Pix4D (Continued): Using GCPs

INTRODUCTION

This report is a continuation of the previous post: Pix4D: Processing UAS Data. In this report, however, data is processed using ground control points (GCPs) to greatly enhance accuracy. For collection methods, visit the previous UAS Platforms and GPS Units for Ground Control post. The purpose of GCPs will first be discussed followed by how they are used in Pix4D and the comparison of results between processing with and without. Finally conclusions are drawn on how GCPs relate to data quality based on processing results. 

GCPs as defined by Pix4D, the current most popular UAS processing software, is a characteristic point whose coordinates are known and used to georeference a project and reduce the "noise." Here, the noise refers to the variance in elevation return producing less accurate results. GCPs purpose is to enhance the positioning and accuracy of processing aerial photography producing more accurate 3D models.  

METHODS

As with the previous post, attention to detail is made first by marking the date, UAS platform, location, and flight elevation as well as noting GCPs will be used. This differentiates the project from other datasets and helps with navigation. Following the last report, the images from the DJI Phantom 4 were uploaded and the image properties switched to linear rolling shutter. 

The next step to incorporate the GCPs collected in the field was accessing the GCP Manager and importing them (Figure 1). 
Figure 1: Importing GCPs into Pix4D GCP Manager
These GCPs should appear on the flight path as blue crosses as shown by Figure 2. If one is missing, it's likely an error in the Y or X field which indicates latitude and longitude, respectively. 
Figure 2: Flight path with imported GCPs in Pix4D
After importing, the basic editor is used for initial processing. Using this editor, the exact location of each point is marked using the cursor to select the point directly on the GCP shown on the UAS imagery. After each GCP is located in one of the UAS images, the data is then ready for initial processing. After this is completed, re-optimization is selected and the Ray cloud in the GCP Manager is used to fine-tune the GCPs. 
For about seven GCPs, the cursor was used to select a more accurate point on the UAS imagery. The rest of the GCPs are then automatically adjusted accordingly to this update. The next two phases of processing--point cloud and mesh and DSM, orthomosaic, and index--are able to run. Upon completion, a quality report is generated.

RESULTS/DISCUSSION

The quality report shows a number of differences in the output by using GCPs. Instead of producing elevation based on an ellipsoidal earth method recorded from the platform sensor, a coordinate system is used (Figure 4). 
Figure 4: The coordinate system used to represent the UAS data shown in Pix4D
The quality report also details the geolocated GCPs. The root mean square error (RMS) shows the accuracy of geolocation. For survey-grade quality, 0.1 is considered acceptable. Using the basic editor, re-optimizing, and Ray cloud editor, the RMS error is well below the accepted rate and can be used with guaranteed accuracy. 

Figure 5: Geolocation details on GCPs to produce survey-grade quality models
After transferring the processed geotiffs to ArcMap, the orthorectified imagery could then be compared with the previous output without GCPs (Figure 6). The total elevation change represented on the bottom map is obvious. A positive elevation gain from 160-140 meters was fairly evenly distributed throughout. This is evident because the change in elevation map retains the outline of most all features. Therefore, the features changed relatively the same together. However, small differences did exist as shown by the top maps. The area to the northeast had less of a positive change in elevation while the southern extent had the opposite effect. 
Figure 6: Change in elevation from processing with GCPs in Pix4D
The GCPs not only affect elevation data, but positional data as well represented by Figure 7. 
Figure 7: The differences in position due to processing with GCPs


CONCLUSIONS

Comments

Post a Comment

Popular posts from this blog

Pix4D: Processing UAS Data

Image
INTRODUCTION Unmanned aerial surveillance photography is an efficient, affordable method in gathering data for small to medium-sized projects for a wide variety of uses. These platforms have a high-degree of customization with various attachments to achieve their goals. Besides controlling the flight and collecting data, the data processing can be an intensive process that involves careful attention to produce accurate results. Pix4D mapper is the most common software in UAS point cloud processing and is demonstrated in this report. Data was collected in the previous report UAS Platforms and GPS Units for Ground Control at the Litchfield mine located southwest of Eau Claire, WI.  METHODS Before processing UAS data in Pix4D, a basic understanding of the software and goal-specific requirements should be obtained. To produce a 3D model, the UAS collects overlapping images and calculates elevation based on these images. The distance of overlap should vary according to th...
Read more

Field Navigation Using GPS, Map & Compass

Image
INTRODUCTION For this exercise, navigation of the University of Wisconsin-Eau Claire's Priory and Children's Nature Center was achieved using a GPS device and coordinates and then using a UTM map and compass. The purpose was to simulate accurate navigation in challenging conditions and terrain during field work. The priory is a wooded, hilly lot owned by the UW-Eau Claire on the Southwest side of Eau Claire, Wisconsin (see Figure 1). The fieldwork was performed on November 4th, 2007 on a rainy morning. Figure 1: The area of interest for the Field Navigation Exercise METHODS GPS and Coordinate Navigation The first round of navigation used a Bad Elf GPS unit and specified coordinates to find marked trees located in the Priory. The class was split into five groups, each with separate UTM coordinates. The coordinates listed below are for Group 1: 1. 617713, 4958075 2. 617767, 4958224 3. 617640, 4958159 4. 617553, 4958074 5. 617579, 4957938 Each group track...
Read more

Survey Techniques (Continued)

Image
INTRODUCTION This post is a continuation of the "Improvised Survey Techniques to Create a Digital Elevation Surface" report. The previous report included a method of sampling elevation points within a sandbox and recording the data to produce a digital elevation surface. For this report, the improvised survey data collected previously is normalized. As defined by ESRI, data normalization is the process of organizing, analyzing, and cleaning data to increase efficiency for data use and sharing. This included restructuring notes from the field journal into standardized columns of x,y, and z values for the sandbox grid to transfer into ArcMap. The elevation data was then modeled using various forms of interpolation. The systematic approach to surveying data points allowed a list of elevations to be made on a grid and left out all elevations between. Therefore, choosing the right interpolation method is important because it performs an estimate of elevations between each po...
Read more