AI Forestry Classification requires a base LP360 (Geospatial, Drone, or Land) license and AI Forestry license - assigned to the user.
The AI Forestry is part of the Advanced AI Classification. The price of the tool and how to use it, is explained in the Advance AI Classification article.
The AI Forestry model applies a generalized deep learning model to classify ground, canopy, tree trunks (including fallen trees if present) and noise points. It assumes unclassified input and will overwrite all existing classes including Class 2 (Ground).
The workflow will also extract feature layers from the classified point cloud and produce a set of forestry-related SHP files (see full list below) and add them back to LP360.
A normalized canopy height model (CHM) showing height above ground of the canopy surface as a raster (GeoTIFF) will also be generated and added back to LP360.The AI Forestry generates a complete forestry analysis. The outputs are:
| Output | Format | Description |
| Classified LAS | LAS | Reclassified forestry point cloud |
| tree_trunk_fitted_centers_at_1.3meter | SHP | Trunk center locations at DBH measurement height |
| tree_trunk_fits | SHP | Fitted trunk axis geometry |
| tree_trunk_centers_and_radia | SHP | Trunk center measurements with radius and diameter |
| crown_points_initial | SHP | Initial crown candidate points |
| crown_shapes_final | SHP | Final delineated crown polygons |
| CHM | GeoTIFF Raster | Canopy Height Model |
Classified LAS
The classified LAS is the primary output of AI Forestry. It contains the point cloud classified into forestry-specific classes.
| Class | Name | Description |
| 1 | Unclassified | Points not assigned to a forestry class |
| 2 | Ground | Ground surface |
| 3 | Low Vegetation | Vegetation and canopy points |
| 18 | High Noise | Noise or outlier points |
| 31 | Tree Trunks | Points classified as tree trunks |
| 32 | Fallen Trees | Points classified as fallen trees |
tree_trunk_fitted_centers_at_1.3meter shapefile
This shapefile contains fitted trunk center locations calculated at 1.3 meters above ground, corresponding to the standard forestry DBH measurement height. This layer provides a point-based reference for detected trunks.
Geometry Type: 3D Point
| Field | Description |
| n_p_all | Total number of points associated with the detected trunk object |
| n_points | Number of points used in the trunk fitting calculation |
| length | Estimated trunk segment length in meters |
| trunk_id | Unique identifier for the detected trunk |
| tree_id | Unique identifier for the associated tree |
| Parts | Geometry component count |
| Vertices | Number of vertices in the geometry |
| Type | Geometry type (Point 3-D) |
| Index | Internal feature index |
| Point X | X coordinate |
| Point Y | Y coordinate |
| Point Z | Z coordinate |
tree_trunk_fits shapefile
This shapefile contains fitted line representations of the detected tree trunks. Instead of individual trunk points, this layer models the estimated trunk axis. This output represents modeled trunk direction and structure.
Geometry Type: 3D Line
| Field | Description |
| n_p_all | Total points associated with the trunk detection |
| n_points | Number of points used in fitting |
| length | Estimated trunk length |
| trunk_id | Unique trunk identifier |
| tree_id | Associated tree identifier |
| trunk_angl | Estimated trunk angle/orientation |
| trunk_str | Trunk structural metric |
| Parts | Number of geometry parts |
| Vertices | Number of vertices defining the line |
| Type | Geometry type (Line 3-D) |
| Index | Internal feature index |
tree_trunk_centers_and_radia shapefile
This shapefile contains measured trunk center locations with estimated trunk dimensions. This is one of the most relevant outputs for forestry inventory analysis.
Geometry Type: 3D Point
| Field | Description |
| n_p_all | Total points associated with the detected tree object |
| n_points | Points used for measurement |
| length | Estimated trunk segment length |
| trunk_id | Unique trunk identifier |
| tree_id | Associated tree identifier |
| height | Measurement height (typically 1.3 m) |
| radius | Estimated trunk radius |
| diameter | Estimated trunk diameter |
| r_redchi | Radius fit residual/error metric |
| Parts | Geometry component count |
| Vertices | Number of vertices |
| Type | Geometry type (Point 3-D) |
| Index | Internal feature index |
| Point X | X coordinate |
| Point Y | Y coordinate |
| Point Z | Z coordinate |
crown_points_initial shapefile
This shapefile contains the initial crown point detections identified during the canopy segmentation process. These are intermediate points used before generating final crown polygons.
Geometry Type: 3D Point
| Field | Description |
| FID | Feature identifier |
| Parts | Geometry part count |
| Vertices | Vertex count |
| Type | Geometry type (Point 3-D) |
| Index | Internal feature index |
| Point X | X coordinate |
| Point Y | Y coordinate |
| Point Z | Z coordinate |
crown_shapes_final shapefile
This shapefile contains the final delineated crown polygons. Each polygon represents an estimated tree crown footprint.
Geometry Type: 2D Polygon
| Field | Description |
| crown_i | Crown identifier |
| area | Crown footprint area |
| height | Estimated crown height |
| points | Number of canopy points contributing to the crown |
| p_10 | 10th percentile canopy height |
| p_20 | 20th percentile canopy height |
| p_30 | 30th percentile canopy height |
| p_40 | 40th percentile canopy height |
| p_50 | Median canopy height |
| p_60 | 60th percentile canopy height |
| p_70 | 70th percentile canopy height |
| p_80 | 80th percentile canopy height |
| p_90 | 90th percentile canopy height |
| Parts | Number of polygon parts |
| Vertices | Number of polygon vertices |
| Type | Geometry type (Polygon 2-D) |
| Index | Internal feature index |
| Perimeter | Crown perimeter |
| Area (m²) | Crown area in square meters |
| Area (hectares) | Crown area in hectares |
| Direction | Polygon orientation |
The percentile fields describe the vertical canopy distribution. For example:
- p_10 = lower canopy structure
- p_50 = median canopy height
- p_90 = upper canopy extent
Canopy Height Model Raster (CHM)
This raster represents canopy height normalized above the ground surface. It is derived from the classified LAS and provides a visual representation of canopy elevation relative to terrain.
Comments
0 comments
Please sign in to leave a comment.