What is photogrammetry?
Photogrammetry is simply accurately measuring objects and surfaces from pictures and other digital images.
What types of distortion does remotely sensed imagery have in its raw form?
Remotely sensed imagery contains geometric distortion such as perspective distortion, field of view distortion, lens distortion, earth curvature, radial displacement, and scanning distortion. These are affected by the sensor or lens being used, the focal length of said lens, and the method in which the sensor captures images. Other distortions can be caused by the angles at which they are photographed. For example, from above, a tall building might appear to be at an angle, whereas should really be just the outline of the building as seen from straight above the building.
What is orthorectification? What does it accomplish?
Orthorectification is the way remotely sensed images are corrected for distortion. This allows for the creation of a an accurate orthoimage that can be used to produce a map.
What is the Ortho Mapping Suite in ArcPro? How does it relate to UAS imagery?
The Ortho Mapping Suite in ArcPro is a set of tools to allow the user to create orthorectified images and products from UAS or satellite imagery. UAS imagery is often taken looking straight down and used for creating orthomosaics or digital surface models. The Ortho Mapping Suite is tailored for specifically these types of applications.
What is Bundle Block Adjustment?
Bundle block adjustment uses ground control and tie point information to adjust the exterior of each image so adjacent images alight properly. Once this is done, all the images are then adjusted to fit the ground. This produces the "best statistical fit" between all of the images.
What is the advantage of using this method? Is it perfect?
The advantage of using this method is the simplicity of combining all of the images into one, where the best statistical fit is used. This method is not perfect, and the process provides the user with a table of residual errors once the process is complete. The user can delete the points with high residual error, or manually move the point in error. The program will redo the adjustment until the overall error and residual error are within an acceptable range.
Part 2 (Methods):
What key characteristics should go into folder and file naming conventions?
Some key characteristics that should go into folder and file naming include date the mission was flown, sensor on board the aircraft, altitude flown, ideally the mission location, and the type of file that it is. For example, if I flew a mapping mission over my parents property on May 20th, 2018 using a DJI Inspire with the Zenmuse X5, I might name the file 20180520_zenmusex5_50m_yoderproperty_ortho.tiff or something similar with a varying location name.
Why is file management so key in working with UAS data?
File management is so key in working with UAS data because of the many file types and extensions used, as well as when it comes to sharing data with other persons or entities. Proper file management would make it easier for other parties to know what they are looking at.
What key forms of metadata should be associated with every UAS mission? Create a table that provides the key metadata for the data you are working with.
Metadata such as date flown, UAS platform, sensor, altitude captured, ground control GPS, coordinate system, and weather conditions should be included with every UAS mission.
Date Flown: Nov 8th, 2018
UAS Platform: Yuneec H520
Sensor: Yuneec E90
Altitude Flown: 70m
Ground Control GPS: Propeller
Ground Control Coordinates: NAD83(2011) UTM Zone 16
UAS Coordinates: WGS 84 DD
Pilot: Joseph Hupy
Part 3 (Results):
Describe your maps in detail. Discuss their quality, and where you see issues in the maps. Are there areas on the map where the data quality is poor or missing?
Overall I would say that the map is not bad. It certainly is not the best and most accurate map that could be produced, but it is not bad. Some quality issues that I notice are in the wooded section of the map. There is quite a bit of discontinuance between areas of trees. Another quality issue is around the edges where ArcGIS Pro rectified the images. The edges have abrupt edges to them, and are not entirely straight. Although we are not using this part of the map, it is still a part of the map. The subject area over that center house, though, is accurate and turned out quite well. In ArcGIS Pro, there is the capability to zoom all the way in and see individual leaves on the ground.
 |
| Figure 14: ArcPro Map |
The initial compilation only took 2 minutes and 8 seconds; however, the block adjustment took 1 hour 4 minutes and 44 seconds to complete.
Initial tie points 2:08
Block adjustment 1:04:44
Part 4 (Conclusions):
Summarize the Orthomosaic Tool.
The orthomosaic tool allows you to complete the process of creating the photogrammetrically correct, orthorectified image. The orthomosaic tool will do color balancing and seamline generation based on the settings that you as the user choose from the menu. For the final orthomosaic output, the user is able to choose file format as well as pixel size and other compression options. In the end, ArcGIS Pro will create a photogrammetrically correct orthorectified image.
Summarize the process in terms of time invested and quality of output.
In summary, for a few hours of time between flying the mission and navigating the process through the ArcGIS Pro software, a user can create and use an orthomosaic to create maps and do other data processing. Depending on the resolution at which the images were captured, the overall quality of output could be extremely high, or mediocre, but altogether, ArcGIS Pro will produce comparable results to the data it was originally given.
Think of what was discussed with this orthomosaic in terms of accuracy. How might a higher resolution DSM (From LiDAR) make this more accurate? Why might this approach not work in a dynamic environment such as a mine?
A higher resolution digital surface model (DSM) could make this more accurate because of the elevation rectification process. One of the distortions mentioned is that of relief displacement, which is basically distortion caused by variable elevation above or below the datum. This causes a slight shift in the images position, which would be corrected based on the DSM. A higher quality DSM would more accurately rectify the images. This approach might not work in a dynamic environment such as this due to some of the terrain features, like the large quantity of trees that would affect the LiDAR dataset.
No comments:
Post a Comment