Project 2 - Prepare Landcover Classification

It took quite a few tries but I was finally able to create a signature file with 50 classes for a mosaic of all clipped photos in the project. The maximum likelihood tool worked well on the mosaic signature but the slice tool did not do a very good job of reclassifying the image. I manually reclassified with reclassify tool by pixel value and used the results in this weeks submitted map.

Project 1 - Report

Project 1 - Base Map

Hospitals?

Special Topics - Project 1 Metadata

Links to new table metadata files for project 1:
Asthma
Census 2000
Ozone
Particulate 2.5

Not being from sea level I feel disadvantaged doing this analysis. The light blue areas to the north seem like they would be water. At .35 m. elevation would this be concidered wet or water? Is ocean elevation 0.0 m. elevation? Is a mile high elevation in the mountains?

Mod 4 - Supervised Classification and Recode

Germantown Land Use Map
The purpose of this map is to view and assess the city of Germantown, Maryland's land use pattern and growth trends in order to locate and build a new town center with respect to Maryland's Smart, Green and Growing Initiative. Satellite imagery has been classified and coded based on areas ground truthed in each category. Classification signatures were evaluated with histogram values and mean plot charts. Spectrally confused classes were edited until the values were accebtable then similar classes were grouped and recoded as reflected in the legend.

Evaluating and revising the signatures waas the hard part of this challenge for me. I wasn't sure if my criteria for evaluating the histograms and mean plots was valid.

Week 3 _ Orthorectification

Mod 3 Challenge Link

A simple undo button on the ERDAS GUI would make this program so much more user friendly. Oh and don't forget to save often and wear your helmet. This was otherwise a fun exercise, I like learning new software.