Lab 5

Goal and Background

          The purpose of Lab 5 was to be introduced basic LiDAR data structure and processing, such as processing and retrieving surface models and digital terrain models (DTM) and processing and creating intensity images and other derivative products from a point cloud. The LiDAR data in this lab is in LAS format. It's important for a student to have experience in working with LiDAR data, as its role and potential is only growing in the remote sensing world. 

Methods

          For the first part of the lab, bring in the .las files in the Lab 5 folder into Erdas Imagine viewer. Change File Type to LAS as Point Cloud so the las. files can be retrieved from the Select Files window. A warning will pop up about the layer not having level of detail (statistics). Uncheck Always Ask and click Ok as this will be computed later. It's essential that LAS data that is being worked on have Metadata and a Tile Index. The Tile Index can indicate which part of a study area the .las file covers. In ArcMap, display the Tile Index shapefile from the Tile Index folder within the Lab 5 folder. Using the tile information in Erdas Imagine, it's possible to locate the point cloud data from the Tile Index of the study area.

          The second part of the lab imagines the student as a GIS manager for the City of Eau Claire, WI, USA. With LiDAR point cloud data of a part of the City of Eau Claire in LAS format, the student is guided to first conduct of quality check of the data by looking at its area and coverage and verify the current classification of the LiDAR.

          In ArcMap, activate the LAS Dataset toolbar and activate the 3-D Analyst and Spatial Analyst extensions. In the LAS folder within the Lab 5 folder, create a new LAS Dataset and name in Eau_Claire_City.lasd. Go to the Eau_Claire_City.lasd properties window and to the LAS File tab. Click the Add File button and select all the LAS files within the Lab 5 folder. The LAS files are now in the LAS Dataset properties window where the individual file information is visible, such as the LAS version, number of points, point spacing, and Z min and max. In the Statistics tab, click Calculate as the statistics for the dataset hasn't been built yet. Now all the individual file data is taken together to form statistical and meaningful data. Statistic data is often used for quality assurance and quality control of the dataset. For example, check the Z min and max values with the known elevation of Eau Claire. They should match closely.

          Go to the XY Coordinate System tab. No coordinate system is specified in the dataset or any of the .las files, which is common for older LiDAR datasets. Review the point cloud metadata and determine the horizontal (XY) and vertical (Z) coordinate system, which is North American 1983 Lambert Conformal Conic and North American Vertical Datum of 1988 in feet respectively. With this information, navigate to Projected Coordinate Systems > County Systems > Wisconsin CRS > US Feet to change the XY information to NAD 1983 HARN Wisconsin CRS Eau Claire (US feet). This is the closest to the information given in the metadata. In the Z Coordinate System tab, navigate to Vertical Coordinate Systems > North American and change the Z coordinate system to NAVD 1988 US Feet (height). Once that's all applied, display the dataset in ArcMap. The red boundaries designate  the tiles.

          Display the Eau Claire County shapefile from the Lab 5 folder with the LAS data. It's clear now that the LAS tiles are over the City of Eau Claire where they should be. Remove the shapefile, it was only for verifying information.

          From the layer properties window, change the classes to 8. Zoom in on the tiles so the points will generate. The further zoomed in, the more percentage of data that will be displayed. The colors of the points are categorized by elevation. From the LAS dataset toolbar, its possible to view the surface points as elevation, aspect, slope, and contour. With Contour activated, in layer properties and the Show tab, manipulate the Index Factor and the Contour Interval and observe the effects on the data. Return to the layer properties window with elevation activated and observe the Classification and Return Filters. 

          Exit the properties window, filter the point data by First Return from the toolbar, and observe the point cloud data. Click on the LAS Dataset Profile view and click and drag a window over a bridge in the data. This will open a window showing the bridge point cloud. Any section of the data can be viewed in this 2-D manner or in 3-D.

          The third part of the lab involves generating LiDAR derivative products. The Point Spacing information of the .las files will help determine the spatial resolution of the derivative products. Take the average of the Point Spacing to estimate the Nominal Pulse Spacing (NPS), which in 1.4103. The spatial resolution will be at 2 meters. The spatial resolution cannot be greater than the NPS. 

          Open the LAS Dataset to Raster tool from the Conversion > To Raster tool sets. Set the output to the Lab 5 folder and name it EC_FR_2m. The value field will read ELEVATION and use Binning interpolation. Run the tool after setting the Cell Type to Maximum, Void Filling to Natural Neighbor, Sampling Type to Cellsize, and Sampling Value to 6.56168 (roughly 2 meters). The DSM image will display. Use the measuring tool to verify the the pixels are 2x2 meters.

          Next, open the Hillshade tool from the 3D Analyst Tools > Raster Surface tool set. Input the DSM image that was just created and run the tool. The resulting image is Figure 1.

          Starting with the LAS dataset on display again, set the point tool to Elevation and filter to Ground. Open the LAS Dataset to Raster tool again and name the output EC_DTM_2m. Set the parameters to Binning interpolation, Cell Assignment Type to Minimum, Void Fill Method to Natural Neighbor, Sampling Type to CellSize, and Sampling Value to 6.56168. Run the tool and then create a Hillside layer of the input. The resulting image in Figure 2. Make sure the DSM hillshade is on top in the table of contents. Activate the Effects toolbar from the Customize tab and select the DSM hillshade layer from the dropdown window. Using the Swipe tool from the toolbar, compare the two images side by side by clicking on the image in the viewer and swiping the arrow. 

          Next, start with just the LAS point cloud data on display. Set the point tool to Elevation and the Filter to First Return. Open the LAS Dataset to Raster tool. Input the LAS dataset. Set the Value Field to Intensity, Binning interpolation, Cell Assignment Type to Average, Void Fill to Natural Neighbor, Sampling Type to CellSize, and Sampling Value to 6.56168. Set Output to the Lab 5 folder and name it EC_Int. Run the tool. The intensity image displays better in Erdas Imagine. Open Erdas Imagine, and in Select File, change the File Type to Tiff so the intensity image is selectable. Bring the Intensity image into the viewer and observe the characteristics of a Intensity image (Figure 3). 

          

Results

          From the First Return points, its possible to create a Digital Surface Model of the LAS point cloud data (Figure 1). Therefore, the surface features on Earth's surface, including buildings, vegetation, and water, are distinguishable. Interpolation on water surfaces can cause shaded anomalies in the image.

Figure 1: Hillshaded DSM image of the City of Eau Claire.

          A Digital Terrain Model can be generated by the Ground Return Points of the LAS point cloud data (Figure 2). The resulting image shows the terrain of the Earth and its relative elevation as it is without buildings and vegetation.

Figure 2: Hillshaded DTM image of the City of Eau Claire.

          First Return points are utilized to generate Intensity images, which closely resemble panchromatic images of optical remotely sensed images. Intensity is the strongest voltage recorded by the LiDAR sensor. the Intensity image can aid in interpretation of features,

Figure 3: Intensity image of the City of Eau Claire.

Sources

Eau Claire County (2013). Lidar Point Cloud. Eau Claire, WI.

Eau Claire County (2013). Tile Index. Eau Claire, WI.

Price, Maribeth (2014). Mastering ArcGIS 6th Edition Dataset. Redlands, CA: McGraw Hill.

Wilson, Cyril (2017) Week 9: Class 2: LidDAR Remote Sensing 2. [Lecture Slides]. Eau Claire, WI.