Unsupervised Classification

Photo Interpretation & Remote Sensing - Mod 9

This week's less in Unsupervised Classification was pretty straightforward and simple, though somewhat time consuming.  Most of the work was done in Erdas Imagine starting with running the Unsupervised Classification tool on a high resolution aerial photograph of the UWF campus.  This

resulted in a thematic raster that allowed us to simplify the image into fewer classes by selecting the pixels and changing them.  Our task was originally to classify the image into four categories; Trees, Grass, Building/Roads, and Shadows.  This seemed simple enough until some of the clusters affected multiple features.  In order to deal with this problem a Mixed class was added as well and anything that affected more than one feature could be added to that.  Sometimes the affect on a second feature was so limited it made more sense to keep that class as the first feature.  Once we had the image classified we merged the classes so we had only the five specified.

Next we added fields to the attribute table for Class Name and Area.  This allowed us to calculate the area for each of the classes so we could determine the percentage of permeable verses impermeable surfaces.  Permeable surfaces include Grass and Trees, impermeable, Buildings\Roads.  The other two classes included a combination of both, so before we could make that calculation we first had to calculate what percentage of each of those classes were made of each surface.

Thermal & Multispectral Analysis

Photo Interpretation & Remote Sensing - Mod 8

This week's assignment was to use image manipulation and interpretation techniques to identify a feature from an image using the thermal infrared band as part of the analysis.  The selection was made in Erdas Imagine using the TM Thermal Infrared Composite band combination of red for the thermal layer Band 6, green for Band 3 and blue for Band 2.  In ArcMap the same image was symbolized with a composite combination of red for Band 1, green for Band 2 and blue for Band 3 for a more real world image, which made the feature clearer and more easily identified as an airport.  The coordinate for this feature are 30° 28' 34.8306' N, 86° 31' 7.3224' W and allowed it to be identified as Destin - Ft. Walton Beach Airport (VPS) in northwest Florida.

Statistical Analysis with ArcGIS - Prepare Week

Special Topics - Project 3 Week 1

The week starts off with a new project:  Statistical Analysis of Methamphetamine Laboratory Busts in West Virginia.  For Prepare Week we had to do a bit of reading up on Methamphetamine, it's

history and its users.  We also created a basemap to use in upcoming weeks to display and report on the analysis of the data we were supplied or downloaded in order to assist the government and law enforcement in anticipating future criminal activity.

The study area of census tracts for Kanawa and Putnum Counties was provided along with a point file of Charleston Meth Labs.  The cities, roads, river, and counties of West Virginia were downloaded from the US Census Bureau's Tiger Shapefile site.

In preparation for the analysis process to occur next week a spatial join was performed on the census tracts and meth lab layers in order to combine the attributes of both tables into one.  Once the attribute tables were combined the unnecessary fields were turned off and this is what we were left with:

Some of the attribute names were a little difficult to interpret.  Best guess is sometimes the best you can do.  It will be interesting to see what we do with them next week.

Image Preprocessing 2: Spectral Enhancement and Band Indices

Photo Interpretation and Remote Sensing - Mod 7

This week was very challenging.  We worked with histograms a lot this week learning how to analyze them in order to interpret images.  Our final exercise was to identify certain areas on the map based on histogram information.  That was very challenging, but by the end I had a little better understanding of this week's lesson.  Here are the maps created:

Mountain Top Removal - Report Week

Special Topics - Project 2 Week 3 & 4

This week was the culmination of all the prep work and analysis performed in previous weeks.  We had to publish our group Mountain Top Removal (MTR) Analysis map on ArcGIS Online UWF Org, complete our Story Map Journal and add our published map to that, along with a link to this blog.

Mountain Top Removal (MTR) is a method of mining for coal that destroys a mountaintop or ridgeline.  All too frequently the plans described in work permits are significantly different from the actually mining activities that occur on the ground.    Remotely sensed data has been used to investigate evidence of human caused changes to landscape as a result of MTR and to compare what was permitted to what has actually taken place.  The data and methods were provided by SkyTruth, a non-profit agency monitoring MTR in the Appalachian Coal Mining Regions within the states of Tennessee, Kentucky, Ohio, Virginia and West Virginia.  This dataset was created by Group 3 of the University of West Florida’s (UWF) Online GIS Certification Program 2016 class and will be shared with SkyTruth for a comparison study.  Group 3 students include; Rachel Hamaty, Maggie Roth, Charmaine Hingada and Austin Adkison.

Satellite imaging was chosen as an independent and cost-effective method of identifying, mapping and quantifying landscapes disrupted and altered by MTR.  2010 Landsat data was used to create this dataset of polygons covering those areas of MTR in eastern Kentucky.  It is a compilation of Landsat images LT50190332010243EDC00 and LT50190342010243EDC00.  The combined accuracy is 97%, with a total acreage of 131,144 acres.  This dataset contains only those areas 40 acres or larger.  Areas within 50 meters of roads and rivers or within 400 meters of major rivers and highways have been removed.  With those exceptions, only those areas that intersect with mountain ridges have been included.

The Story Map Journal is basically a compilation of slides with a column on the side for descriptive text.  My Story Map Journal, A Journal of Mountain Top Removal, starts with an introduction of MTR, then background of MTR and the role of GIS in defining and analyzing it, an image of the study area my group was responsible for, an image of our analysis that we published though ArcGIS Online, and finally a discussion through a link to this blog.  Below is the link to my Story Map Journal:

http://arcg.is/2daNCgz

Image Processing 1: Spatial Enhancement and Radiometric Correction

Photo Interpretation and Remote Sensing - Mod 6

This week we used radiometric and spatial enhancements to enhance an image and reduce striping.  

The first step was to perform a Fourier transformation in order to run some of the Fourier tools in a Fourier Transform Editor.  Prior to running this step the image was just a big white blur with a few black splotches until it was zoomed to 1:150000.  After running the Fourier Transform Editor tools it was a complete image that could be zoomed to its extent, but it still had striping.  I tested numerous tools trying to find that right combination that would lessen the stripes without diminishing the clarity of the image but nothing I tried work.  Finally I settle on an image that still had all the stripes it came out of the Fourier Transform Editor with, but the clearest image I had managed to achieve.  The tools I used to accomplish this after the Fourier Transform Editor were the Convolution tools Sharpen and Haze Reduction in ERDAS Imagine and in ArcMap I used the Spatial Analyst Focal Statistics tool with a width and height of 3 and a Statistics type of Range.