteristics is supported:

• Resolution—High resolution (10–40 centimeters)
• Bit depth—8 Bit Unsigned
• Bands—Three bands (for example, red, green, and blue)

Note:

Off-nadir imagery or imagery with a high obliquity angle will not produce suitable results.

Preprocess the imagery (optional)

If you have a multispectral product, follow the steps below to preprocess the imagery before using it with the model:

1. Open ArcGIS Pro and browse to the image product in the Contents pane.
2. Expand the product file (.imd file) and add the Pansharpen layer to the map.

   

3. Right-click the newly added layer and select Edit Function Chain.
4. Click Stretch Function in the function chain window to edit the stretch properties.

   

5. In the Stretch Properties pane, do the following:
   1. Change Output Pixel Type to 8 Bit Unsigned on the General tab.

      

   2. Keep the default settings on the Parameters tab.

      

6. Click the OK button to apply the changes.

   Click the play button in the top menu of the tool to apply the changes.

   

Extract building footprints

Use the following steps to extract building footprints from the imagery:

1. Download the Building Footprint Extraction—USA model and add the imagery layer in ArcGIS Pro
2. Zoom to an area of interest.

   

3. Click Tools on the Analysis tab.

   

4. Click the Toolboxes tab in the Geoprocessing pane, expand GeoAI Tools, and select the Extract Features Using AI Models tool under Imagery AI.

   

5. On the Parameters tab, set the parameters as follows:
   1. Input Raster—Choose the input raster on which processing will be performed.
   2. Area of Interest(optional) — Set the geographical extent that will be used to extract features.

      Only features within the area of interest will be extracted.

   3. Mode—Select Infer and Postprocess.
   4. Output Location (optional) — Select the file geodatabase where the outputs will be stored.
   5. Output Prefix— Set the prefix prefix that will be added to the name of outputs and name of group layer.
   6. Models— Select the pretrained model.
      • Set Pretrained Models— Select the BUILDING FOOTPRINT EXTRACTION - USA pretrained model from the drop down list.
      • Set Polygon Regularization OptionsPolygon Postprocessing is the default postprocessing option for this model.
        • Regularization Method— Use the default regularization method. The default is Right Angles.
        • Tolerance—Use the default value. The default is 1 meters.

        

6. On the Environments tab, set the values as follows:
   1. Processor Type—Set GPU ID to the GPU to be used.

      

7. Click Run..

   The output group layer with postprocessed building footprints is added to the map.