Lesson 10b2: Exploring and Classifying Lidar Data in ArcGIS Pro

Video Transcript
Download Video
Right-click and save to download

Detailed Description

This lesson will cover additional lidar point cloud processing and analysis features in ArcGIS Pro. If necessary, please review lesson 10b1 for guidance with importing LAS files, filtering, and visualizing lidar point cloud data in ArcGIS Pro.

Please note that some of our URLs in the video have changed. All URLs beginning with 'viewer.nationalmap.gov/...' have been updated to 'apps.nationalmap.gov/...'. These URLs should reroute automatically, but if you have an issue replace 'viewer.nationalmap.gov/...' with 'apps.nationalmap.gov/...'.  If you have questions, please visit https://usgs.gov/NationalMap/data 

This lesson contains links to data hosted at an ftp site which was active when this video was published. That site is no longer functional.  Please substitute the following address in place of the ftp address:  https://rockyweb.usgs.gov/Training_Data/


Date Taken:

Length: 00:24:49

Location Taken: US


The National Map is a collection of mapping products and services produced by the USGS National Geospatial Program. The products and services are accessed via the internet through service calls and graphical user interfaces. The National Geospatial Program has released a series of “how to” videos for people who are working with The National Map data and services. The videos show how to use the services and interfaces to access data and tools for viewing, analyzing and printing geospatial data. Lessons 1 and 2 serve as an introduction to the training materials, and the remaining “how to” videos may be viewed as needed, in any order, depending on the user.

Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government.

Lesson 10b2: Exploring and Classifying Lidar Data in ArcGIS Pro. Lesson 10b1 introduced the basics of interacting with lidar point cloud data in ArcGIS Pro.

This lesson will cover additional lidar point cloud processing and analysis features in ArcGIS Pro. If necessary, please review lesson 10b1 for guidance with importing LAS files, filtering, and visualizing lidar point cloud data in ArcGIS Pro.

By the end of this lesson, you will be able to view LAS classes, select points to reclassify using polygon selection and class filtering, edit LAS classes with manual and automatic classification methods, and understand the fundamental differences of a LAS file and a LAS Dataset. In this tutorial we are using uncompressed LAS data. ArcGIS Pro does not have the ability to natively read LAZ files, but there is an interoperability extension in the ArcGIS Environment that can convert LAZ to LAS for additional cost. Please refer to Lesson 10e2 on how to convert LAZ files into LAS files for free.

If you’re interested in learning more about using USGS lidar data in other software packages, additional videos show how to use lidar data in Global Mapper and LP360. For this lesson, we will be using the same tiles of LAS data from Lesson 10b1. If you have not done so already, please download the USGS_LPC_CO_SoPlatteRiver_Data_for_Lessons.zip file from our FTP site at: ftp://rockyftp.cr.usgs.gov/Training_Data/.

Note that Google Chrome no longer supports ftp so you will need to enter this link into a file explorer search bar and navigate to the folder or use Microsoft Internet explorer to access the data.

After you’ve downloaded the lidar data, extract the zip file into a folder on your local computer to use during the lesson.

The National Map has a download client where you can find USGS products including elevation data such as lidar point clouds at https://apps.nationalmap.gov/downloader. If you are interested in learning more about downloading products on The National Map, be sure to check out our training videos located at http://www.usgs.gov/NGPvideos.

To begin, open ArcGIS Pro on your computer.

This exercise specifically uses ArcGIS Pro version 2.5. However, any version 2.1 or newer should suffice for this tutorial, but there may be a few interface and layout differences from version 2.5.

Use your enterprise or personal login information if prompted by the ArcGIS Pro log-in prompt. Once you have successfully logged in, click the Map tab under ‘New—Blank  Templates’. In the ‘Create A New Project’ window, name the project“ArcGIS_Pro_LiDAR_Training_Lesson_2”. Use the same folder we downloaded the lidar data to in the previous lesson 10b1. Then, click OK. 

On the map tab, in the Layer ribbon, select the add data icon. In the ‘Add Data’ window, browse to the location of the lidar point cloud LAS files that were downloaded from the FTP site. Click on the file name ending in ‘13SDD483396_LAS_2015.las’ and click ‘OK’. The map extent is zoomed out too far, so none of the points are able to draw. The red box shows the extent of the lidar point cloud data. To see the point data, zoom in or type 1:2000 in the map scale box in the lower left corner of the map.

The map now shows the point cloud data colored by elevation, which is shown by expanding the LAS layer dropdown in the Table of Contents. At this scale of 1:2000, only 10% to 15% of points are being drawn. This will vary based on your screen size.

If you have not already, highlight the LAS layer in the Table of Contents by clicking on it once. Now that the layer is activated, click the ‘Appearance’ tab and select the ‘Symbology’ dropdown and choose ‘Class’ under the ‘Symbolize your layer using points’ heading. In the symbology pane that has appeared, Under the ‘Draw Using’ dropdown, make sure it says ‘Classification’ and the ‘Draw Using’ box is checked. Read through the legend to see the various classes listed in the Table of Contents legend for the LAS layer. These classes include: ground, low vegetation, medium vegetation, noise, overlap, and high noise. Close the symbology pane when you’re finished.

Next, we will explore how ArcGIS Pro stores metadata for the LAS file. Right click the LAS layer and select ‘Properties’. If you select ‘Metadata’, you will notice that there is no information for this layer. Now click on the ‘Source’ tab. This tab shows the file locations, number of LAS points, and Vertical Units all stored under the ‘Data Source’ dropdown. Expand the ‘Spatial Reference’ dropdown. This shows the projected coordinate system : “NAD 1983 UTM Zone 13N”, and Vertical Coordinate System : “NAVD88 – Geoid 12A (Meters)”. On the layer Properties dialogue box, select the LAS Filter tab and explore the classification codes. Unfortunately, all of the codes listed pertain to the  ASPRS LAS version 1.4, which DO NOT pertain to this particular file’s LAS version. All classification codes are defined by the Americn Society for Photogrammetry and Remote Sensing (or ASPRS) for LAS formats 1.1 -1.4. Each version has changes to the class numbers and classification of certain criteria, as we will discover later. In order to  verify the LAS version of these LAS files, we must access the vendor report to verify the LAS version. As a part of the USGS 3DEP project, various contractors are hired to collect and deliver lidar data to the USGS. Vendor reports are required with every contract of the lidar missions. For this tutorial we have accessed the vendor report from the same ftp site where the data was downloaded in order to observe the LAS version and metadata information.

Under the Deliverables section of this vendor report, the uncalibrated, unclassified raw point cloud swath is listed as LAS version 1.2, the vendor report has been accessed in advance for reference.

Now that we know the LAS version is ASPRS version 1.2 format, we will compare the ASPRS LAS data versions 1.2 and 1.4 and note some of their differences.

The American Society for Photogrammetry and Remote Sensing periodically updates the LAS file format including changes to point classes. For information about the current LAS file version, please see the LAS Specification v.1.4 - R15 document found at the ASPRS link shown here.

As you can see, in LAS version 1.2 the classes highlighted in blue were reserved but not yet defined. But, by LAS version 1.4, many of these previously reserved classes were defined.”

For example, in LAS version 1.2 class 17 is ‘reserved’, and by version 1.4 it is defined as Bridge Deck, and so on.

The acquisition vendor used some of these reserved classes and they do not depict the correct features in LAS 1.4. For example, class 18 was used for Overlap ground points instead of High Noise. This is shown in the vendor’s airborne topographic Lidar report. Note the classification scheme of Class 17 as Overlap Default, and class 18 classified as Overlap Bare-earth by the vendors.

However, if you go back to the properties window and select LAS Filter again, the default classification reads LAS files as version 1.4. If the LAS Filter window is not currently pulled up, make sure the LAS file is selected in the Table of Contents, and select the Appearance tab on the ribbon menu. Then, click the ‘LAS Points’ Icon. Make sure to select the icon and not the drop-down arrow. This will open the LAS layer properties window.

Uncheck the “All” box to turn off all classes, then check the “class 17 Bridge Deck” to turn it on. Then, click OK. With the LAS Filter you can see that this lidar collection used class 17 to classify points within the overlap between flight lines. This explains the reason for some of the classification discrepancy.

Now we’ll visualize some of the other point classifications. Open the LAS Layer Properties by clicking on the LAS Filter window under the Appearance tab of the LAS Dataset Layer ribbon, and click on the top icon, not the dropdown arrow.

Again, if the “All” box is checked for the Classification codes, click it again to deselect all classes to turn them off. Check only the ‘2 Ground’ classification code from the list, then click “OK”.

We will now view our ground surface in a 3D Scene. To do this, we must create a separate 3D map window, or local scene. Navigate to the View tab and select the ‘Convert’ icon, in the drop down, select ‘To local scene’.

In the local 3D scene, click the top icon of the Symbology button under the Appearance tab, and make sure the ‘symbolize layer by using points’ icon is selected and that draw using ‘elevation’ is selected.

If the ‘draw using’ box is checked and the layer has not shown up on the 3D Scene, check the Draw Using box to turn it off and then check it once more to draw the layer again. The layer should draw immediately after rechecking the ‘Draw Using’ box. Because the file is so large, it might take a while for the LAS point cloud to draw completely.

We will convert the LAS file to a LAS Dataset to better improve the functionality of the processing and viewing in ArcGIS Pro. A LAS Dataset is designed to use lidar data in .las or .zlas file types. It supports any LAS file version 1.0 -1.4. It is important to note that the LAS format supports the classification codes for each point according to the American Society for Photogrammetry and Remote Sensing and ArcGIS applies the class codes specified for LAS version 1.4. A LAS Dataset has the capability to store and reference many LAS files, compiling them to one LAS Dataset, similar to a mosaic of raster surfaces. The LAS Dataset enables the user to view the LAS file or files in their native binary format of stored airborne lidar data. By using a LAS Dataset, users can more efficiently conduct QA/QC on lidar data, and quickly visualize many LAS files at once, as either a TIN or point cloud, in both 2D and 3D views. The LAS dataset also allows the user to manually fix or reclassify errors in class codes through editing techniques provided by ArcGIS Pro. We will cover more on LAS Datasets in another video. For more information on what a LAS Dataset is or LAS Dataset capabilities and properties, visit these ArcGIS Pro help page.

Select the ‘Tools’ icon in the ‘Analysis’ tab. In the Geoprocessing ‘Search’ window type ‘Create LAS Dataset’ then select the option to “Create LAS Dataset”. In the Input Files section, browse to the folder containing your downloaded lidar data, and select the file ending in ‘13SDD483396_LAS_2015.las’ then click okay. In the ‘Output LAS Dataset’ section, browse to your project folder and name the LAS Dataset ‘Tutorial_10b2.lasd’. We will not be working with clipping extents or breaklines for the LAS Dataset in this tutorial, so leave the Surface Constraints blank. Verify that the Coordinate System corresponds to your original las file. This can be found by right clicking the original .las file and selecting properties, then source, and expanding the ‘Spatial Reference’ drop down arrow. In the Geoprocessing window, if the coordinate system does not read ‘NAD_1983_UTM_Zone_13N / VCS:NAVD88’ then select the appropriate projections and coordinate systems. Alternately, you may select the original .las file from the Coordinate System drop down menu and it will automatically choose the .las file’s datum and projections. Make sure ‘Compute Statistics’ is checked, then click ‘Run’.

Now is a good time to save your project. Go to the upper left corner and select the save icon or hit ‘ctrl’ + ‘s’.

Turn off the LAS layer ending in ‘13SDD483396_LAS_2015.las’. Make sure the LAS Dataset layer, Tutorial _10b2 is highlighted by left clicking it once.

Right click the ‘Tutorial_10b2.las’ layer and select ‘properties’. In the ‘LAS Filter’ tab, Under Classification Codes” Check ‘All’ to deselect all classes and then check the ‘11 Road Surface’ and ‘2 Ground’ boxes to add them to the display. Notice the floating red points that are classified as ‘Road Surface’.

If you are using a mouse, the left mouse button will allow you to pan around the map view window. Click and hold the middle mouse button to rotate the 3D scene to view the point cloud from the side and click and hold the right mouse button while moving the mouse to control the zoom. If you are using a laptop pad without a mouse, access the on-screen-navigator tool, to control the tilt, pan, north orientation, and rotation around a target as shown in the slide. For more information for the ‘on-screen-navigator’ tool , visit this link: https://pro.arcgis.com/en/pro-app/help/mapping/navigation/the-on-screen-navigator.htm.

There are also very helpful keyboard shortcuts that can be found here https://pro.arcgis.com/en/pro-app/help/mapping/navigation/keyboard-shortcuts-for-navigation.htm.

Under the ‘Map’ tab, make sure the ‘Explore’ tool is highlighted. In the map extent window, Click and hold the middle mouse button to rotate the 3D scene to view the point cloud from the side. Click and hold the left mouse button to pan the scene and click and hold the right mouse button while moving the mouse to control the zoom. If you are not using a mouse or haver a laptop touchpad, access the on-screen-navigator tool, and select the ‘full navigation’ drop down arrow on the screen.

Click the LAS Points button again to view the LAS Filter properties. Turn off the ‘2 Ground’ classification and add the ’18 High Noise’ classification to the view. The ‘High Noise’ classification appears to be ground points in areas of swath overlap, whereas the ‘Road Surface’ classified points appear to be actual High Noise points.

Next let’s visualize class 17, check the ’17 Bridge Deck’ points classification and turn off all other classes. The points classified as ‘Bridge Deck’ appear to be buildings, vegetation and roads in areas of swath overlap. From visualizing these classifications independently, we can see that there are some misclassifications within this dataset.

While there may appear to be large mistakes in classifications, remember that these project data were delivered in LAS format version 1.2, whereas the LAS version used by the software is the current version 1.4.

Now that we have identified some misclassifications within the dataset, we can modify them. ArcGIS Pro provides several automatic and manual classification Tools to edit and classify LAS classes. The LAS Dataset is just like a mosaic dataset. It stores the lidar footprints and statistics, but it is referencing the original LAS files. So, when we modify the codes, we will be changing the original data. If you look in Windows Explorer, you’ll see the LASD is only 7 KBs and the Date Modified for your fixed LAS file will be more recent than the other five tiles. In case there are any errors when reclassifying, we will create a copy of the LAS Dataset. In the Geoprocessing pane, search ‘Copy’ and select the ‘Copy (data management)’ tool. For the input select the ‘Tutorial_10b2.lasd’ by navigating to “Project” and then “Folders” on the left column of the ‘Input Data’ window. In this project folder location, you can access the ArcGIS Pro project geodatabase, output data files, and other items specific to your map project.  Name the output file ‘Tutorial_10b2_copy.lasd’ and run the tool.

Because the points have been previously classified, we will change the classification codes using the ‘Reassign Classification’ tool located in the Classification tab, in the top left corner. We will start by reclassifying the misclassified High Noise Points into the corresponding 12 Overlap Points, Filtering the point cloud to only display the ‘18 High Noise’ class as we did previously.”

Select the Classification Tab and click the Reassign Classification button. This will open the ‘Change LAS Class Codes’ geoprocessing pane. Input the current class as ‘18’ for the high noise points that are turned on and set the New Class to ‘12’ for Overlap points. Accept the defaults ‘No change’. Click ‘Run’.

A red box will appear over the extent showing no points in the current LAS point filter because we have reassigned class 18.

Next, change the LAS filter to only show ‘11 Road Surface’. Using the Reassign Classification tool, which should still be open in your Geoprocessing window, repeat the same procedure for input class 11, and new class 18. This will change the high noise points in class 11 Roads to high noise points in the correct High Noise Point class 18. To view the new classifications, go to the LAS filter and select only class 18 High Noise. Now we will repeat the same process again for class 17 Bridge Deck. Change the filter to only show Class 17 Bridge Deck. Click the Reassign Classification and input 17 for current class and 12 for the new class in order to reclassify these overlapping above ground points into class 12 Overlap. Click Run.

ArcGIS Pro also offers some automated classification tools. We will now classify buildings using the Automated Classification tool located to the right of the Reassign Classification tool. Click on the Automated Classification drop down arrow and select ‘Classify Buildings’. Make sure the input LAS Dataset reads ‘Tutorial_10b2_copy.lasd’.  Change the minimum Rooftop Height to 4 feet and Minimum area to 5 square feet. Make sure Compute Statistics is checked and the other boxes are not, then click ‘Run’. Once the tool has finished running, filter only class 6 buildings and zoom in to see the building filter. You may change the basemap to Imagery to better delineate the building footprints.

After viewing the buildings, right click the Tutorial_10b2_copy.lasd layer and select ‘Zoom to Layer’. There is a light green bridge/overpass classified as a building, located in the right central edge of the point cloud. We will manually zoom in and select these points and reclassify them as 17 Bridge Deck.

Go to the Classification tab under the LAS Dataset Layer ribbon and select the ‘Select LAS Points by Rectangle’ tool in the Selection section.

Draw a rectangle around all points in the bridge feature by left-clicking and dragging the mouse to draw the rectangle. The initial selection will not select all the points, so you may need to zoom in and hold the ‘shift’ key and repeat the select by rectangle multiple times to select all points in the bridge. The total points should be around 4,000. Once all the bridge points are selected, click the classification code drop down arrow and choose 17 Bridge Deck. Click the highlighted arrow labeled Apply changes. The points should disappear from our filtered extent. Go to the LAS Points filter and choose only 17 Bridge Deck. The bridge we just manually reclassified should draw.

Congratulations! You’ve finished Lesson 10b2: Exploring and Classifying Lidar Data in ArcGIS Pro.

In this lesson, we discussed how to view LAS class codes, select points of interest using polygon selection and class filtering, use manual and auto-classification methods to edit LAS classes in a LAS Dataset, and how to generate and use a LAS Dataset. If you are interested in learning more about using lidar data, please see the additional USGS lidar training videos in ArcGIS Pro, LP360 and Global Mapper at:  http://www.usgs.gov/NGPvideos.