Use of Ground Control Points And Cartographic Fundamentals

 Use of Ground Control Points And Cartographic Fundamentals  

    We are going to explore the use of Ground Control Points (GCP) in data collection and how it corresponds to post-processing orthomosaic generation. Next, we will take the orthomosaic data and use it to create a professional cartographic map.

    One of the most desired skills in the UAS world is having the capability to take UAS data and put it into the context of a Geographic Information System software package for further analysis and use. Beyond that, you should know there are fundamental differences between a proper map and a picture taken from the air. I have become familiar with Geospatial core concepts and I created a map that is functional and meets cartographic fundamentals. I am presenting two maps that I created in AcrPro. One I created using data I was given, and one where I went out and collected the data.

Field Notes:

Before we went out into the field, we were assigned our flight crews. I was a member of crew #7 along with Cameron Dine. Our mission was to fly a grid pattern over a select area at Purdue Wildlife Area (Figure 58) at 100ft AGL with 80% front and lateral overlap. We placed 5 ground control points in the flight area and the location of those can be seen below (Figure 59). We placed the ground control points because they allow us to get more accurate location data which will help us get better results when we process the data in post-processing.

Figure 58: Flight Area At Purdue Wildlife Area

Figure 59: Shows Location Of Ground Control Points


Conditions:

  • Overcast
  • Wind 7mph N-NE
  • Potential Rain In Area

Hazards:

  • Multiple Operations Going On At The Location
  • Treeline To The Right
  • Few Birds Present

Flight 1:

  • PIC: Cameron Dine
  • VO: Hunter Donaldson
  • Flight 1 was conducted at 100ft AGL with 80% overlap
  • Takeoff: 2:45pm
  • Landing: 2:54pm

Ground Control Points (GCPs):

Ground Control Points, or GCPs, are marked points on the ground that have a known geographic location. A drone can be used to autonomously collect photos of the survey area, like the data collection we did at Purdue Wildlife Area. If used, GCPs must be visible in these aerial photos and then processed in the cloud using drone mapping software. For us, we are going to use Pix4D mapper. For aerial survey applications GCPs, are typically required as they can enhance the positioning and accuracy of the mapping outputs. 

Key Points For Using GCPs:

  • The bare minimum number of GCPs you want in a project is 3
  • The recommended number of GCPs you want in a project is 5
  • It is recommended that you use 5 GCPs that are visible in at least 5 images
  • The GCPs should be evenly spaced on the landscape to minimize error
  • It's not a good idea to just place them along the edge of the operation area because they will only be visible in a portion of the images


First, you want to open your processing software (I will be using Pix4D) and upload your data. Once your data loads in, you will want to set the image properties.

Figure 60: Image Properties Page


Once the data has loaded in, you should see your data and flight path on the map. You must uncheck point cloud and mesh, and DSM, orthomasaic, and index because that will take a lot longer to process, you only need the initial processing completed before you set the points with rayCloud editor.

Figure 61: Data Loaded In And Beginning Initial Processing


*Note* While processing my map, I kept getting errors that a large portion of my images was not calibrated and would not load in (Figure 62). I mentioned it to a TA, and he tried to help, but we still could not get it to work. I ran the process 5 times, and it did the same thing every time. I believe that is why my map turned out the way it did. 


Figure 62: Shows Uncalibrated Images

Nevertheless, now that the data has finished initial processing, it's time to add in our GCPs. 

  1. Make sure that the GCP file has the correct input file format

  2. On the GCP/MTP Manager window, on the GCP/MTP Table, click

    Import GCPs

  3. Select the order of the coordinates

  4. Click browse, to navigate to the GCP file

  5. Click the GCP file

  6. Click open

  7. Insert accuracy


Figure 63: GCP/MTP Manager Window


* If your GCPs show up in the wrong location, you need to check your coordinate system * 

Once your GCPs are added to your project, you need to tie them together using RayCloud Editor. The raycloud provides seamless integration between a 3D representation of the project area and the original images. The 3D reconstruction is created by the intersection of multiple rays, creating a cloud of rays. Once you open the raycloud editor, you want to go in and select each GCP individually. Once you select one, all the images that captured the GCP will show up. In order to make it more precise, you need to manually correct the location of each GCP. To do this, you zoom in on the images and click the center of the GCP. Once you do this for 3 of the images, it will start to auto-correct itself. 

When using the RayCloud Editor:

  • Green cross: Reprojecting of the 3D point on the images 
  • Orange cross: Position where the associated 2D keypoints have been automatically detected
  • Orange circle: The radius of the orange circle indicates the size of the area that has been considered to detect the keypoint


Figure 64: Shows Manual GCP Correction


Once you're finished correcting the GCPs, it's important that you reoptimize the data. You need to reoptimize it because it reoptimizes the camera positions and the internal camera parameters. It does not compute more matches between the images; therefore, it is a fast step that improves the accuracy of the project. It should be done after changes have been applied to the GCPs after initial processing.


Figure 65: The Map With GCPs After It Was Reoptimized


After it's been reoptimized, you need to uncheck initial processing (because we already did it) and then reselect point cloud and mesh, and DSM, orthomasaic, and index, Then click start. 

Figure 66: The Final Product After initial processing, GCPs added, reoptimized, point cloud/mesh, and DSM, orthomosaic, and index have finished



 
Now that Pix4D has finished creating the data, we can get the orthomosaic file that we will use to create our map (Figure 67). You can see that my orthomosaic is missing some data, that can be contributed to the uncalibrated images that I had.

Figure 67: Orthomosaic Imagery Of PWA


Cartographic Fundamental And Map Creation: 

Key Features:
  • Title
  • North Arow
  • Scale Bar
  • Locator Map
  • Watermark (For you to state that you made the map)
  • Data Sources/ Metadata
  • Legend

To start off making our map, we are going to open ArcPro and add the orthomosaic data that we created in Pix4d. This was added on top of a world topographic basemap because it shows geographic positions and elevations. It also shows the different land shapes by contour lines. Next, we will add our title, watermark, and metadata because adding text is one of the easier things to add. To add text in ArcPro, right-click on a layer in the contents pane and click the label feature. Then use the text tools in the graphics ribbon to add your text. You can format your text by right-clicking and going to properties. 

Now that our text is inserted, we are going to add our North Arrow and Scale Bar. On the Insert tab, in the Map Surrounds group, click Scale Bar. You can then drag the scale bar to where you'd like and adjust the size. Just like the text, you can right-click and go to properties to format your scale bar.

Since we have our text and scale bar, it's time to add in our locator map. To do this, you need to add a new map. From now on, we will call this map "Locator Map". On the locator map, I added a pin on PWA to show where it is located. I then went into properties to adjust the size and format of the pin. In order to add the locator map to our map, we need to add a map frame. Go to insert, map frame, and select the locator map we just created. Then simply click and drag to add the locator map to the main map. 

The last step isn't required but is good practice, and that is to add a reference scale. Right-click on the name of the map in the contents pane and select the reference scale you want to add. 

That's it! Your cartographic map is complete.

Figure 68: Purdue Wildlife Area Map


I also created another map using the data that was provided. This map has the title in the top left “Wolf Paving Excavation Operation” followed by my watermark “Map Created By Hunter Donaldson”. In the top right, we have the locator map with a red triangle marking the operation location, and it also has a scale bar for distance reference. In the bottom right, there is a legend that marks the bands and the GCP identifier. On the bottom of the map there is another scale bar for the operation site, and in the bottom left there is a cardinal direction symbol. For this map, I also added  DSM in the bottom left. DSM is a digital surface model (DSM) that better shows the elevation of the terrain, as well as above-ground features like buildings and other infrastructure. You add this the same way you did the locator map. By creating a new map, adding the DSM data, and then adding the map frame to the original map.

Figure 69: Wolf Paving Excavation Map



















 

Comments

Post a Comment