Supervised Classification

 

Spatial Enhancement, Multispectral Data, and Band Indices

 
This week's lab focused on my neck of the woods in the Pacific Northwest.  It was fun getting to learn about different spectral combinations to identify features in a region that I am very well acquainted with.

The focus this week was to learn how to use different processes to identify features in an image. Some of these processes included the use of histograms, image enhancement methods, and the use of spectral characteristics.

Introduction to ERDAS and Digital Data

This weeks lab had us focusing on using ERDAS Imagine. This was my first time using this software so, this lab was a little slow going at first. Especially since the lab instructions were clearly created using an older version of the software which resulted in a lot of "hunt and peck" on my end. However, once I got used to it, it became more intuitive. 

I appreciated the second half of the lab because it was focused on showing the functionality and utilization of tools in ERDAS. There are quite a few similarities to ArcGIS but overall, ERDAS is a little more user friendly.
 

Land Use Land Cover and Ground Truthing

 

Ariel Photography: Visual Interpretation

The objective for exercise 1 was to identify the tone and texture of an aerial photograph. Tone can be described as the brightness or darkness of an area so, the task was to create polygons around areas in the photograph that could be described as very light, light, medium, dark, and very dark. The next task was to identify areas of varying texture which involved a similar process to that of identifying tone. Texture in this case is defined as how smooth or rough the surface appears to be. For this task, polygons were drawn around areas perceived as either very fine, fine, mottled, coarse, and very coarse.

In these tasks, I did accomplish identifying areas of varying tones and textures. I had some challenges because I had to relearn how to create polygons but once I figured it out, the process went smoothly. My next challenge was labeling the polygons. In my first few attempts, the labels did not appear. I believe the problem was simply that the text was too large for the polygons so after adjusting a few settings, I was able to see the labels. 

 The objective of exercise 2 was to identify areas in an aerial photograph by the shape and size (an easily recognizable object like a tree or building), shadow ( an object that can be recognized by the shadow it casts like a water tower), pattern (an area that is recognizable not by a single object but by multiples of an object such as a farm), and association (areas in which conclusions can be drawn based on surrounding features).

With these tasks, new features had to be created, similar to exercise one, but these features were not drawn polygons and were instead points being made on the photo. This exercise went much smoother for me as I had worked out any difficulties in the process for the first exercise.

Google Earth

Screenshot of South Florida Map in Google Earth
 

Well, here it is, the final project for Computer Cartography. This was a great project to end on, I certainly enjoyed working on it.

In this lab, we were instructed to convert data in ArcGIS to KML, use that file to help create a Google Earth Map, and then create a tour with the finished map. 

Using ArcGIS, population, surface water, and county data were uploaded and then converted to a KMZ file using the 'Layer to KML' geoprocessing tool. The KMZ file was transferred and displayed in Google Earth. The resulting map was of the population of South Florida using dot symbology. The dots were layered over a map of Florida which included county boundaries and bodies of water.

The next step in this assignment was to make a tour of the map. This is where it got fun. Within google earth you can elect to add placemarks to areas of interest. That placemark will be shown at the extent in which it was placed. For the purposes of our map tour, we began with an overview of 23 counties in Southern Florida, then zoomed a little closer to show the Miami metropolitan area. From there, we zoomed to the placemark in downtown Miami. Likewise, we showed the cities of Fort Lauderdale, St. Petersburg, and Tampa during this tour as well.

Creating this map and the tour within Google Maps was much easier than I had anticipated it being. That being said, it was still very engaging, and I learned a lot from it. 

Isarithmic Map

 This weeks module covered isarithmic maps. In this lab, we were tasked with creating a precipitation map of Washington state depicted with hypsometric tints, hillshade effect, and contours. Data was acquired from the USDA Geospatial Data Gateway and the PRISM Group at Oregon State University.

In order to show continuous data throughout the map, the data needed to be interpolated. The precipitation data was interpolated using the Parameter-elevation Relationships on Independent Slopes Model (PRISM), which is an analytic tool that uses point-based climate measurements to predict monthly or annual climate behavior. Another useful feature of PRISM is that is accounts for variables in the map such as terrain and elevation by utilizing a digital elevation model (DEM) which enhances the accuracy of climate data.

To complete this map, hypsometric tinting, also known as elevation coloring, was implemented. Hypsometric tinting is used to show elevation differences on a map by using different colors on different elevation ranges. For this map, the hypsometric tinting was implemented by using the geoprocessing tool called Int (Spatial Analyst Tool) and inputting the annual precipitation raster data. Next, contours were overlayed on the hypsometric tints. This was performed using another geoprocessing tool called Contour Listand run using the same annual precipitation raster data.

The result is a map that illustrates how precipitation is impacted by elevation and terrain, which is important for understanding patterns in the climate of any given area.