Lab 8: spectral signiturre analysis

In this lab students conducted exercises that better their understanding of spectral reflectance signatures of various Earth surface and near surface features.  To practice this skill students used a Landsat ETM+ image of north central Wisconsin and eastern Minnesota to digitize and record spectral tendencies of various features.

specficaly we were asked to collect spectral data on the following features:

To obtain these spectral signatures, all one has to do was draw a polygon containing a uniform piece of the land/feature you are interested in and run an operation using the signature editing tool under the raster tab to collect the data.  after collecting the information this is what I was able to obtain:

The data collected showed that plants generally reflect a lot of green light and absorb longer wavelengths.  The only differences was that some plants absorbed more levels of light, indicating that they were more active/were producing more food for themselves to consume.  man made features like roads, parking lots and runways had high reflectivity in the longer wavelengths, but absorbed more blue light than other features. 

This skill could be one of the most important we've learned so far in this class, in that it can be applied to track the fluidity and function of various systems in the natural environment.

 

Lab 7: Photogrammetry

The goal of this laboratory exercise was to develop the skills necessary to preform photogrammetric tasks on aerial photographs or satellite images.  Specifically, this lab is tailored to train students to better understand the mathematics involved in calculating scales, measuring areas and perimeters of features, and calculating relief displacement.  Moreover this lab also offers an introduction of to stereoscopy and satellite image orthorectification.

In the first 3 sections of the lab, students conducted used both mathematical and computer generated methods to discover either the scales, record area/perimeters, and measure relief displacement.  The measurement of area/perimeters was done via the ERDAS imagine digitizing tool. 

The calculating of scales was done using the equation: s = f/H' where;
S= scale
f = focal length
H'= height of image taken - height of area above/below sea level.

calculating relief displacement was done using the equation: D = (hxr)/H where;

D= displacement
h= height of object
r= radial distance object is from principle point
H= height of plane taking photo



Part 2: Stereoscopy

In this portion of the lab, students uploaded two images into ERDAS imagine, one was an image of the city of Eau Claire and the other was a digital elevation model of the same area (DEM).  Students than used an anaglyph operation to create stereoscopic image of the area.  With polarized glasses, looking at this image produces a 3d image of the area showing both the features of the land, and the differentiating relief across the city. 

The image produced was accurate in its relief, but also introduced an element of geometric scaling error.  This error is produced when an image has height variation, but has the same general scale.  The result is taller features displaying relief displacement, where they are oriented at odd angles in relevance to their actual appearance in the real world

here we see a smoke tower that is altered due to relief displacement
on the upper campus UWEC
 

Part 3: Orthorectification

orthorectification is a very intensive process that involves heavy input from the user.  The subject area of the images being corrected was that of Palm Springs, California.  By using the Lecia Photogrammetric Suite  (LPS) digital photogrammetry tool, students used control points and tie points to triangulate, and ultimately orthorectify the images we collected into one.  we used multiple scale images and also a digital elevation model to create the output image where relief displacement and other geometric errors have been removed, along with an overall improvement of accuracy.