Extract Building Footprint
Operation Steps
1.Before extracting the building footprint, first use Classify by Deep Learning to classify the point cloud and obtain a classification result that distinguishes building points from other point classes, to be used as the input data for this function.
2.For ground-based point cloud data, check Slice Elevation and select a height with a complete outline and high point cloud density as the horizontal section location, to be used as the extraction data. This is generally not required for airborne data.
3.You can hide non-building point class in Display by Classification; hidden point cloud will not be included in the extraction.
4.Set the parameters on the Basic tab; to adjust building separation and area filtering, switch to the Advanced tab.
5.Two modes are supported: Polygon Extraction and Global Extraction.
- Polygon Extraction: Extracts the building footprint within the selected box (extraction runs automatically once the selection is complete).
- Global Extraction: Extracts all point cloud data.
Parameter Description
Basic
- Detail Size: Notches narrower than this width will be filled in. Decrease the value to retain more detail; increase it for a simpler outline.
- Regularize: If checked, the outline is regularized into straight wall lines, with adjacent walls aligned and intersected to generate regular corner points; suitable for regular buildings. This is not recommended for curved or irregular outlines.
- Ground Class: The category of the ground points. This needs to be set after checking Fill Height.
- Fill Height: If checked, the extracted outline result will be snapped to the ground, and the building height will be written to the BuildingHeight field of the result layer.
Advanced
- Building Gap: The distance interval between buildings, used to separate buildings.
- Minimum Building Area: Outlines smaller than the minimum building area will be filtered out.