Spectral Signature Analysis

Introduction:

The main goal of this lab is to give the student experience in the measurement and interpretation of spectral reflectance of various Earth surfaces with satellite images. This lab also teaches the student how to collect spectral signatures, graph them and perform analysis to determine if they pass the spectral separability test. This type of analysis is a prerequisite for image classification.

Methods:

This lab will have the student plot the spectral reflectance of twelve different earth surfaces. These surfaces include standing water, moving water, vegetation, riparian vegetation, crops, urban grass, dry soil (uncultivated), moist soil (uncultivated), rock, asphalt highway, airport runway and a concrete surface.

Once each feature had been identified, a small polygon was then digitized around each feature in Erdas Imagine. Once the polygon was created, the raster processing tools were then activated, this enabled the signature editor window to be opened. The signature editor window lets the user name each digitized feature as well as open the mean plot window. This plot allows for the student to identify which bands have the greatest and least reflectance for each feature.

Results:

The figures below are the results of spectral analysis performed on all twelve features.

Figure 1. This is the mean plot window for the
airport runway feature. 

Figure 2. This is the mean plot window for the
asphalt highway feature.

Figure 3. This is the mean plot window for the
concrete surface feature.

Figure 4. This is the mean plot window for the
crops feature.

Figure 5. This is the mean plot window for the
dry soil feature.

Figure 6. This is the mean plot window for the
moist soil feature.

Figure 7. This is the mean plot window for the
moving water feature.

Figure 8. This is the mean plot window for the
riparian vegetation feature.

Figure 9. This is the mean plot window for the
rock feature.

Figure 10. This is the mean plot window for the
standing water feature.

Figure 11. This is the mean plot window for the
urban grass feature.

Figure 12. This is the mean plot window for the
vegetation feature.

Photogrammetry

Introduction:

The goal of this lab was to teach the student how to perform key photogrammetric tasks on different aerial photographs and satellite images. The student will learn how to calculate photographic scales, measure area and perimeter of features as well as calculating relief displacement. The last task of this lab is to introduce stereoscopy as well as perform orthorectification

Methods:

The first section of the lab had the student finding the scale of different photographs. Data was provided and the student had to use formulas to find the scale for two different aerial photographs. The second task was measuring perimeters and areas of different polygon features within ERDAS Imagine. The third task was calculating the relief displacement from an object height. The student had to find the radial distance as well find its real world height by measuring the height in the photograph. The student then had to perform conversions in order to used their measured data within the provided equations. 

The second part of the lab introduced the student to Stereoscopy. The specific area being analyzed was the city of Eau Claire, WI. After performing stereoscopy on the images, polaroid glasses needed to be used in order to see the results of the analysis. The images produced allowed for the student to see the different elevation changes in levels, rather than gradual changes. 

The third part of the lab involved orthorectification. There were many tasks for the student to complete in this section of the lab. Some tasks included collecting GCPs, performing automatic tie point collection, triangulating images, selecting a horizontal reference source and orthorectifing images. 

Results:

All of the processes performed in the third part of the lab were intertwined. They all compiled on each other in order to come to an end result. The figure below (Figure 1) show the results of all the placed GCPs as well as the generated tie points. 

Figure 1 shows the different GCPs and tie points for both images.

The second image (Figure 2) is the results of all the processes combined together. The two images that were orthorectified were overlaid with each other to produce one image. 

Figure 2 is the result of the two orthorectified images being overlaid. 

Sources:

United States Department of Agriculture, 2005.

United States Department of Agriculture Natural Resources Convservation Service, 2010.

Erdas Imagine