Lab 12: UAS 3D Imagery

    

Introduction:

    Lab for this week was in preparation for a future UAS or Unmanned Aerial Systems lab. This lab focused on the image data the UAS take. Dr. hupy gave the class two file folders of images taken by a previous class, one of a track field and another of a baseball field both within the city of Eau Claire, Wisconsin.  The objective this week was to transform these images into 3D composite views in an attractive map form. 

              

                Figure 1  Selecting the aerial photos for PixMapper              Figure 2 Selecting 3D Models in the New Project window

Methods:

      To begin the project Pix4DMapper Pro was opened to a new project the entire file folder of images of the track field was added first while Dr. Hupy explained some of the steps. (see figures 1 & 2)  After making sure the project will save to the desired folder, continuing through the setup windows  and choosing to create a 3D model, a map view with points appears. (see Figure 3)

Figure 3 first view of data in mapview on Pix4D Mapper Pro

 This is a view showing where the images are "tied down" to the earth. In the bottom left of the screen is a Processing window, here is where each picture is matched up through points, vertices and geotags which are associated with each image and looking at each pixel's intensity values to make bands and putting them all together ​ to create a point cloud.

Figure 4 Processing window

 A point cloud is the 3D visualization of an object on an attached coordinate system. Each of the processes were run and automatically saved to the designated folder for this lab. Once the processing was finished a new image appeared on the screen along with a quality report. (see Figures 5, 6, 7 & 8) 

                                                                               Figures 5, 6 & 7 from the quality report:                

Figure 8 Image after processing showing camera points

              

            Figure 9 Measuring 3D area volume

Dr. Hupy also wanted each student to measure the length of a linear feature, calculate the area of a surface,  the volume of a 3D object and create a video showing different views of the project. (see video below) This was completed under the RayCloud tab and new polyline, new surface and new volume.(see Figure 9)

  

Figure 10 Measuring a New Polyline

Once selected, manual pinpoints were placed on distinctive areas of the image to calculate each. (see figure 10) When all the measurements were complete the project was opened in ArcScene by adding the tif in the orthographic folder. To make the image 3D the base height was set to custom float to the DSM tif in its folder. (see Figure 11)

Figure 11 Adding float to base level in ArcMap

 From there the 3D image was saved as a layer file and brought into ArcMap where it was transformed into a map. These steps were completed the same for the baseball field image folder as well.

Results:

Above is the 3D image created from the photographs taken by the unmanned aerial system. Each photo becomes like a pixel as the matched areas accurately tie the imagery to the globe and compute the height of the objects pictured.

Conclusions:
         Through various programs a fairly accurate 3D representation can be created from a UAS image files in a fairly simple manner.