Lab 8: Spectral Signature Analysis & Resource Monitoring

Goals and Background
The main goal of this lab was to introduce the concept of spectral reflectance and perform basic functions for the interpretation of spectral signatures. This included collecting spectral signatures  for a dozen different surface features, graphing them, and analyzing them. Also, spectral analysis would be used for the classification and analysis of vegetation health and ferrous soil concentration.

Methods
Part 1: Spectral Signature Aanlysis
First, the eau_claire_200.img Landsat ETM+ image was brought into a Erdas Imagine viwer. Then, using the drawing tool, a polygon was drawn around a lake located in the image. Then, using the Signature Editor tool, a new Signature form was created from the AOI of the drawn polygon. This signature was labeled "Standing water". This process was repeated for eleven more surface materials: moving water, forest, riparian vegetation, crops, urban grass, dry soil (uncultivated), moist soil (uncultivated), rock, asphalt highway, airport runway, concrete surface (bridge, parking lot, or any type of concrete surface).
Part 2: Resource Monitoring
In this section, simple band ratio was used to monitor and classic both vegetation and soils. The Raster Unsupervised tool "NDVI" was first opened in Erdas. The image ec_cpw_2000.img was input into the the Indices interface. The sensor was set to Landsat 7 Mulispectral and the function was set to NDVI. The process was then run. The image was then used to create a visually appealing map illustrating vegetation in the image area.
A similar process was completed again, this time for the ferrous mineral content of soils of the ec_cpw_2000.img image. The Raster Unsupervised tool Indices was launched, with the ec_cpw_2000.img image being input into the tool. The sensor was set to Landsat 7 Multispectral and the function set to "Ferrous Minerals". The tool was then run, and the output image used to cartographically pleasing map.

Results
Part 1
The resulting signature mean plots, taken from the twelve created spectral signature, show that features made up of similar materials seem to have similar spectral signatures. Man made surfaces have similar reflectance, vegetation types share similarities, exposed soils are similar. Several of the surface feature types, like exposed rock and cropland, are different from most other signature types.

Part 2
Based on the resulting image, the majority of health vegetation lies to the eastern half of the counties, or lines the very edge of the river. This is likely because the west is more densely populated than the east, and it is difficult to populated the area directly on the edge of the river.

Based on data taken from the map, the densest quantity of ferrous minerals lies to the direct west of the Chippewa river. This is likely what used to be the floodplain of the Chippewa River, and the area least covered by dense vegetation, exposing the soil.

Sources

Earth Resources Observation and Science Center. In United States Geological

Survey. Retrieved 12/14/2016, from http://eros.usgs.gov/usa

Wilson, C. (2016). Lab 8: Spectral signature analysis & resource monitoring. Eau Claire, Wisconsin

Lab 7: Photogrammetry and Orthorectification

Goals and Background
The primary goal of this lab was to introduce and develop skills portraying to aerial and satellite images. Specifically, these skills included learning how to calculate photographic scales, measure the area and perimeter of surface features in an image, and calculate the relief displacement. However, the primary focus of this lab was to introduce and perform the stereoscopy and the function of orthorectification on satellite images. Orthorectification is a valuable tool that removes image perspective and relief effects from an image.

Methods
Part 1: Scales, Measurements and Relief Displacement
The first part of the lab dealt primarily with performing measurements and calculations on images. With the first image, we were tasked with calculating the scale of a photograph, Eau Claire_West-se.jpg. This was done by measuring the distance between two points on the image while it was viewed under specific size parameters, in this case the basic maximized view of Windows Photo viewer, and using the real world distance between the two points, 8822.47 feet. Scale was then calculated using the equation "s= pd/gd".
The photographic scale was determined for another photograph, ec_east-sw.img, was calulated as well. This was done using the known quantities of the aircraft altitude of 20000 feet, the area's elevation of 796 feet, the focal lens length of 152 millimeters, and the equation "s= f/(H-h)".
Using the ec_west-se.img photograph, we were tasked to measure the perimeter and surface area of a lagoon marked "X" in the image. This was done using the "Measure Perimeters and Areas" digitizing tool in Erdas.
As a final section of part one, the relief displacement of a tower labeled "A"in the ec_west-se. image needed to be calculated. This was done by measuring the distance between the tower and the principal point of the image while it was viewed in a maximized Windows Photo viewer and using the known scale of the photograph, the height of the camera above the datum, and the equation "d = (h*r)/H".
Part 2: Stereoscopy
In this segment, the goal was to generated a three dimensional image using an elevation model. This would allow for the visual identification of relief displacement. This was done using the Terrain tool "Anaglyph" in Erdas Imagine. The input DEM for the tool was the ec_dem2.img. file, and the input image was the ec_city.img. The verticle exaggeration was set to "1" and the tool was run, with the output being saved as ec_anaglyph_sec1.img. The output anaglyph was then opned in an Erdas viewer and examined. At this point, it is worth noting that a second anaglyph would have been created. However, the eau_claire_quad.img file had become corrupted some how, and many students were unable to access or load the file.
Part 3: Orthorectification
This part of the lab introduced the Erdas Imagine Photogrammetric Suite (LPS), which was in this case used primarily for the function of orthorectification. This process would create a planimetrically true orthoimage. This was done in a multistage process.
First, a new block file was created in the LPS Project Manager of Erdas and the model setup to a Polynomial-based SPOT pushbroom. The horizontal reference coordinate system was then properly selected. The Spot_pan.img file was then added as a block and the sensor was specified.
The point measurement tool of the LPS Project manger was then selected. From here, the xs_ortho.img file was selected as the reference image. From here, a series of Ground Control Points (GCPs) were added to the Spot_pan image and the corresponding location on the xs_ortho image. A total of nine pairs of GCPs were collected. An addition two pairs, labeled 11 and 12, were collected using a different horizontal reference source, that of the image NAPP_2m-ortho.img. The skipping of point "10" was done to create a clear distinction between the two horizontal surfaces. The elevation information of the points was collected by setting the vertical reference source to that of the DEM file palm_springs_dem.img. The "Type" of each point was then set to Full, and the "Usage" was set to control. The point measurement tool was then saved and closed.
The second image of the area, spot_panb.img, was added to the frame, with the frame properties being defined. The GCPs which existed in both pot_pan and spot_panb had there coordinates defined in spot_panb. The point measurement tool was then saved. The Automatic Tie Point Generation Properties tool was then opened, with its parameters then being properly defined. This was used to generate additional GCPs without manual collection. Afterwards, the newly generated points were examined for correctness and the Point Measurement tool once again saved. Then, the Triangulation properties tool was opened, with many of its parameters needing to be defined. After this was completed, the triangulation was run and the report saved. In the IMAGINE Photogrammetry Project Manager, the project was then saved.
The "Start Ortho Resampling" tool was then selected, with the DEM being defined to the palm_springs_dem.img file and the cell sizes set to 10 x 10. The output was saved as orthospot_pan.img. The resampling method was set for bilinear interpolation. A second output was added for the spot_panb.img file, with the output being saved as orthospot_panb.img. The process was then run, with the outputs then displayed in a Erdas Imagine viewer.
Results
Part 1
The photographic scales of the Eau Claire_West-se.jpg and the ec_east-sw.img phot graphs were determined to be "1:38498.05" and "1:38059.07", respectively. The area of the lagoon in the ec_west-se.img file was determined to be 35.9395 hectares, or 88.80843857 acres, and its perimeter was determined to be 4149.01 meters, or 2.57807529 miles. The relief displacement of the tower in the ec_west-se.img file was determined to be .3527 inches.

















Part 2
The anaglyph shows a clear shift in elevation near the river and along an incline near the center of the image. However, buildings and surface features other than the terrain do not appear three dimensional in the image. Instead, these features remain two dimensional in appearance.

Part 3
The orthorectification process has clearly fixed most of the issues with the original two images. The images, which originally did not overlap very well, now line up almost perfectly.




























Sources

Wilson, C. (2016). Lab 7: Photogrammetry. Eau Claire, Wisconsin.

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 Conservation Service, 2010.

Lidar-derived surface model (DSM) for sections 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.