An overview of the Point Cloud toolset

The Point Cloud toolset contains tools for classifying, managing, and processing LAS-format point cloud data. Esri's proprietary ground and building classification algorithms provide high-quality results that enable digital elevation model (DEM) production and 3D feature extraction workflows. Additional classification methods support:

• Classifying points based on height above the ground or a custom elevation surface.
• Classifying points using their proximity to point, line, or polygon features.
• Classifying points based on their overlay with a classified raster.

The toolset also includes deep learning tools for classifying and detecting objects from point clouds. This allows you to build and deploy custom solutions tailored for your aerial and terrestrial lidar and photogrammetric point clouds.

Additionally, the toolset provides functionality for performing the following data management operations:

• Divide large LAS files into more efficient tiles.
• Thin point clouds to eliminate hot spots of high density point collections which can degrade performance and enlarge the data size.
• Colorize aerial point clouds using overlapping imagery.

Tools in the Point Cloud toolset

Tool	Description
Colorize LAS	Applies colors and near-infrared values from orthographic imagery to LAS points.
Extract LAS	Creates LAS files from point cloud data in a LAS dataset or point cloud scene layer.
Thin LAS	Creates new LAS files that contain a subset of LAS points from the input LAS dataset.
Tile LAS	Creates a set of nonoverlapping LAS files whose horizontal extents are divided by a regular grid.

Tools in the Classification toolset

Tool	Description
Change LAS Class Codes	Reassigns the classification codes and flags of .las and .zlas files.
Classify LAS Building	Classifies buildings in LAS format point cloud data.
Classify LAS By Height	Reclassifies lidar points based on their height from the ground surface.
Classify LAS Ground	Classifies ground points from LAS data.
Classify LAS Noise	Classifies LAS points with anomalous spatial characteristics as noise.
Classify LAS Overlap	Classifies LAS points from overlapping scans of aerial lidar surveys.
Set LAS Class Codes Using Features	Classifies LAS points that intersect the two-dimensional extent of input features.
Set LAS Class Codes Using Raster	Classifies LAS points using cell values from a raster dataset.

Tools in the Classification (Deep Learning) toolset

Tool	Description
Classify Point Cloud Using Trained Model	Classifies a point cloud using a deep learning model.
Evaluate Point Cloud Classification Model	Evaluates the quality of one or more point cloud classification models using a well-classified point cloud as a baseline for comparing the classification results obtained from each model.
Prepare Point Cloud Training Data	Generates the data that will be used to train and validate a point cloud classification model.
Train Point Cloud Classification Model	Trains a deep learning model for point cloud classification.

Tools in the Conversion toolset

Tool	Description
LAS Dataset To TIN	Exports a triangulated irregular network (TIN) from a LAS dataset.
LAS To Multipoint	Creates multipoint features using one or more lidar files.

Tools in the Object Detection (Deep Learning) toolset

Tool	Description
Detect Objects From Point Cloud Using Trained Model	Detects objects captured in a point cloud using a deep learning model.
Prepare Point Cloud Object Detection Training Data	Creates point cloud training data for object detection models using deep learning.
Train Point Cloud Object Detection Model	Trains an object detection model for point clouds using deep learning.