How to import trees based on an Inference Overlay

General steps:

• Add a Satellite Overlay

and configure an inference overlay

• • Select neural network named foliage
  • Specify prequel to satellite image
  • Lower mask and score thresholds to 0.3
  • Add all result type overlays
• Calculate the overlay
• Building -> import geo data
  • Import method: A Grid Overlay Legend
  • Overlay: Inference Overlay, frame 0, Select all legend entries
  • Check: Convert Geometry to Buffered Centerpoint
  • No filtering
  • Import as new Building with attributes
  • Assign one function: Standaard Loofbomen (Default Decidious tree)
  • One Stakeholder

Editor → Current Situation (Ribbon tab) → Overlays (Ribbon bar)

How to detect and import trees based on a Inference Overlay:

1. First add a Satellite Overlay. This will be used as input for the Inference Overlay.
2. Add an Inference Overlay
3. With the Inference Overlay added and selected, click on the wizard button in the Overlay's detail panel
4. Once the wizard is ready, click next
5. In the Select Model step, click on the Select button to open the neural network selection panel.
6. In the selection panel, select the Neural Network named foliage.onnx and click next to advance the wizard.
7. In the Input Tensors step, verify that multiple input tensors are recognized and click next. These should be correctly configured.
8. In the Input prequels step, select the Input Prequel 'A', and under Overlay, select the Satellite Overlay. Click next.
9. In the Output Tensors step, verify that output tensors are present, click next.
10. In the Settings step, select the Score Threshold and adjust it to 0.3. This will lower the threshold to accept identified trees.
11. In the Settings step, select the Mask Threshold and adjust it to 0.95. This will raise the threshold to include pixels of identified trees.
12. Verify that the Inference Mode is set to BBOX_DETECTION and click next.
13. In the Overlays step, select all result overlay types and click finish.