LIDAR Remote Sensing

Goals and Background

           The goal of this lab is to become familiar with the structure and processing of LIDAR data. In order to gain knowledge about LIDAR lab 5 emphasized how to process and retrieve different surface and terrain models. Additionally, processing and creation of intensity images was done along with deriving outputs from a point cloud. In this lab the LIDAR data that was used was in the LAS file format. Working with LIDAR is an essential tool as it is an extremely quickly growing field.

Methods

         For this lab ArcMap and Erdas will both be used. To begin lab open up Erdas, in the viewer go to open and select the files that were provided in the LAS file in the lab 5 folder. From there be sure to change the files of type to LAS as Point Cloud (*.las). After the file type is changed all of the data can be brought in to Erdas. Be sure to uncheck the always ask button and to click no. This step takes a while for the point cloud to load. When working with an unprojected data set such as this it is important to take a look at the tile index and the metadata. By opening ArcMap and bringing in the QuarterSection_1.shp one can be sure that the point cloud was displayed in the correct area.

       Next, close Erdas and ArcMap and open a blank page in ArcMap. The goal of this next objective is to create a LAS dataset, explore the properties of the LAS dataset and to visualize the dataset as a point cloud in both 2D and 3D. After connecting to the student folder using ArcCatalog a new LAS dataset was created named Eau_Claire_City. The same files from Erdas were then added into ArcMap by clicking on add files. After the data is in, click on calculate which is under the statistics tab, this will calculate the statistics for the dataset. These statistics are used to ensure data quality and help make sure that the LAS Dataset is accurate.

        The next step is to add a coordinate system to the LAS Dataset. No coordinate system was specified, so the metadata was used to find the information regarding the coordinate system. After consulting the metadata it is possible to define the (XY) and (Z) coordinate systems. NAD 1983 HARN Wisconsin CRS Eau Claire (US Feet) is used for the (XY) coordinate system, while NAVD 1988 US feet is used for the Z coordinate system. To make sure that the dataset is in the correct spatial location a shapefile of Eau Claire county is brought into ArcMap. After a little examination it is clear the dataset is in the correct location.

        Next, be sure the LAS dataset tool bar is active, this will be used to visualize the point cloud and it will be used later to help generate other products. Under the properties tab of the Eau_Claire_City shapefile that was created earlier change the number of classes from 9 to 8. When zoomed out at the full extent the points may not be visible, this is done to make the software faster and not bog it down, upon further inspection by zooming in the data will appear.

         Using the LAS dataset tool bar expand the surface menu, from there aspect, slope and contour will be assessed one at a time. Next, the contours will be used to help give a better idea of what will be generated with the DSM. The contour interval can be changed and it is essential to see how it affects the display. Another way to change the contour is to go into the layer properties tab and click the filter tab, on the bottom right of the tab is four predefined settings that use different methods of classification.

        In this next step there will be DSM's and DTM's created using the same pointcloud. The raster products were created at a 2 meter spatial resolution. There was four products created, a DSM, DTM and a hillshade of both the DSM and DTM. The LAS dataset to raster tool was used here which is found under conversion tools. Some defaults were accepted, though the sapling value field was changed to 6.56168 each time which is approximately 2 meters. The hillshade tool is found under 3D analyst tools and raster surface. The defaults there are all acceptable, just be sure the place it will be saved is somewhere easily accessible.

        The final step of this lab is to create a LIDAR intensity image from a point cloud. This was done very similarly to the step above. The LAS dataset to raster tool was used and the value field was changed to intensity, the void fill changed to natural neighbor and the same cell size used in the DSM and DTM is acceptable. From here the image was brought into Erdas, where it was automatically enhanced, take note that the file needs to be brought in as a Tiff.

Results

       

Figure 1 DSM

          Figure 1 above is a screenshot showing the grid that was used throughout the exercise that is picture by the red lines. The image is a Digital surface model, meaning that it is bare earth, there is no vegetation or trees included. In contrast, figure 2 below is a Digital terrain model that shows the terrain of Eau Claire including the trees and buildings. Each has there own uses. The DSM shows the elevation changes much more clearly, while figure 2 gives a better representation of how populated the area is and what the density of vegetation is.

Figure 2 DTM

       Figure 3 below shows the intensity output that was created. In ArcMap the image was very hard to see anything, it is nearly impossible to differentiate anything.

Figure 3 shows the intensity output on ArcMap

        Figure 4 below shows the intensity image that was created in ArcMap, though it is displayed in Erdas in figure 4. When bringing the image into Erdas as a Tiff. Erdas automatically enhances the image. When looking at the image below it is clear that it has very high spatial resolution, this is an image that would be very well suited for aerial image interpretation. It is a very clear image and it is easy to tell what is water, vegetation or man made buildings.

Figure 4 shows the intensity output in Erdas

Sources

Lidar point cloud and Tile Index are from Eau Claire County, 2013.
Eau Claire County Shapefile is form Mastering ARcGIS 6th Edition by Margaret Price, 2014.