Lidar sensors generate 3-D scans of their surrounding environments as collections of points in space called point clouds. Though point clouds are accurate and robust, which makes them useful for robotics applications, raw point cloud data is large, contains high density noise, and has a scattered distribution. Lidar Toolbox™ includes preprocessing features that enable you to better to store and use point clouds.
Lidar Toolbox includes preliminary processing algorithms to downsample, filter, transform, align, block, organize, and extract features from point clouds. These algorithms improve the quality and accuracy of the data, and can accelerate and improve the results of advanced workflows.
When your point cloud data is too large to process at once, you can divide and process the point cloud as small blocks by using the
For advanced workflows that require organized point clouds, such as object detection, and segmentation, you can convert unorganized point clouds to the organized format by using the
pcorganizefunction. For more information on the distinctions between organized and unorganized point clouds, see What are Organized and Unorganized Point Clouds?
Lidar Toolbox includes functions that generate surface meshes, digital elevation models (DEM) and 2-D scans from point cloud data. You can also create and process surface mesh data by using the
surfaceMeshobject. Lidar Toolbox includes functions that read, write, and visualize a surface mesh.
You can also interactively visualize, analyze, and preprocess point cloud data using the Lidar Viewer app.
|Visualize and analyze lidar data (Since R2021b)
|Downsample a 3-D point cloud
|Median filtering 3-D point cloud data (Since R2020b)
|Remove noise from 3-D point cloud
|Find points within a cylindrical region in a point cloud (Since R2023a)
|Remove invalid points from point cloud
|Remove hidden points from point cloud (Since R2023a)
|Align array of point clouds (Since R2020b)
|Concatenate 3-D point cloud array (Since R2020b)
|Estimate normals for point cloud
|Transform 3-D point cloud
|Undistort point cloud affected by ego motion (Since R2023a)
Organize Point Cloud
Find Points in Point Cloud
Detect and Extract Features
|Extract eigenvalue-based features from point cloud segments (Since R2021a)
|Extract fast point feature histogram (FPFH) descriptors from point cloud (Since R2020b)
|Detect ISS feature points in point cloud (Since R2022a)
|Detect LOAM feature points from 3-D lidar data (Since R2022a)
|Detect rectangular plane of specified dimensions in point cloud (Since R2020b)
|Detect road angles in point cloud (Since R2022b)
Register Point Clouds
|Register two point clouds using LOAM algorithm (Since R2022a)
|Register two point clouds using FGR algorithm (Since R2022b)
|Register two point clouds using ICP algorithm
|Register two point clouds using CPD algorithm
|Register two point clouds using phase correlation (Since R2020b)
|Register two point clouds using NDT algorithm
Convert Point Cloud
Surface Mesh Workflow
|Create surface mesh (Since R2022b)
|Construct surface mesh from 3-D point cloud (Since R2022b)
|Read 3-D surface mesh data from STL or PLY file (Since R2022b)
|Write 3-D surface mesh into STL or PLY file (Since R2022b)
|Display surface mesh (Since R2022b)
|Smooth surface mesh (Since R2023a)
|Cluster connected faces (Since R2023a)
- Introduction to Lidar
High-level overview of lidar concepts and applications.
- Coordinate Systems in Lidar Toolbox
Overview of coordinate systems in Lidar Toolbox.
- What are Organized and Unorganized Point Clouds?
Define unorganized and organized point clouds and how to convert the former to latter.
- Get Started with Lidar Viewer
Interactively visualize and analyze lidar data.
- Create Custom Preprocessing Workflow with Lidar Viewer
Create custom preprocessing workflows for interactive use within the app.
- Estimate Transformation Between Two Point Clouds Using Features
This example shows how to estimate a rigid transformation between two point clouds.
- Estimate Stockpile Volume from Aerial Lidar Data
This example shows how to estimate the volume of a stockpile from aerial point cloud data.