Lab 04 - Ground Truthing and Accuracy Assessment

 

GIS 4035

 

 

In this weeks lab, I continued with the Pascagoula, Mississippi Land Use/Land Cover map that I created.  The class was to focus in on Truthing, and verifying that the data we surmised last week was true and accurate.  Utilizing Google Street Views I was able to look at sample site locations and determine if the classifications that I determined last week were in fact accurate.  Overall for this sampling I was 70% accurate when comparing last weeks assessments to this weeks samplings of the areas. 

 

In the above map, you can see my sample locations.  Those locations that I found to be of an accurate assessment were colored green.  However, if I looked at a sample location and found that the area represented in the classification was false, I symbolized the dot in a red. 



Lab 3 - LU/LC Classification

GIS 4035

 

 

 

This weeks lab, we looked at Land Use and Land Cover.  The above map of the Pascagoula, Mississippi area was created utilizing an aerial photograph.  Based on that aerial photograph and utilizing the skills we learned in the past two lessons I was tasked with identifying different locations within the photograph and classifying them based on their looks.  I utilized size, color, shape, surrounding features to identify each location.  The Minimum Mapping Units (MMU) that I used on this map was approximately 2.5 hectares (roughly 6 acres).  If something was smaller than this standard it was typically lumped in with the other surrounding classification.

 

This was a time challenging map.  I utilized a lot of time trying to get as much detail as I could into the islands in the marshland area.  I could have cut corners, but in the end, I think it looks better and presents the data in a more professional manner.



Lab 2 - Visual Interpretation

GIS 4035

 

 

 

In this weeks lab and lecture we learned about identifying tone and texture in an aerial photograph.  Above you will see areas that are defined by their tone.  Tones are defined by the brightness or darkness an area displays on an aerial image.  This varies from a very light tone, all the way up to the very dark tones.  Also in the above aerial, I show different textures.  Texture is how rough or smooth a surface looks when viewing the aerial photo.  I have shown a range from very fine texture (water) all the way up to a very rough course texture (housing development).

 

 

 

 

The map above identifies features based on shape/size, shadows, patterns, or associations.  Shape and Size of features is a common way that analysts look at aerial photos and quickly determine what they are looking at.  Roadways, Buildings, Water bodies can all be identified in the photo by the way they look and the size they are relative to their surroundings.  Shadows help show the viewer what may not be clear with an initial viewing of the photograph.  Take for instance the water tower, sign and Utility Pole, all of which can easily be identified on the ground but from the air, may be more difficult.  Shadows help the analyst make a better determination of what is on the ground.  Pattern is also used by analysts to determine what they may be looking at.  Housing developments, Vegetation, and Parking Lots can be looked at and by a pattern show the user what they are looking at.  Finally, through association we can make a determination of what we are looking at.  The end of a major water body and the pier makes it easy to associate a beach to the coastline.  While a pool and building shape, with cars and a parking lot helps the analyst determine a hotel is present.