Overview of Pix4D software
The objective in this lab was to get acquainted with using Pix4D software package to process imagery. An Unmanned Aerial System (UAS) was used to collect the imagery provided for processing in Pix4D.
What is the overlap needed for Pix4D to process imagery? What if the user is flying over sand/snow, or uniform fields? For a general case, at least 75% frontal overlap (with respect to flight direction) and 60% side overlap (between flying tracks) is needed. If flying over sand/snow, or uniform fields then at least 85% frontal overlap and 70% side overlap are necessary.
What is Rapid Check? Rapid check is a function that speeds up processing by lowering the resolution of the original images. It is recommended for use in the field because it is much quicker than full processing.
Can Pix4D process multiple flights? What does the pilot need to maintain if so? Yes, the pilot needs to maintain 3 things: 1. each plan captures the images with enough overlap, 2. there is enough overlap between 2 image acquisition plans, and 3. the different plans are taken as much as possible under the same conditions (sun, weather, new construction, etc.)
Can Pix4D process oblique images? What type of data do you need if so? Yes, it can process oblique images as long as the user knows what angle off nadir the images were collected.
Are GCPs necessary for Pix4D? When are they highly recommended? GCP's are not necessary, but they provide information on scale, orientation, and absolute position, so they are highly recommended when working with images which don't have any geolocation.
What is the quality report? The quality report basically provides all of the metadata and allows the user to see if there are any errors to take note of before moving forward.
(Answers provided by Pix4D Software Manual)
Methods
Walkthrough of Pix4D software
Step 1: Open the Pix4D Mapper software and create a new project. When naming the project, it is important to include the date, site, platform or sensor, and altitude. The name for this project was 20160621_litchts_phantom3_60m ("ts" was added to identify the creator's initials).
Step 2: Select images to add. All of the imagery needed for this project was located inside a folder, so the "add directories" function was used to add all images at once. Pix4D will then read the metadata associated with the imagery and note if they are geotagged as well as indicating the coordinate system used. Camera properties are also included but are often unreliable and will need to be edited. Change the camera model to Linear Rolling Shutter. Parameters should then look like Fig. 1.
Figure 1. Camera parameters as viewed in Pix4D. |
Step 3: Leave the coordinate system in default, it should be WGS84/UTM zone 15N. Next, choose the 3D maps processing template. Now the set-up is complete and processing can begin.
Figure 2. Prior to beginning steps 2 and 3 (Note: step 1 is UNCHECKED!) |
Summary of quality report: 100% of the images were used (68 total), none were rejected. Georeferencing was used however there were no 3D GCP's. Most of the area had very good overlap (Fig. 3), the only areas with low overlap were the left and bottom edges. This may be due to lack of full coverage from some images but this is a very minor detail.
Figure 3. Shows overlap of imagery used in Pix4D. Green areas have high overlap which is good, yellow/red areas have poor overlap. |
Results
Figure 4. Map of Litchfield Mine created in ArcMap. |
Conclusion
Final Overview of Pix4D software
Pix4D software is very user friendly for those who have had an intermediate amount of GIS training. It makes turning imagery into a real map with metadata much easier than old school processes. Through this lab, a new understanding of aerial imagery was gained after seeing all of the metadata that goes into creating maps. Using Pix4D in this activity was solid preparation for future field surveying with drones and transforming imagery.
References
https://support.pix4d.com/hc/en-us/articles/204272989-Offline-Getting-Started-and-Manual-pdf-#gsc.tab=0
http://hagenfieldmethods.blogspot.com/2016/05/processing-uas-data-in-pix4d.html
No comments:
Post a Comment