Symbolizing LiDAR data using points

Click on a point somewhere to the north of the stadium. Then click on another point somewhere south of the stadium, and the view will change to show a profile of points along that line.

Using the Measure tool, determine the height difference from the field level to the top of Death Valley stadium.

Q. How tall is the stadium? Are both sides equal in height?

Close the Measure Distance window and Profile Viewing window.

3D visualization helps to clarify where the first, second, third, and fourth returns occur. Change the Symbology to Return and adjust the Color scheme as needed.

Q. Where do you notice multiple returns happening in 3D space?

In the images below, notice that there are lots of multiple returns in trees, shown by the red, orange, and yellow points. There are very few around buildings or in open areas, such as Littlejohn Coliseum on the right.

Save the project.

6. Classifying and Reclassifying the Point Cloud (video link)

Raw data from a LiDAR scanner does not have any classification applied. Classification codes are added to each point using a combination of processing tools and manual classification. There are several tools for classifying point cloud data in ArcGIS Pro.

One example is using vector data to apply class codes by overlay. For example, we have building footprints for Clemson and can overlay these on the point cloud. Points within the footprints will be assigned the building class code. There will be discrepancies between which buildings are present in the LiDAR data from 2011 and building footprints from 2016, but most of the classification will be correct.

In the Catalog pane, go to Using_Visualizing_LiDAR_in_GIS.gdb and expand the geodatabase. Add the layer titled Buildings_CU_2016_2D.

Click on the Clemson_Campus_Projected.lasd in the Contents panel. In the LAS Dataset Layer group and Classification tab, click on Feature Proximity and 2D Proximity.

Set the following parameters for the tool:

Input LAS Dataset: Clemson_Campus_Projected.lasd

Input Feature Class: set Features to Buildings_CU_2016_2D\Buildings_CU_2016.

Buffer Distance: 2

New Class: 6 which represents Buildings in the LAS standard.

Click Run.

When the tool completes, remove the LAS dataset from the scene and add it once again from the Catalog pane. View the result using a Classification symbology.

In some cases, you may have LiDAR data without any classification codes. In ArcGIS Pro, there are tools for classifying points representing the ground surface, buildings, and vegetation. Let’s try it out.

First, we will use a geoprocessing tool to perform automatic classification of ground points. In a nutshell, the tool will identify a subset of points that are reasonably flat and smooth without spikes (trees) or big discontinuous changes in elevation (rooftops).

From the LAS_Files folder, add the 20110319_SC_PickensCo_2011_000041_NOCLASS.las to the scene from the Catalog pane.

Click on it in the Contents pane and in the Classification group, click on Automated Classification> Classify Ground.

Leave the Ground Detection Method set to Standard Classification. This will allow for some undulation in topography. Click Run.

Now let’s classify buildings using the automated tool.

Select 20110319_SC_PickensCo_2011_000041_NOCLASS.las in the Contents pane, go to Classification > Automated Classification > Classify Buildings. Accept the default options and click Run.

When the tool finishes, remove the LAS file from the Contents pane, add it once again, and set the Symbology to Classification. View the result, filtering to only show Building points.

Scenario: A large flag is to be raised to commemorate Clemson’s 125th anniversary. The goal is for the flag to be visible across as much of campus and the immediate surroundings as possible, but needs to be placed on an existing building. It will ideally be visible from the three primary highways leading to Clemson: highway 123 from the west, highway 93 from the east, and highway 76/28 from the south. Where should it be placed to meet most, if not all, of these criteria? Can it be done from a single location? Where would you try first?

Solution: We can do a line-of-sight analysis with our point cloud to evaluate whether some feature is visible from another point. Performing a viewshed analysis using our DSM of Clemson and the proposed location(s) will show us what regions are visible. From this, we can determine whether the goal is feasible using our proposed locations.

To start, we will use the Line of Sight tool to evaluate whether Tillman Hall can be seen from the three highways leading to campus.

From the Using_Visualizing_LiDAR_in_GIS geodatabase, add the Sight_Lines feature class, which contains 3D line features connecting each highway to the top-most point representing Tillman Hall.

Click on the Clemson_Campus_Projected.lasd in the Content pane and then the Data tab. Click Visibility > Line of Sight to open the tool.

Set the Sight_Lines as the Input Line Features, and save the output as Tillman_LOS. Click Run and then view the result.

The lines are split into segments, with a green color indicating a visible segment from the observation point and red color indicating not visible.

Input Raster: Clemson_DSM

Input Observer Features: click the pencil icon and then click Points to draw in an observer feature. Navigate to Tillman Hall and draw in a point as near as you can to the top of Tillman Hall, using the LiDAR points as a guide.

Output raster: Tillman_viewshed in the geodatabase.

Click Run and then view the result, which will be symbolized by visibility.

Scenario: The University wants to build an array of solar panels on a building rooftop on campus and needs to know which buildings would be best suited for this purpose.

Solution: We can calculate the insolation (solar radiation) received on each of the buildings using the DSM, which gives us the elevation as well as the aspect (direction) and slope of the buildings. The LiDAR data provide high-resolution elevation information to calculate incoming solar radiation varies across individual rooftops, enabling us to optimize the placement of solar panels.

To begin our analysis, insert a new Map view (go to Insert > New Map > New Map). Add the Clemson_DSM raster and the Buildings_CU_2016_2D dataset to the map. Save your project.

We will use the Area Solar Radiation tool to calculate incoming energy. We will calculate for a specific day – the winter solstice on December 21, to show the minimum amount of energy expected. This calculation is fairly intensive, so to speed things up, we will clip out our DSM to only those areas where we have buildings.

In the Geoprocessing pane, search for and open the Extract by Mask tool. Use these tool parameters:

Input raster: Clemson_DSM.

Input raster or feature mask data: Buildings_CU_2016_2D/Buildings_CU_2016_2D.

Output raster: Bldgs_DSM Click Run.

The output raster is clipped to only the locations of the building footprints.

Next, search for and open the Area Solar Radiation (Spatial Analyst) tool in the Geoprocessing pane. Use the following parameters:

Input raster: Bldgs_DSM. Notice that the Latitude value automatically populates.

Output global radiation raster: Clemson_Solar_Rad in the geodatabase.

Time configuration to Within day, and select day 355 (December 21).

Explore the other parameters for the tool, then click Run. This may take a few minutes to process.

The Clemson_Solar_Rad raster is added to the map showing the insolation received by building rooftops.

9.Exercise on your own: Determining Forest Canopy Height

You can add new classification codes to your LAS files based on their height. In this way, the points themselves can be quickly visualized using the tools in ArcGIS as well as processed a bit differently.Hint1: Inspect the source LAS file, the statistics, and note the current classes. Class 4 will represent our ideal tiger-bird habitat.