GIS 4043

Unsupervised and Supervised Classification

 For this week's lab, we were tasked with classifying a satellite image using unsupervised and supervised classification methods in ERDAS Imagine. For the supervised classification final portion of the lab, I created spectral signatures for multiple classes in Germantown, Maryland utilizing the Region Growing Properties tool, Polygon tool, and the Signature Editor tool. Once I created Areas of Interests (AOIs) for each class, the histograms and mean plots of all the signatures were examined for spectral confusion. My conclusion was that bands 4, 5, and 6 were the most separate and least confused. These bands were then utilized for the supervised classification using Maximum Likelihood. I recoded the supervised image to merge the AOIs and calculated the area in square miles for each of the classes. Below is the final result of supervised classification. 

GIS4035 Module 4

 This weeks lab focused on Spatial Enhancement, Multispectral Data, and Band Indices

We downloaded and imported satellite images and performed spatial enhancements in both ERDAS and ArcGIS Pro. We explored the Histograms, utilized various tools like Convolution, Focal Statistics, created spectral band indices and utilized various filters. 

In the last portion of the lab we were tasked with using all the previous skills/methods outlined in the prior exercises to find 3 features in the provided image and create a map layout for each one. Below are the results:

Searched the image for a feature that produced a spike between pixel values of 12 to 18 in Layer_4. The feature was water make and for the best visibility I chose Short Wave Infrared Color Composite band combination where dark blue is water, green is vegetation and pink is bare soil.

For the second map we looked for a feature that represents a small spike in layers 1-4 around pixel value 200 and a large spike between pixel values 9 and 11 in layer 5 and layer 6. This feature was snow and True Color band was chosen. 

Last, we searched the provided image for areas of water where layers1-3 become much brighter than normal, layer 4 becomes somewhat brighter than normal and layers 5-6 remain unchanged. The above water transition area was located and Near Infrared was chosen to best highlight the feature

GIS 4035

 Lab 3 Intro to ERDAS Imagine and Digital Data Lab

This week se stepped outside of the ArcGIS Pro box for a moment and were introduces to ERDAS Imagine.  This is a remote sensing software with a wide range of tools for imagery and geospatial analysis. We opened a provided image into an ERDAS Imagine viewer and explored the basics for zooming, adding, images, preferences and various settings. We altered the bands to a TM False Natural Color band combination and explored the difference with other color settings. Next, we learned how to open the attribute table in ERDAS and add a field.  For our purposes it was a new area field which we would use once we opened the image in an ArcGIS map project. We used the Inquire Box tool in ERDAS to select one section of the image to focus on. In our map project we added the image we worked with in ERDAS and changed the symbology. We were then asked to created a map layout highlighting land cover types and the total area of each one.  Below is my map layout:

It was nice to be introduced to software outside of ArcGIS. ERDAS Imagine has a wide range of capabilities and we barely scratched the surface. I didn't find it as user friendly as ArcGIS Pro but I'm sure that is a novice reaction.  

Part b of our lab involved further examination into some additional ERDAS tools and functions. We touched on the four different types of resolution in ERDAS, and analyzing, and interpreting thematic rasters. No new map layout was required for this section.

GIS 4043 Land Use/Land Cover Classification

 Module 2 Lab

This week we were introduced to Land Use/Land Cover Classification. We were given an aerial photograph and asked to digitize an area which was located in Pascagoula, MS. We created polygons using a Level II classification system. Once that was completed we selected 30 points to use for ground truthing to check the accuracy of the mapping classification. Below is the map generated from my results from the lab:

This exercise gave me a chance to practice many map making skills that I have not used recently. Making polygons was a little challenging but we did use new tools, auto complete for polygons and also the clip tool when smaller polygons were created within larger ones.