The goal of this lab was to acquire basic knowledge on LiDAR data structure and processing. This lab worked with the retrieval and processing of various terrain and surface models. Also this lab worked with the processing and creation of intensity image and other derivative products from point cloud.
This lab was divided into 3 parts. Part 1 worked with visualizing point cloud data in Erdas Imagine. Part 2 worked with generating a LAS dataset and exploring the properties of LiDAR point clouds with ArcGIS. In Part 2, I worked under the scenario that I was a GIS manager completing a project for the City of Eau Claire, WI, USA. I will be using LiDAR point cloud data in the LAS format for a section of the City of Eau Claire. I will need to check the quality of the data with an initial quality check by considering its areas and coverage and validating the current classification of the LiDAR data. Part 3 worked with generating LiDAR derivative products.
Methods:
Part 1:
I opened an Erdas Imagine viewer and opened each individual LAS file. To access each LiDAr point cloud tile I had to convert the file type to LAS as Point Cloud (*.las). I brought the point cloud tiles into the Erdas Imagine viewer and examined the features through zooming in and out.
Part 2:
I started in an ArcMap viewer and activated the LAS Dataset toolbar. I connected ArcCatalog with the LAS folder containing the LiDAR point cloud tiles used above in Part 1. By right clicking on the LAS folder I created a new LAS Dataset. I opened the properties of this new dataset and added the LAS files from in the LAS folder in the LAS Files tab of the window. I assigned a XY coordinate system to the data under the XY Coordinate System tab in the window. Using Metadata given with the lab information, I was able to determine I needed to assign the NAD 1983 HARN Wisconsin CRS Eau Claire (US Feet) as the horizontal coordinate system and the NAVD 1988 US feet coordinate system as the vertical coordinate system. I then displayed the LAS dataset in ArcMap. To confirm the data was spatially located correctly I added a hollow symbolled shapefile of Eau Claire County to the data frame. The data was spatially located correctly over the Eau Claire County shapefile. I removed the shapefile from ArcMap and zoomed in to the LAS tiles and observed the point cloud being shown by color coded elevation. I activated the LAS Dataset Tool by going under customize in the main toolbar and then selecting extensions and checking the boxes next to 3D Analyst and Spatial Analyst. I experimented with using the aspect, slope and contour tools on the LAS Dataset Toolbar. I also tried using different filters available in the Layer Properties under the Filter tab. I lastly worked with the Profile View Tools available on the LAS Dataset Toolbar. I started by setting the LAS dataset to full extent in the ArcMap viewer. I set the Points to Elevation and the filter to First return. I zoomed in on two tiles in the dataset and selected the LAS Dataset Profile View tool from the LAS Toolbar. I clicked and dragged my mouse over the bridge crossing the river in the image and moved my douse to the right to expand the red box and clicked. A new profile window opened up and displayed the point cloud bridge at a profile view. I also opened the 3D interactive view by selecting the LAS Dataset 3D View tool in the toolbar next to the Profile View Tool. Another window opened up displaying the 3D view.
Part 3:
Part 2: Generating a LAS dataset and exploring LiDAR point clouds with ArcGIS
Part 3:
Section 1: Deriving DSM and DTM products from point cloud
I started by setting my workspace to a specified location for all of my outputs for this section. I did this by going to the Geoprocessing toolbar in the main toolbar and selecting Environments and setting my output location in the Workspace section. To create a digital surface model by using first returns I started by zooming to the full extent of the LAS dataset in an ArcMap viewer and displaying the image in points color coded by elevation and setting the Filter to First Returns. I opened Arc Toolbox and opened the Conversion Tools and then opened the To Raster tools and selected the LAS Dataset to Raster tool. In the tool popup window I named my output and inputted the LAS Eau Claire Dataset as the input. I set the Value Field to Elevation, Cell Type to Maximum, and Void Filling to Natural_Neighbor. I made sure Sampling Type was set to Cellsize and entered 6.56168 feet as the Sampling Value. I selected OK to run the tool. Once the processing was complete the data surface model opened in the ArcMap viewer. Using the measuring tool I measured the cell size to check that the pixels were 2 X 2 meters. I turned off the LAS Dataset in the table of contents in ArcMap so that only the newly made raster was displayed. Next, I made a hillshade of the digital surface model I just created. To do this I first made sure my 3D Analyst extension was activated. I then went into Arc Toolbox and went into the 3D Analyst Tools and selected Raster Surface and choose the Hillshade option. I selected the raster I just created as the input and clicked OK to run the tool. The new image when the tool completed its run was brought to the viewer. The next image I created was a digital terrain model from the LiDAR point cloud dataset. In the ArcMap viewer I turned on the Eau Claire LAS Dataset in the Table of Contents and turned off both the digital surface model and the hill shade product. In the LAS Dataset toolbar I set the filter to Ground and set the Point Tool to be colored by elevation. By using the LAS Dataset to Raster tool I was able to generate a digital terrain model. I used the tool by setting the name and location for the output to save to. I set the Interpolation to Binning, the Cell Assignment Type to Minimum, the Void Fill Method to Natural_Neighbor, the Sampling Type to CellSize, and the Sampling Value to 6.56168. After running the tool the output raster was brought into the ArcMap data frame. I lastly, created a hillshade of this raster following the same steps I used for the digital surface model hillshade, but changed the input to the digital terrain model raster. When the tool completed its run the hillshade was brought into the viewer. I brought both of the newly created hillshades into the viewer; with the first return hillshade return as the top layer and the bare Earth (ground) as the second layer. I customized the main toolbar and added the Effects Toolbar. In the Effects Toolbar I was able to select the data surface model in the pulldown menu and selected the Swipe tool. This allows the cursor to reveal the bottom layer as the mouse is swiped back and forth across the screen. This tool helps with comparing two similar images to notice slight changes.
Section 2: Deriving LiDAR Intensity image from point cloud
I started by setting my workspace to a specified location for all of my outputs for this section. I did this by going to the Geoprocessing toolbar in the main toolbar and selecting Environments and setting my output location in the Workspace section. To create a digital surface model by using first returns I started by zooming to the full extent of the LAS dataset in an ArcMap viewer and displaying the image in points color coded by elevation and setting the Filter to First Returns. I opened Arc Toolbox and opened the Conversion Tools and then opened the To Raster tools and selected the LAS Dataset to Raster tool. In the tool popup window I named my output and inputted the LAS Eau Claire Dataset as the input. I set the Value Field to Elevation, Cell Type to Maximum, and Void Filling to Natural_Neighbor. I made sure Sampling Type was set to Cellsize and entered 6.56168 feet as the Sampling Value. I selected OK to run the tool. Once the processing was complete the data surface model opened in the ArcMap viewer. Using the measuring tool I measured the cell size to check that the pixels were 2 X 2 meters. I turned off the LAS Dataset in the table of contents in ArcMap so that only the newly made raster was displayed. Next, I made a hillshade of the digital surface model I just created. To do this I first made sure my 3D Analyst extension was activated. I then went into Arc Toolbox and went into the 3D Analyst Tools and selected Raster Surface and choose the Hillshade option. I selected the raster I just created as the input and clicked OK to run the tool. The new image when the tool completed its run was brought to the viewer. The next image I created was a digital terrain model from the LiDAR point cloud dataset. In the ArcMap viewer I turned on the Eau Claire LAS Dataset in the Table of Contents and turned off both the digital surface model and the hill shade product. In the LAS Dataset toolbar I set the filter to Ground and set the Point Tool to be colored by elevation. By using the LAS Dataset to Raster tool I was able to generate a digital terrain model. I used the tool by setting the name and location for the output to save to. I set the Interpolation to Binning, the Cell Assignment Type to Minimum, the Void Fill Method to Natural_Neighbor, the Sampling Type to CellSize, and the Sampling Value to 6.56168. After running the tool the output raster was brought into the ArcMap data frame. I lastly, created a hillshade of this raster following the same steps I used for the digital surface model hillshade, but changed the input to the digital terrain model raster. When the tool completed its run the hillshade was brought into the viewer. I brought both of the newly created hillshades into the viewer; with the first return hillshade return as the top layer and the bare Earth (ground) as the second layer. I customized the main toolbar and added the Effects Toolbar. In the Effects Toolbar I was able to select the data surface model in the pulldown menu and selected the Swipe tool. This allows the cursor to reveal the bottom layer as the mouse is swiped back and forth across the screen. This tool helps with comparing two similar images to notice slight changes.
Section 2: Deriving LiDAR Intensity image from point cloud
In section 2 I started with the LAS dataset in an ArcMap viewer. I set the dataset to points and filtered of first returns in the LAS Dataset Toolbar. In Arc Toolbox I accessed the LAS Dataset to Raster tool. In Input I entered the LAS dataset of the City of Eau Claire. I set the Value Field to INTENSITY and set the Binning Cell Assignment Type to Average and Void Fill to Natural_Neighbor. I set the cell size to 6.56168 and set a location to save my output image to. I ran the tool and once it was completed it was automatically added to the viewer. The image was displayed too dark in ArcMap so I opened the image in Erdas Imagine, which enhanced the display. To be able to display the image in Erdas Imaging I had to change the File Type to TIFF first.
Results:
Part 1: Pont Cloud Visualization in Erdas Imagine
Part 1: Pont Cloud Visualization in Erdas Imagine
Figure 1
Figure 1 is the Erdas Imagine viewer displaying all of the LiDAR point cloud tiles.
Figure 2
Figure 2 is a zoomed in image of Figure 1. In the Erdas Imagine viewer with Figure 1 displayed it was zoomed in a great amount to be able to see the image at point level. In Figure 2 you can see some of the individual points that make up a section of Figure 1.
Part 2: Generating a LAS dataset and exploring LiDAR point clouds with ArcGIS
Figure 3
Figure 3 is what the LAS dataset first looks like when displayed in ArcMap. The extent of the LAS dataset tiles are drawn out with red boundaries.
Figure 4
Figure 4 shows what the LAS dataset looks like after zooming in while displayed in ArcMap. In the Table of Contents on the left side of the image shows the symbolized color elevations of the points in the image.
Figure 5
Figure 5 shows the LAS dataset of Eau Claire displayed in the aspect tool from the LAS Dataset toolbar. The aspect tool calculates the direction of the slope for TIN faces. In the Table of Contents on the left side of the image shows the symbolized colors for the different directions of slope of the TIN faces in the image.
Figure 6
Figure 6 shows the LAS dataset of Eau Claire displayed in the contour tool from the LAS Dataset toolbar. The contour tool shows the LAS Dataset surface with contours. In the Table of Contents on the left side of the image shows the symbolized by lines for contour and index contour lines.
Figure 7
Figure 7 shows the LAS dataset of Eau Claire displayed in the elevation tool from the LAS Dataset toolbar. The elevation tool symbolizes the LAS Dataset TIN faces by elevation. In the Table of Contents on the left side of the image shows the symbolized colors for the different elevations.
Figure 8
Figure 8 shows the LAS dataset of Eau Claire displayed in the slope tool from the LAS Dataset toolbar. The slope tool calculates the slope for TIN faces. In the Table of Contents on the left side of the image shows the symbolized colors for the different slopes of the TIN faces in the image.
Figure 9
Figure 9 shows the profile view and 3D view of a bridge in Eau Claire found in the original image or the farthest back data frame in the image, in the red rectangle. The profile view was created by using the LAS Dataset Profile View tool and the 3D view was created by using the LAS Dataset 3D View tool. Both tools are located in the LAS Dataset Toolbar.
Part 3: Generation of LiDAR derivative products
Section 1: Deriving DSM and DTM products from point cloud
Figure 10
Figure 10 shows the digital surface model with first returns. This image was made from turning the LAS Dataset into a raster by using the LAS Dataset to Raster tool located in Arc Toolbox.
Figure 11
Figure 11 shows the hillshade of the digital surface model using first returns. This output image was created by using the 3D Analyst tool of Raster Surface to Hillshade found in Arc Toolbox.
Figure 12
Figure 12 shows the digital terrain model using ground returns. This output image was created by turning the LAS Dataset into a raster by using the LAS Dataset to Raster tool located in Arc Toolbox.
Figure 13
Figure 13 shows the hillshade of the digital terrain model using ground returns. This output image was created by using the 3D Analyst tool of Raster Surface to Hillshade found in Arc Toolbox.
Section 2: Deriving LiDAR Intensity image from point cloud
Figure 14
Figure 14 shows what the LAS dataset looks like after being converted to a raster intensity image. This image was displayed in Erdas Imagine. This image is very sharp and this was accomplished by using a narrower spectral resolution.
Sources:
Lidar point cloud and Tile Index are from Eau Claire County, 2013.
Eau Claire County Shapefile is from Mastering ArcGIS 6th Edition data by Margaret Price,
2014.