Get Started with Lidar Viewer

Use the Lidar Viewer app to view, analyze, and perform preprocessing operations on lidar data. Use the app to prepare data for advanced workflows like labeling, segmentation, and calibration.

This topic provides an overview of the app workflow and underlying features. To open the app, at the MATLAB^® command prompt, enter this command.

lidarViewer

Alternatively, you can open the app from the Apps tab, under Image Processing and Computer Vision.

Load Data

The Lidar Viewer app can import pointCloud objects from the workspace and read point cloud data from PLY, PCAP, LAS, LAZ, PCD, and rosbag files. You can load lidar data from multiple sources at once. Use this process to load data into the app:

• On the app toolstrip, select Import > From File. Choose a data source from the list.

• In the Import dialog box that appears, specify the location of the point cloud data from the selected data source.

  • Point Cloud Sequence — In the Provide path to a folder containing PCD/PLY file(s) box, specify the path to the folder containing your point cloud data. Alternatively, select Browse, browse to the folder containing your data, and then select Select Folder.

  • Velodyne Lidar — For Velodyne^® lidar data, select the device model name from the Device Model list. In the Provide path to a calibration file box, specify the path to the calibration file of the sensor.

    

  • LAS/LAZ File— In the Provide path to a LAS/LAZ file box, specify the path to the folder containing your point cloud data. Alternatively, select Browse, browse to the folder containing your data, and then select Select Folder.

  • Rosbag — For rosbag files, select the topics that contain point cloud data from the Point Cloud Topics list. To load data from a rosbag file, you must have a ROS Toolbox license.

    

  • Custom Point Cloud — To load point cloud data from a custom source, create a custom reader function and add it to your MATLAB path. In the Custom Reader Function box specify the path to the reader function, and in the Source Name box specify the path to the data folder.

    

• Alternatively, select Import > From Workspace on the app toolstrip. In the Import From Workspace dialog box that appears, select the variables/ objects from the workspace you want to import.

  

The app loads and plots point cloud data in the Point Cloud Display pane.

Data Visualization

The Lidar Viewer app organizes loaded data, based on the order you import it, as a list in the Data Browser pane. To delete the imported data, right click on the data you want to delete and select Delete Data.

Select the data that you want to visualize. The metadata of the point cloud populates in the Point Cloud Information pane. The information and fields change based on the data source and available metadata.

The visualization pane displays the point cloud data along with playback controls in the Playback Slider.

The Playback Slider pane contains a slider that indicates your current position in the point cloud sequence, as well as buttons to control the playback of the point cloud data. The buttons, from left to right, are:

• First Frame

• Previous Frame

• Play

• Next Frame

• Last Frame

If you select Play, the Play button becomes the Pause button, and all the playback controls except Pause are disabled.

For point cloud data with timestamp information such as PCAP, rosbag files, the app displays start time, current time, and end time on the Playback Slider pane.

You can adjust the point cloud view using the x-, y-, z-axes at the bottom left corner of the point cloud pane.

Color Controls

Lidar Viewer provides various visualization features to analyze point cloud data. The app uses color to visualize more details about the displayed point cloud.

You can control the color of the displayed point cloud by using these options in the Color section of the app toolstrip:

 Colormap

Choose the color profile for the point cloud data, from these options.

The app enables the Color option only when the input point cloud data has color information.

 Colormap Value

Choose how the color profile applies to the point cloud, from these options:

• Z Height — The color changes as the distance between points increases.

• Radial Distance — The color changes as the distance between points and the ego vehicle increases.

• Intensity — The color changes as the intensity value of points increases.

• Classification — The color changes based on the classification value of the point. For more information, see Classification.

  

• LaserReturn — The color changes as the number of laser returns increases. The return number is the number of times a laser pulse reflects back to the sensor.

• ScanAngle — The color changes as the sensor scan angle increases. The scan angle is a value in degrees between –90 and 90. At 0 degrees, the laser pulse is directly below the aerial lidar sensor. At –90 degrees, the laser pulse is to the left side of the sensor, relative to the direction of flight. At 90 degrees, the laser pulse is to the right side of the sensor, relative to the direction of flight.

  

• GPSTimeStamp – The color changes as the timestamp of a point increases.

Note

The app enables all the above options for Colormap Value with LAS/ LAZ data only. The default options for any other data format include Z Height, Radial Distance, Intensity.

 Color Variation

Choose the color variation for the point cloud data as Linear or Custom.

When you select Linear, the color variation is linear along the z-axis of the point cloud. When you select Custom, you can choose from one of these options. You can also modify the color variation using the colorbar below the graph.

 Point Size

Adjust the display size of points.

 Background Color

Select the background color for the point cloud display.

Camera View Options

The app provides various predefined camera angles for viewing the point cloud data, as well as the option to create custom views:

 XY View

View the xy-axes of a point cloud. This is the top view of the scene, line of sight is along z axis.

 YZ View

View the yz-axes of a point cloud. This is the front view of the scene, line of sight is along x axis.

 XZ View

View the xz-axes of a point cloud. This is the side view of the scene, line of sight is along y axis.

 Bird's Eye View

View from a high angle above a point cloud.

 Chase View

View the point cloud from a fixed distance behind the ego vehicle (actor).

 Ego View

View a point cloud from the perspective of the ego vehicle.

 Ego Direction

Use the Ego Direction list to select the direction the camera faces for Ego View and Chase View.

 Custom Views

Select Custom Views to save and reuse custom views of the point cloud data. You can interactively rotate, pan, and zoom the camera to create a view, then save the view by clicking Custom Views and selecting Save Camera View. Specify a name for the view and select OK. You can return to the saved view at any time by clicking Custom Views and selecting the saved view from the list. Select Organize Camera Views from the list to delete or rename the saved views.

 Restore Default View

Restore the point cloud display to the default view.

 Hide Ground

Select Hide Ground to remove ground points from the point cloud. This enables the Ground Settings option. Click Ground Settings to open a dialog box in which you can select from these ground removal algorithms and specify their parameters.

• Select Range-based floodfill to remove ground points using the segmentGroundFromLidarData function.

• Select Fit ground plane to fit and filter the ground plane using the pcfitplane function.

• Select Segment ground SMRF to filter the ground points using the segmentGroundSMRF function. The app enables this option for only organized point clouds.

To visualize the segmented ground and nonground planes on the point cloud, select View Ground Data.

 View Clusters

Select View Cluster to view point cloud clusters. This enables the Cluster Settings option. Click Cluster Settings to open a dialog box in which you can select from these cluster-based coloring algorithms.

• Select Range-based clustering to cluster point cloud data using the segmentLidarData function.

• Select Distance-based clustering to cluster point cloud data using the pcsegdist function.

• Select k-means clustering to perform k-means clustering on the points.

You can specify the algorithm parameters in the dialog box, and visualize the output point cloud clusters.

Measure Point Cloud

Select the Measurements option from the app toolstrip. The app opens a Measurements tab. You can measure point cloud attributes such as distance, angle, and volume using the measurement tools in this tab.

To measure an attribute of the point cloud, select a measurement tool on the app toolstrip, then interactively select points on the point cloud to return the corresponding measurement.

• Distance — Measure the distance between any two points in the point cloud.

• Elevation — Measure the elevation between any two points in the point cloud.

• Location — Obtain the xyz-coordinates of a point.

• Angle — Select three points on the point cloud to measure the angles of the triangle defined by the points.

• Volume — Measure the volume inside a cuboid region on the point cloud. You can interactively adjust the limits and the size of the cuboid.

The app deletes all your measurements in the current signal when you exit the Measurements tab.

Keyboard Shortcuts

You can use these keyboard short cuts in the Measurements tab of the Lidar Viewer app.

• Click Esc to undo your measurement tool selection.

• Click R to enable rotation. You can use these controls in the rotate mode.

  Action	Keyboard Shortcut	Example
  Move forward or backward	w — Forward s — Backward
  Move left or right	a — Left d — Right
  Look around the scene (pitch/yaw rotation of the camera)	Up arrow — Camera pitches up Down arrow — Camera pitches down Left arrow — Camera turns left in yaw Right arrow — Camera turns right in yaw
  Roll the scene (roll rotation of the camera)	q — Camera moves clockwise. (Scene appears to move counterclockwise.) e — Camera moves counterclockwise. (Scene appears to move clockwise.)
  Rotate the scene (rotate the camera around the point cloud)	Shift + Left or Right — Camera orbits horizontally around the point cloud. (Scene appears to stand still as camera moves externally around it horizontally.) Shift + Up or Down — Camera orbits vertically around the point cloud. (Scene appears to stand still as camera moves externally around it vertically.)
  Zoom in or out	z — Zoom in x — Zoom out
  Look at a view plane	1 — XY 2 — YX 3 — XZ 4 — ZX 5 — YZ 6 — ZY Camera positioned in the selected plane with a line of sight to the center of the plot.
  Switch between rotate and pan	Spacebar — Press and hold the spacebar to switch the interaction to pan. Release the spacebar to switch back to rotate.
  Rotate in X-, Y-, and Z- dimensions	T — Adds rotate interaction in the third dimension.

Edit Point Cloud

Apply preprocessing operations to a point cloud by selecting on the Edit Point Cloud from the app toolstrip. The app opens the Edit tab. This tab retains the color and visualization features present in the Lidar Viewer tab . You can select built-in preprocessing operations from the Algorithm section of the toolstrip:

 Denoising

Remove noise from a 3-D point cloud using the pcdenoise function.

 Downsampling

Downsample a 3-D point cloud using the pcdownsample function.

 Filtering

Median filter 3-D point cloud data using the pcmedian function.

 Cropping

Crop a 3-D point cloud using cuboids. You can interactively adjust the cuboid limits, or specify them using the Cuboid Limits option. Select a Crop Method to specify whether to crop inside or outside the cuboid. Then, click OK. You can add multiple cuboids using the Add Cuboid option, and visualize the cropped point cloud.

 Ground Removal

Segment the ground plane from 3-D point cloud data. You can select from these algorithms:

• Fit Ground Plane — Fit and filter ground plane using the pcfitplane function.

• Hide Ground — Hide ground using the segmentGroundFromLidarData function.

• Segment Ground — Segment ground using the segmentGroundSMRF function.

 Organizing

Convert the 3-D point cloud into an organized point cloud using the pcorganize function. For more information, see What are Organized and Unorganized Point Clouds?

After selecting an algorithm, the app populates the Algorithm Parameters pane with the corresponding tunable parameters.

Algorithm Parameters

The Lidar Viewer app dynamically updates the point cloud as you tune the parameters, enabling you to see the results in realtime. Select Apply All Frames to apply the algorithm to all the frames in the data source. After tuning your parameters select OK. The History pane records all preprocessing operations applied to the current frame. You can apply the same algorithm multiple times on the data using OK. Select Cancel to exit the edit algorithm.

You can discard all applied algorithms by selecting Discard Operations. You can also export the selected preprocessing steps and the parameters, as a function by selecting Export To Function. The app creates a MAT file containing your custom preprocessing function. The function accepts a pointCloud object as input and outputs the processed pointCloud object.

Custom Preprocessing Algorithms

You can create a new custom preprocessing algorithm or import an existing one into the app. Follow these steps to apply a custom preprocessing algorithm to your data.

1. Click Add Algorithm. To apply your algorithm to only a single point cloud frame, select an option from the Spatial Algorithms section. To apply your algorithm to multiple point cloud frames, select an option from the Temporal Algorithms section.

2. To create a new algorithm, select New. Select whether to create an algorithm using a Class Template or a Function Template. MATLAB opens a new MAT file that contains a code framework and directions to create your custom algorithm. With the template, you can also define user interface (UI) elements for parameter tuning. These UI elements appear in the Algorithm Parameters pane.

3. To import your algorithm into the app, first select Add Algorithm and, in the Spatial Algorithms or Temporal Algorithms section, select From File. Then, select Import Class or Import Function. In the dialog box, select the file that contains the algorithm you want to import.

To finalize your edits to the point cloud and return to the Edit tab, on the app toolstrip, select Accept.

You can also export the history of edit operations to a function-based algorithm using Export To Function.

For more information on creating custom preprocessing workflows, see Create Custom Preprocessing Workflow with Lidar Viewer.

Export Point Cloud

You can export point clouds as PCD or PLY files. After processing your point clouds, on the app toolstrip, select Export Point Cloud. Lidar Viewer opens the Export Point Cloud dialog box.

Select the point clouds you want to export. Then, in the Provide path to the destination folder text box, specify or browse to the destination folder.

Note

If the input point cloud data is in PLY format, the app exports it as a PLY file. If you load input data of any other format, the app exports them as PCD files.

Note

A Windows^® OS machine must have WebGL™ 2 support for graphics rendering.