Human Geography
Analyzing Earth With Satellites
Remote sensing is the art of identifying, observing, and measuring an object without actually coming into direct contact with it. This process involves measuring radiation of different wavelengths that is reflected or emitted from distant objects. These objects are then organized by class, substance, and spacial distribution.
Everything on this earth, unless it is at a temperature of absolute zero, emits energy in the form of electromagnetic radiation. As the temperature of an object increases, the electrons vibrate faster which decreases the peak of its emitted wavelengths. The smallest possible amount of electromagnetic energy of a particular wavelength is known as a photon, which moves at the speed of light. The more light a photon possesses, the more energy it contains. The electromagnetic spectrum contains the entire array of electromagnetic waves. The spectrum has been divided into different regions that are organized by frequency and size of wavelength. Some wavelengths are visible, while others are only detected as heat.
Our atmosphere contains different gases that absorb different types of radiation. The transmissivity of each atmospheric gas depends on the type of radioactive wavelength that is passing through. There are certain areas, known as windows, in the electromagnetic spectrum where the atmosphere is transparent to certain wavelengths. Most remote sensing instruments measure the wavelengths that pass through these windows, however some devices measure the absorption of the wavelengths in the atmosphere.
Remote sensing is often used to separate the features of an image into different categories or classes. Different surface materials have different spectral signatures, meaning they each reflect different distinguishable wavelengths. Because of this, remote sensing can often detect the type of surface cover in a specific area and gather detailed information to create a thematic map.
Satellite imagery is made up of tiny squares that are called pixels. These pixels, short for picture elements, are each a different shade of gray color and represent the relative reflected light energy for a specific part of an image. Computers convert the reflected light into pixels and keep track of every number for every pixel.
Color is used in satellite imagery to gather more detailed information from an image. Computers can display the gray tones as red, blue, and green light to gather more data. We must understand what each of the colors mean in order to translate the data correctly.
Tutorial #1: Display & Image Inspection of Image Data  In the first tutorial, I learned how to open an image and how to perform a few other basic functions in the MultiSpec program. When I first opened the file I changed the display to the "3-Channel Color" display type. After the image was displayed, I learned how to zoom in and out and then learned how to change the display to "side by side channels." After changing the display and zoom, I learned how to change the map parameters and also experimented with a few different coordinate views. The last step was to change the view to "1 Channel Thematic Display Type," which is what is shown in my final image. Tutorial #3: Unsupervised Classification  The third tutorial involves Cluster analysis and the use of the ISODATA algorithm. As you can see in the image above, the similar pixels have been grouped into clusters or categories. I began this tutorial by setting the Cluster Specifications and selecting the ISODATA algorithm. I chose 7 clusters and set the convergence percentage to 100. After the pixels were sorted into their different classes, I opened the Thematic Display Specifications box which created a legend. With this legend I analyzed the different classes and learned how to put the clusters into groups. | Tutorial #2: Image Enhancement  In the second tutorial, I learned how to control the enhancement of an image by experimenting with 5 different options: Bits of color, Stretch, Min-maxes, Treat 0 as, and Display levels per channel. I began by adjusting the "Bits of Color" to 24 and the "Display levels per channel" to 256. Next, I changed the "Treat 0 as" setting from data to black and then to white. Following that experiment, I moved on to the "Stretch" and "Min-maxes" options. I changed the Stretch from Linear to Equal Area to Gaussian and compared the differences. I then tried some different "Min-maxes" options to see what they did. My final image is with an "Entire range" setting. Tutorial #4: Supervised Classification  The final tutorial for this assignment involves selecting training areas for specified classes from known areas. It is not shown in the image above, but I labeled each training field as either trees, image blank, weeds, soil, or light soil. I began this tutorial by selecting training fields. I dragged a box on each field and labeled them accordingly. After this, I set the classification specifications and viewed them on text output to verify accuracy. Next, I opened the classification map and added my field outlines and then opened the classification probability map to ensure that all portions of the image were classified correctly. |
Assignment Reflection:
Remote sensing can be used for many reasons. Remote sensing images could be used to monitor the surface of the earth, which can be helpful in predicting natural disasters such as volcanic eruptions, earthquakes, and landslides. Remote sensing is also useful in measuring vegetation and the spacial distribution of other elements on the earth including the ocean. A big reason we know so much about global warming is because of remote sensing.
Remote sensing can help us understand a specific region physically and culturally by gathering detailed images that can be analyzed and interpreted without having to be physically present in that specific area. In the video "Geospatial Revolution," remote sensing images were used to view the conflict in Darfur. Because of these images, we were able inform their government that we could see the destruction of their villages.
This assignment was a lot of work, but I'm glad I did it because I learned a lot about remote sensing. I didn't know much about what remote sensing was or how it worked before I did this assignment. I didn't understand radiation and the electromagnetic spectrum until I read the NASA article. I also learned a lot from the tutorials and was able to see in more detail what remote sensing can tell us about a specific location. Geospatial technology is fascinating to me and I'm excited to learn more about it.
