My Geographic Information Journal

  This map was created as a LULC for the city of Germantown, Maryland for educational purposes only. I classified the pixels of this image by creating classes from "seed", which involved finding different examples of class types and essentially making all of the similar pixels across the image have the same color. After doing all the required class types, I realized there was some classes being grouped together due to the band combinations being used. I utilized the histogram and mean plot tools to figure out the best band combination (im my opinion) to separate all of the classes and make them easier to pick out. This one ended up being Red - 5, Green - 4, Blue - 3. The biggest issue with these classes is separating the roads from the urban areas, but I think it is somewhat easy to pick out the roads as they are a different shade of purple. The map also has the calculated areas of each land use type, shown in the legend. The lower right map is the Distance Image that I creat...

  This layout was made for educational purposes to learn how to identify features through only descriptions of reflectance properties. I was given criteria for each type of feature to find, such as spikes in pixel values. The first feature (top left) is a very large river or lake of some kind. It was the largest spike on the histogram as well as the furthest left, meaning that it was very large and dark! The second feature (bottom left) is a snowy mountain top. This type of feature has large contrasts in brightness and size depending on what layer is being viewed. The third feature (bottom right) is of a water sandbar area that has large spikes in brightness all the sudden. This is because water is typically very dark, but having a large amount of sand right under the water will quickly cause it to change brightness. 

  This map was for educational purposes to help learn to find data about features of an image through ERDAS Imagine. I learned how to capture a specific location on a larger map and then find area data about that location. Once the new area had been clipped out and created as a new layer, all new data was able to be extrapolated about it. This makes it helpful when looking at an image to understand how much of it actually being covered by these different features, instead of having to guess. 

  This was my first time doing a LULC map. I started on the largest areas I saw, which were the natural areas. To the left of the map was bays and estuaries, streams and canals, forests and wetlands. I clumped some of these together because they looked to all be intertwined into a couple different habitats. I then followed some different streams into the city and found any natural areas within the city. Then I started at the residential neighborhoods in the bottom left of the map and followed it until it was obstructed by a commercial building, industrial complex, or something else other than residential houses. I labeled these other areas as I went. Then I did the main road and some other miscellaneous areas that hadn't been filled out yet. This was an interesting project that taught me about the layout of a city or overall area.