Lab 7: Photogrammetry
Goal:
The goal for this lab was to become familiar with preforming photogrammetric tasks. Also, this lab is meant to understand the mathematics behind photogrammetry. The topics that this lab covers are scale calculation, relief displacement calculations, measuring features, to create stereoscopic images, and to preform rectification on satellite images.
Methods:
Part 1:
Section 1:
The first section of the first part of this lab is about finding the scale of the satellite image. For this section, an image was provided with two points marked on it. The real world distance between the points was given and the distance between the points needs to be found. Once this number is found, the real world distance and the distance on the image need to be converted to the same units. Then divide the screen distance by the real world distance. This is the scale of the image, 1 to the result.
The second images scale needs to be calculated from the the focal length and the aircraft altitude above ground level. This is found by dividing the focal length by the flying height of the aircraft above ground level. The information that was given was the focal length and the altitude of the aircraft above sea level as well as the elevation of the study area. The height above ground level can be found by finding the difference between the altitude above sea level and the elevation of the study area.
Section 2:
This section is about calculating areas and perimeters in aerial photographs. This can be done in ERDAS under the measure tool set. Using the point drop down and selecting the polygon tool, the desired shapes outline can be digitized. Once the polygon is created, the window the shows units in the measure tool set will give the measurements of the digitized shape. The units can be changed by selecting the unit drop down and selecting the desired unit.
Section 3:
This section is about calculating the relief displacement in an image. This is found with the formula: relief displacement = (height of object in the real world * radial distance from the top of the object to the principle point) / height of the camera above the local datum. This will provide the relief displacement in the image.
Part 2:
Section 1:
This section of the lab is about creating anaglyphs, special type of image used with polaroid glasses to make the image look 3 dimensional. The first anaglyph that is create uses a DEM created from satellite images. In the ERDAS viewer, under the terrain toolbox, the anaglyph tool is used. The input images should be the DEM created from satellite images and a satillite image of the same area showing the visible spectrum. After the photos are uploaded and the output file is named, the tool can be run. This will result in a anaglyph image of the area.
Section 2:
The section of the lab is creating anaglyphs the same way as part 2 section 1 but instead of using a DEM, a DSM created from LiDAR is used. The DSM is used in place of the DEM. This results in another anaglyph.
Part 3:
Section 1:
This section of the lab is creating block file for preforming orthorectification. In ERDAS, under the toolbox tab, there is a tool called Imagine Photogrammetry. Selecting this tool opens a new window. In the new window, the left most symbol needs to be selected. This will start the process to create a new block file. The new block file opens up a model setup window. In this window, the sensor type needs to be chosen. For this lab, spot was used so the polynomial based push broom is selected as well as spot push broom. Next, the horizontal reference system needs to be defined by selecting the projection type, spheroid name, datum name, zone, and cardinal direction. Once this is complete, clicking ok will save the block file with the proper coordinate system.
Section 2:
This section is about adding the desired imagery to the block file. This can be done while in the image photogrammetry window. By selecting the images folder in the table of contents and clicking the add from icon. From here, the desired image can be added to the file.
Section 3:
This section is about collecting the GCPs to correct the image. This is done by selecting the start point measurement tool and then selecting the classic point measurement tool. Next, clicking the reset horizontal reference icon bring up a new window. In the new window, selecting image layer will open a new window. In this new window, the reference image can be navigated to and selected. Once the image is uploaded, clicking the use viewer as reference to view both images. Next, GCPs are collected and placed in the same real world location on both images. Once a couple of GCPs are placed, using the automatic (x,y) drive tool, the program will assist in finding the rest of them by trying to estimate where the GCP should be on the original image when a GCP is placed on the reference layer. Next, clicking on the reset vertical reference source is selected which will bring up a new window. In the new window, DEM can be selected and the desired DEM file can be navigated to and selected. Next, highlighting the entries for the GCPs, the update z values on selected points can be clicked. This included the z value for each GCP to go along with the x and y values.
Section 4:
The first step in this section is to change the type and usage for the points. This can be done by highlighting the desired column and right clicking the title and select formula. In the formula window, the desired designation can be typed and it will apply that to all the entries in that column. Once the entries are correct the file can be saved to ensure there is no data loss. Next, a new image is added to the frame in the same way as part 3 section 3 and GCPs need to be collected in the same way.
Section 5:
This section is about using automatic tie points collection, triangulation and ortho resampling. Automatic tie point collection can be done by selecting the automatic tie point generation properties icon. This opens a new window with options for the tie point collection. For this section, the images used part is set for all available and the initial type is set to exterior/headers/GCP and under the distribution tab, the intended number of points/image is set to 40. Then run can be selected to start the automatic tie point collection. This adds more GCPs and adds tie points. Once the points are verrified accurate, the save button can be clicked. Back in the project manager window, under edit, the triangulation properties can be opened. In the window, the iterations with relation value needs to be changed to 3 and under the point tab, set the type to same weighted values and X Y and Z values need to be set to 15. Once this is done, the tool can be run. A summary report will apear when the tool is done running. The report will have various information about the triangulation. Next, the start ortho resampling process icon can be selected. In the general tab, the DTM source can be chagned to DEM and the desired DEM can be navigated to and uploaded. The output cell size needs to be changed to 10 and the output file name needs to be declared. Next, clicking add allow input of an image. For this, two images from spot were used.
Section 6:
This section is about viewing the orthorectified images. They can be brought into the viewer by clicking the add data button and navigating to the new images created in the previous step. Viewing the images allows and mistakes to be seen because the images will not fit together.
Results:
Part 1:
Section 1:
This section resulted in the scale for the two images. The first images scale was found by comparing the same areas length in the real world and on the map. The real world distance was given as 8822.47 ft and the distance on the map was 2.75 inches. Both values need to be in the same unit so the real world distance was converted to inches by multiplying 8822.47 by 12 giving 105869.64 inches. Then take 105869.64 inches and divide it by 2.75 inches. This will give a scale of 1:38,498.05.
The second image used the formula of scale = focal length / flying height of the aircraft above ground level. To find the flying height above ground level, find the difference between the height of the aircraft (20000) and the elevation of the study area. The focal length also needs to be converted from mm to ft to keep all units the same. The scale = .498/19204 which gives a scale of 1:38510.59.
Section 2:
This section resulted in the area and perimeter of a lake in Eau Claire. After digitizing the lake the dimensions can be found. The lake is 94.8 acres or 38.3 hectares and has a perimeter of 4038.22 meters or 2.5 miles.
Part 2:
Section 1:
This section resulted in an anaglyph of Eau Claire created from a DEM created from satellite imagery. This anaglyph can be seen below in figure 1. While wearing polaroid glasses, looking at the image causes it to look 3 dimensional. However only extreme elevation changes can be seen because of the lower resolution of the satellite image the DEM was created from.
Section 2:
Part 3:
Part 3 resulted in orthorectified images. The images started as raw satellite images and through photogrammetry and orthorectification, the images are accurate and have a coordinate system to they can be used in other work. The corrected images can be seen below in figure 3. The images have an almost seamless transition and without the black border, the would look like the same image.
Sources:
National Agriculture Imagery Program (NAIP) images are from United States Department of Agriculture, 2005.
Digital Elevation Model (DEM) for Eau Claire, WI is from United States Department of Agriculture Natural Resources Consercation Services, 2010.
Lidar-derived surface model (DSM) for section of Eau Claire and Chippewa are from Eau Claire County and Chippewa County governments Respectively.
Spot satellite images are from Erdas Imagine, 2009.
Digital elevation model (DEM) for Palm Spring, CA is from Erdas Imagine, 2009.
National Aerial Photography Program (NAPP) 2 meter images are from Erdas Imagine, 2009.
The goal for this lab was to become familiar with preforming photogrammetric tasks. Also, this lab is meant to understand the mathematics behind photogrammetry. The topics that this lab covers are scale calculation, relief displacement calculations, measuring features, to create stereoscopic images, and to preform rectification on satellite images.
Methods:
Part 1:
Section 1:
The first section of the first part of this lab is about finding the scale of the satellite image. For this section, an image was provided with two points marked on it. The real world distance between the points was given and the distance between the points needs to be found. Once this number is found, the real world distance and the distance on the image need to be converted to the same units. Then divide the screen distance by the real world distance. This is the scale of the image, 1 to the result.
The second images scale needs to be calculated from the the focal length and the aircraft altitude above ground level. This is found by dividing the focal length by the flying height of the aircraft above ground level. The information that was given was the focal length and the altitude of the aircraft above sea level as well as the elevation of the study area. The height above ground level can be found by finding the difference between the altitude above sea level and the elevation of the study area.
Section 2:
This section is about calculating areas and perimeters in aerial photographs. This can be done in ERDAS under the measure tool set. Using the point drop down and selecting the polygon tool, the desired shapes outline can be digitized. Once the polygon is created, the window the shows units in the measure tool set will give the measurements of the digitized shape. The units can be changed by selecting the unit drop down and selecting the desired unit.
Section 3:
This section is about calculating the relief displacement in an image. This is found with the formula: relief displacement = (height of object in the real world * radial distance from the top of the object to the principle point) / height of the camera above the local datum. This will provide the relief displacement in the image.
Part 2:
Section 1:
This section of the lab is about creating anaglyphs, special type of image used with polaroid glasses to make the image look 3 dimensional. The first anaglyph that is create uses a DEM created from satellite images. In the ERDAS viewer, under the terrain toolbox, the anaglyph tool is used. The input images should be the DEM created from satellite images and a satillite image of the same area showing the visible spectrum. After the photos are uploaded and the output file is named, the tool can be run. This will result in a anaglyph image of the area.
Section 2:
The section of the lab is creating anaglyphs the same way as part 2 section 1 but instead of using a DEM, a DSM created from LiDAR is used. The DSM is used in place of the DEM. This results in another anaglyph.
Part 3:
Section 1:
This section of the lab is creating block file for preforming orthorectification. In ERDAS, under the toolbox tab, there is a tool called Imagine Photogrammetry. Selecting this tool opens a new window. In the new window, the left most symbol needs to be selected. This will start the process to create a new block file. The new block file opens up a model setup window. In this window, the sensor type needs to be chosen. For this lab, spot was used so the polynomial based push broom is selected as well as spot push broom. Next, the horizontal reference system needs to be defined by selecting the projection type, spheroid name, datum name, zone, and cardinal direction. Once this is complete, clicking ok will save the block file with the proper coordinate system.
Section 2:
This section is about adding the desired imagery to the block file. This can be done while in the image photogrammetry window. By selecting the images folder in the table of contents and clicking the add from icon. From here, the desired image can be added to the file.
Section 3:
This section is about collecting the GCPs to correct the image. This is done by selecting the start point measurement tool and then selecting the classic point measurement tool. Next, clicking the reset horizontal reference icon bring up a new window. In the new window, selecting image layer will open a new window. In this new window, the reference image can be navigated to and selected. Once the image is uploaded, clicking the use viewer as reference to view both images. Next, GCPs are collected and placed in the same real world location on both images. Once a couple of GCPs are placed, using the automatic (x,y) drive tool, the program will assist in finding the rest of them by trying to estimate where the GCP should be on the original image when a GCP is placed on the reference layer. Next, clicking on the reset vertical reference source is selected which will bring up a new window. In the new window, DEM can be selected and the desired DEM file can be navigated to and selected. Next, highlighting the entries for the GCPs, the update z values on selected points can be clicked. This included the z value for each GCP to go along with the x and y values.
Section 4:
The first step in this section is to change the type and usage for the points. This can be done by highlighting the desired column and right clicking the title and select formula. In the formula window, the desired designation can be typed and it will apply that to all the entries in that column. Once the entries are correct the file can be saved to ensure there is no data loss. Next, a new image is added to the frame in the same way as part 3 section 3 and GCPs need to be collected in the same way.
Section 5:
This section is about using automatic tie points collection, triangulation and ortho resampling. Automatic tie point collection can be done by selecting the automatic tie point generation properties icon. This opens a new window with options for the tie point collection. For this section, the images used part is set for all available and the initial type is set to exterior/headers/GCP and under the distribution tab, the intended number of points/image is set to 40. Then run can be selected to start the automatic tie point collection. This adds more GCPs and adds tie points. Once the points are verrified accurate, the save button can be clicked. Back in the project manager window, under edit, the triangulation properties can be opened. In the window, the iterations with relation value needs to be changed to 3 and under the point tab, set the type to same weighted values and X Y and Z values need to be set to 15. Once this is done, the tool can be run. A summary report will apear when the tool is done running. The report will have various information about the triangulation. Next, the start ortho resampling process icon can be selected. In the general tab, the DTM source can be chagned to DEM and the desired DEM can be navigated to and uploaded. The output cell size needs to be changed to 10 and the output file name needs to be declared. Next, clicking add allow input of an image. For this, two images from spot were used.
Section 6:
This section is about viewing the orthorectified images. They can be brought into the viewer by clicking the add data button and navigating to the new images created in the previous step. Viewing the images allows and mistakes to be seen because the images will not fit together.
Results:
Part 1:
Section 1:
This section resulted in the scale for the two images. The first images scale was found by comparing the same areas length in the real world and on the map. The real world distance was given as 8822.47 ft and the distance on the map was 2.75 inches. Both values need to be in the same unit so the real world distance was converted to inches by multiplying 8822.47 by 12 giving 105869.64 inches. Then take 105869.64 inches and divide it by 2.75 inches. This will give a scale of 1:38,498.05.
The second image used the formula of scale = focal length / flying height of the aircraft above ground level. To find the flying height above ground level, find the difference between the height of the aircraft (20000) and the elevation of the study area. The focal length also needs to be converted from mm to ft to keep all units the same. The scale = .498/19204 which gives a scale of 1:38510.59.
Section 2:
This section resulted in the area and perimeter of a lake in Eau Claire. After digitizing the lake the dimensions can be found. The lake is 94.8 acres or 38.3 hectares and has a perimeter of 4038.22 meters or 2.5 miles.
Part 2:
Section 1:
This section resulted in an anaglyph of Eau Claire created from a DEM created from satellite imagery. This anaglyph can be seen below in figure 1. While wearing polaroid glasses, looking at the image causes it to look 3 dimensional. However only extreme elevation changes can be seen because of the lower resolution of the satellite image the DEM was created from.
 |
| Figure 1. Anaglyph created from DEM from a satellite image. |
Section 2:
This section resulted in an anaglyph created from LiDAR data. This anaglyph can be seen below in figure 2. Looking at this image while wearing polaroid glasses causes the images to look 3 dimensional. This anaglyph shows a lot more detail and small variation in the terrain including trees and small structures. This is because the LiDAR data that the DEM was created from has a better spatial resolution than the previous image had. This causes more of the individual features to pop out more.
 |
| figure 2. Anaglyph of Eau Claire created from a DEM created from LiDAR data. |
Part 3:
Part 3 resulted in orthorectified images. The images started as raw satellite images and through photogrammetry and orthorectification, the images are accurate and have a coordinate system to they can be used in other work. The corrected images can be seen below in figure 3. The images have an almost seamless transition and without the black border, the would look like the same image.
 |
| Figure 3. Orthorectified satellite images overlapping to show accuracy. |
Sources:
National Agriculture Imagery Program (NAIP) images are from United States Department of Agriculture, 2005.
Digital Elevation Model (DEM) for Eau Claire, WI is from United States Department of Agriculture Natural Resources Consercation Services, 2010.
Lidar-derived surface model (DSM) for section of Eau Claire and Chippewa are from Eau Claire County and Chippewa County governments Respectively.
Spot satellite images are from Erdas Imagine, 2009.
Digital elevation model (DEM) for Palm Spring, CA is from Erdas Imagine, 2009.
National Aerial Photography Program (NAPP) 2 meter images are from Erdas Imagine, 2009.
Comments
Post a Comment