LIDAR Data processing

The article gives a practical example of laser scanning data processing to obtain
 an initial road model, suitable for road maintenance project execution.


LiDAR Data Processing
Introduction VRMesh provides a powerful point cloud classification and feature extraction solution for airborne and mobile LiDAR.It automatically extracts vegetation, buildings, and ground points from LiDAR data.99% classification jobs can be automatically done in one click by setting a few simple parameters.
Unlike traditional LiDAR data processing workflows that have to extract ground points first before classifying vegetation and buildings, the workflow of point cloud classification in VRMesh is to detect vegetation first, then extract building roofs, and finally clean ground points.A basic LiDAR processing workflow in VRMesh is shown below: In addition, VRMesh supports batch processing of multiple files.It is recommended that using a sample file to figure out a customized workflow with tailored parameter settings.The batch mode will process all files in a specified folder one at a time using the same settings.Output results will be automatically saved to a subfolder at the same location as the original folder.The suggested procedure is: 1. Using the Index/Attach command to retrieve inquiring points from a sample file into the window.2. Using the Detect Ground Points command to classify visible points on the screen into vegetation, buildings, and ground.The purpose of this step is to figure out the appropriate parameter settings for your project.3. Using the Survey Wizard to batch process multiple files in a selected folder.

Attach
There ar *.las, *.p Index/A compute more tha

Click the panel ap
Step 1: Press the and file t to the sa take a fe are read Step 2: Step 3: Refresh Display There are three ways to refresh the display.The Auto Update option automatically refreshes the display if the number of blocks shown on the screen is less than or equal to the value of Max.Block Update.The Lock IDs option allows you to define a list of blocks that will be displayed on the screen.The Lock by Selection option allows you to select blocks by directly drawing a box around a region of interest.
Note: You only need to create an indexed PCP file once for the same original file, which means that you can directly press the [Attach] button if the indexed file has already existed.www.vrmesh.com

Bare Earth Extraction
The creation of a bare-earth surface involves classifying vegetation, buildings, and ground points in LiDAR data and then generating a grid surface or a triangulated surface to represent the topography.The following instruction will guide you through the bare-earth generation process.
Step 1: Detect Ground Points Click the Detect Ground Points command in the Survey menu.A working panel appears in the Parameter Window on the right side of the screen.This is a composite command containing three steps for classifying point clouds:

Detect Vegetation
This is the first fundamental step in VRMesh workflow for point cloud classification.The program calculates the roughness on point clouds to identify vegetation in LiDAR data.Normally, the value of roughness is set to 0.30 for an urban area, and 0.6~0.8 for a mountainous terrain.The value of minimum points required for a valid region is normally 5~50.

Detect Building Roofs
This step automatically detects building roofs in point clouds.The maximum value of roof slope is normally set to 50~85.You also need to define the minimum height of a roof and the minimum / maximum number of points required in a roof.
You may ignore this step if your data is from a mountainous region without buildings.

Clean Ground Points
This step automatically deletes floating parts to get well classified ground points.There are two clean types.The Flat Region type is preferred to a flat urban region.The Steep Region type is preferred to a steep mountainous region.Both methods are working well for point clouds derived from airborne, mobile, terrestrial LiDAR, or from stereo imagery.
The composite command can be executed step-by-step or by one-click batch processing.Meanwhile, you may select the "Only visible points" checkbox to classify visible points on the screen only, which can be used to figure out the appropriate parameter settings for your project.
After the three steps, the ground points will be successfully identified.You may use the Show Points by Classification command in the Analyze menu to view each class of points.If ground points are not enough, you may use the Add Point to Ground command to assign more points to the ground class.
Step 2: Create a Bare-Earth Surface There are two ways to generate a bare-earth surface in VRMesh.You may directly create a grid surface to represent the bare earth, which is simple and fast.Or, you may choose to triangulate point clouds into meshes, which is more accurate in representing the bare-earth topography.
For creating a grid surface, click the Create Grid Surface command in the Survey menu.Specifying the cell size and sampling method to create a grid surface from the visible points on the screen.You can use the Show Points by Classification command in the Analyze menu to control which class of points are shown on the screen.Please note that a grid surface generated using this command may not accurately represent the topography of a steep mountain.It's better to create a meshed surface for steep terrain.
For creating a triangulated bare-earth surface, use the Split Classified Points command in the Survey menu to split the ground points.Then, use the Denoise Point Cloud command in the Cloud menu to smooth the point cloud.And finally, use the Point Cloud to Mesh command in the Cloud menu to convert the point cloud into a triangle mesh.If the generated mesh has a large number of triangles, you may use the Decimation command in the Modify menu to reduce the number of triangles into an acceptable amount depending on your computer system.www.vrmesh.com

Building Footprint Extraction
VRMesh provides an automatic solution for extracting building footprints from LiDAR data and generating 3D building models.It doesn't require a good bare earth before identifying building roofs in VRMesh workflow.Therefore, VRMesh can successfully extract building footprints from areas where there is very few ground points available.The following instruction will guide you through the footprint extraction process.
Step 1: Detect Vegetation Click the Detect Vegetation command in the Survey menu.Specify the parameter settings.Normally, the value of roughness is set to 0.22 for an urban area, and 0.6~0.8 for a mountainous terrain.The value of minimum points required for a valid region is normally 5~50.
Step 2: Detect Building Roofs Click the Detect Building Roofs command in the Survey menu.The maximum value of roof slope is normally set to 50~85.You also need to define the minimum height of a roof and the minimum/maximum number of points required in a roof.

Step 3 (Optional): Clean Ground Points
Click the Clean Ground Points command in the Survey menu.The command automatically deletes floating parts to get well classified ground points.There are two clean types.The Flat Region type is preferred to a flat urban region.The Steep Region type is preferred to a steep mountainous region.

Note: You may use the composite command Detect Ground Points
in the Survey menu to finish the above three steps in one click.
Step 4: Extract Building Footprints Once you get the classified building roofs, click the Extract Building Footprints command in the Survey menu.Specify the number of points in each grid cell, normally 8 ~25, which is used to create a grid surface for footprint extraction.Define a minimum area for a valid roof.Additionally, there are two options for squaring up a footprint.If All Polygons is selected, the program will refine all corners to 90 degree.If Quadrilaterals is selected, the program will only refine the corners of quadrilateral shaped polygons to 90 degree.
Step 5: Create 3D Buildings The program can automatically create 3D building models based on the building footprints.You may select the Create Buildings checkbox in parameter settings of the Extract Building Footprints command, or click the Create Building from Footprint command in the Survey menu.You can define a roof shape as an Approximate type, or a Flat type.The approximate type means the roof shape is close to the actual shape.The flat type means all roof shapes are horizontal.You may also specify the maximum number of triangles to represent a roof.
Step 6: Export You can export building footprints or 3D building models to a ShapeFile or a DXF file.

Click the appears automat
Step 2 (

Those hi step 1. I you can you may the Surv
Step 3: Step 5: Analyze Clearance The Classify by Tunnel command in the Survey menu allows you to classify points based on the shape of a tunnel, which can be used to conduct a clearance analysis between powerlines and surrounding vegetation.The program will use a tunnel section template to classify points which are inside the tunnel, outside the tunnel or close to the tunnel section.You need to select a catenary curve to represent the centroid of a tunnel before using this command.If the Apply to All Curves option is selected, the program will execute the command on all catenary curves.In addition, the visibility and transparency of the tunnel object can be adjusted using the right-mouse menu in the Object Tree Window.

Point C
Learn ho Step 1:

VRMesh distribute vegetatio
The sugg is 3 million.The attached file appeared in the Object Tree Window will be marked with an asterisk, e.g., *Object.
www.vrmesh.comnormally set to a little bit more than half of the width of the powerline.Define a search gap between consecutive hits on a wire.It is recommended that you first try a relative short search gap for the detection.If the Apply to All Curves option is selected, the program will execute the command on all catenary curves.
After automatic classification, you may need to adjust some wrong classified points.VRMesh Survey provides several manual classification tools to reclassify those points.You can classify points on the fly in a cross section view or directly in the Main Window.The manual classification tools include: 1. Classify in section viewport Click the Classify in Section Viewport command (Survey > Classify in Section Viewport), or click the icon in the Toolbar.A widget will be shown in the Main Window.Move/Resize the widget to define a region.Double-click on the widget to create a cross sectional profile shown in the Cross Section Viewport.Then, select From class and To class.Click the [Classify by Polygon] button, draw a polyline to surround target points in the Cross Section Viewport, and double-click to start the classification.This tool allows you to classify target points accurately.