Automatic Tree Trunk Classification – Global Mapper

Automatic Tree Trunk Classification is found in the Classification section of the Automatic Point Cloud Analysis tool.

This function requires Global Mapper Professional

The Tree Trunk Classification tool detects and classifies points representing trunks of trees of user-specified size. The classification uses a segmentation-based Max Likelihood method, developed for modern, dense point clouds such as collected by terrestrial, programmatic, and drone-mounted lidar hardware. The model segments the point cloud into clusters of points (called segments) and then each segment of points that meets the requirements of the tree trunk model (as configured by the classification settings) has the Tree Trunk classification (lidar class 74) applied. Additional segmentation parameters are available in the Segmentation tool.

Successful Tree Trunk Classification requires sufficient point density and scanner angle such that the circumference of the trunk is visible in the point data. You can filter your point cloud by elevation or height above ground to check that the trunks have been sufficiently captured by the scan.

Use the settings in the top of the Automatic Point Cloud Analysis window to select which layers are to be processed, specify the bounds of the processing, filter by points by attributes, or enable a Live Classification Preview on points currently selected by the Digitizer. You can choose to run multiple classifications at once by checking multiple options (noise, ground, etc). They will run in a pre-specified order based on method. Once your settings have been determined for all desired classifications, click Classify Features to begin processing.

Note: Prior to using the Tree Trunk classification tool, the point cloud should have ground classified. If the ground is not classified, check the box to run ground classification as well. Well-classified ground points are necessary for correct identification of non-ground features.

Classification and Extraction Shared Settings:

The Tree Trunk classification shares multiple settings with the Feature Extraction tool:

Resolution

Resolution (or Neighborhood Range) defines the spatial radius around a target point used to compute its local geometric and surface properties (such as slope, curvature, or planarity). Set the Resolution that will be used for determining the localized neighborhood of points evaluated during Principal Component Analysis (PCA) to generate reliable statistical properties for a surface. It must be set roughly to the physical scale of the feature being classified.

Due to the point density required for trunk classification, you will likely set this in meters or feet instead of point spacings. Make sure the Resolution value is low enough to identify the desired size trunk; trunks under .5 meter in diameter will require a Resolution set at .5 meter or lower. The equivalent size in "point spacings" is displayed below; this value is data-dependent, but as a rule of thumb, ensure the Resolution is large enough that it equivalent to at least a few point spacings. Changing the Resolution will have the largest impact on the classification results.

Min DBH

Set the minimum required diameter at breast height. Only segments of points that meet this minimum diameter will receive the Tree Trunk classification.

Max DBH

Set the maximum required diameter at breast height. Segments of points larger than this will not receive the Tree Trunk classification

Min Length

Set the minimum length required for a segment of points to be classified as Tree Trunk.

Max Length

Set the maximum length for a segment of points to be classified as Tree Trunk.

Min Height Above Ground

Set a minimum height above ground for the classification; points below this threshold will be ignored.

Classification

Check the box next to Tree Trunk to enable classification.

Most of the parameters from the Shared Settings (above) are mirrored in the Classification parameters; you may enter values in either place.

Reset Existing Points to Unclassified at Start

Resets any previously classified Tree Trunk points to Unclassified before running the classification. Recommended when re-running the tool so that any false positives can be reevaluated.

Additional Options

Enable Report

Enable this option to save a text summary of the classification settings and results to the Reports folder, and export a JSON settings file for future reuse or sharing.
To load the saved settings .json file: Open the automated Point Cloud Analysis tool, check the box to Enable Custom Feature Models, then from the list of options that appear, choose Load Models.

Report information

Class Changes: How many points of each classification category were reassigned a different classification.
Input : Input point cloud layers and how many points were processed.
Classifier Setup: Classification settings as entered into the Automatic Point Cloud Analysis dialog.
Report Artifacts: File path to the saved report and json model.

The following report items pertain to segmentation-based (Max-Likelihood) classification methods, and may or may not appear, depending on the results of the classification:

Processing: Models run and any tiling of the processing
Classification Health: Provides information about the results. For example, "Healthy" is a successful classification, versus "Critical" can indicate no points were assigned and suggestions for improving results. "Notable" can indicate that segments best matched a class that was not enabled in this run.
Unclassified reasons: When a segment doesn't get a class, the report explains why:
- LowConf (Low Confidence): The segment best matched that class, but its probability score fell below the classifier's required confidence threshold (e.g., segments looked like Medium Vegetation, but were not clear enough to officially classify).
- NotAssign (Not Assigned / Recipe Disabled): The segment matched a class, but that class was not enabled in your current run settings (e.g., segments matched Pole, but you only ran the Building classification recipe).
- Other: The segment was excluded due to geometric, height, or spatial filtering criteria (such as falling outside specified area or footprint limits).
Per-class confidence: Shows the distribution of the classifier's top prediction confidence scores across each class, grouped into 10% percentage brackets (from <10% up to >=90%). Here, p1 refers to the highest probability score calculated for a segment's best-matching class.
Preferred class: Indicates which class the remaining unclassified segments are likely to be. Even though these segments failed to earn an official label (due to low confidence scores, disabled recipes, or geometric rules), the algorithm tracks which class was their top candidate.
Assigned with low confidence: Identifies point segments that were successfully assigned to a class, but where the classifier's top choice barely beat out the second-best candidate.
Columnar drop-reason breakdown: Summary explaining why potential vertical, column-like objects (such as utility poles) were rejected by the classifier rather than saved as confirmed features.

Use Custom Segmentation

The classification of vegetation points involves a statistical analysis (Principal Component Analysis) of clusters of nearby points to determine what is likely part of a tree feature. Points are gathered into clusters based on a measure of similarity between nearby points, with the intent to group points that belong to the same object or feature.

Once neighborhoods are established and quantified using Principal Component Analysis, a measure of point-to-point similarity is evaluated to guide clustering. This similarity measure is evaluated as a generalized statistical distance between points, based on the distributions defined by the statistics collected over their local neighborhoods. Clustering aims to group points with similar surface characteristics into groups that can be classified based on those characteristics.

The settings in the Geometric Segmentation tool can be used in Max Likelihood classifications. Simply open the Geometric Segmentation tool, choose your settings, and check the box to Use Custom Segmentation Parameters. These settings will influence how the point cloud is segmented.
To train the algorithm for your specific feature, check the box to enable Custom Feature Models. and train a model using sample segments.

Tree trunk extraction is discussed on Features to Extract.