Wednesday, October 18, 2017

Spatial Enhancements and Radiometric Correction

Module 6 looks at enhancing imagery between ERDAS Imagine and ArcMap.  These are used in conjunction with one another and provide a wide array of possibilities when it comes to enhancing  or correcting aerial and satellite images.  My finished map is seen below:



For this image I first used the Fourier Analysis in ERDAS Imagine, used the transformation editor with wedge tool and low-pass tool and subsequently used the Sharpen filter and Low pass 5x5 filter.  The image started out with significant striping and the preceeding tools and filters allowed for the striping effect to be minimized.  The Fourier Analysis was my favorite part albeit the tricky part.  It was difficult to position the wedge and low-pass tools correctly.  Once all these were in place the image was then opened in ArcMap for final editing.  The end result still has stripes but the image is distinguishable.  Using ERDAS Imagine, particularly in conjunction with ArcMap, shows the vastness of options when editing remote sensing images.  The level of detail is easily dictated and the desired result is always possible with such a plethora of enhancements and corrections. 

Friday, October 13, 2017

MTR - Report Week

The conclusion of the Mountaintop Removal project is seen this week in "report week" and includes a number of external links to further support the study. The final group map is seen below:




As a group we analyzed four landsat images and compiled them into the map shown above.   The areas of Mountaintop Removal are seen along with the major rivers.  I chose to include the major rivers because by living in this region and seeing this mining process daily I am reminded constantly of the harm done to the rivers and those downstream as well as the environment.  To complete this map, my data was initially problematic but after reworking the issues were resolved and we were able to obtain the composite image.  Many tools were used in order to accurately depict the areas of Mountaintop Removal and to erase areas of interference such as river beds and roads.  The use of ERDAS Imagine was also helpful in categorizing the image with areas of Mountaintop Removal and areas that are not.  Despite setbacks and a highly complex study, the lab has taught me how valuable this analysis can be to the communities affected, the entire region and the environment.  The following are additional resources accomplished throughout this project. 

Link to group web map:  http://arcg.is/XjSjG

Link to story journal:  http://pns.maps.arcgis.com/apps/MapJournal/index.html?appid=a4076329addd44c084fcc8b6b71e8838
(or, http://arcg.is/1q0Gyn)

Link to story map: http://arcg.is/2xmpU9T

Tuesday, October 3, 2017

Intro to ERDAS Imagine and Digital Data

The first part of Module 5 begins to look at ERDAS Imagine and its use in conjunction with ArcMap.  The finished product, seen below, began in ERDAS Imagine and was then brought into ArcMap for final editing.  This lab also began looking at Electromagnetic calculations and how those translate into interpreting images/digital data. 




The map shows a section of Washington State and the various types of features are shown in representative colors.  The image was first opened in ERDAS Imagine and various tools were used in order to gain experience in the software.  It is definitely different from ArcMap and other applications.  It took some getting used to to find tools, I found it particularly important to check the group labels in the toolbars in order to find some of the tools.   For the legend (done in ArcMap) I chose to edit the description of each feature in the layer properties and symbology tab.  This allowed for the inclusion of the units (square miles) in the legend and therefore gives the legend more value.  The most difficult part of ERDAS Imagine was determining how to save the work.  Work can be saved as a session or in different types of layer files. 

Although ERDAS Imagine was a completely different type of program than anything I've used previously, I enjoyed how this software allows for in depth interpretation and analysis of imagery. 

Saturday, September 30, 2017

Mountaintop Removal - Analyze Week

This week continues to look at and analyze the practice of Mountaintop Removal in the Appalachian Mountains of the Southeastern United States.  As stated previously, this is precisely where I live and have grown up and therefore the issue is an incredibly important one to me. 

The lab worked with the Landsat images using the Composite Bands tool and the Extract By Mask tool to prepare the image for work in ERDAS IMAGINE software.  Once in the IMAGINE software, the unsupervised classification tool was used to classify each pixel as either mountaintop removal (MTR) or non-mountaintop removal (NMTR).  The image was then sent back to ArcMap to complete the reclassification.  I have included screenshots below.  I was unable to successfully run the reclassification tool (was unable to use the "class_names" as the reclass field") and therefore will be doing further work in this lab.  The first image is the reclassification image I did obtain using "value" in the reclass field and the second image was prior to the reclassification, showing the end result of classifying pixels in ERDAS IMAGINE:



Tuesday, September 26, 2017

Truthing - Land Use and Classification

Module 4 continued on the concepts of Module 3.  The land classifications determined in Module 3 are here further analyzed.  Seen below is my final product:



I used sampling protocol using the codes and locations.  Thirty random points were selected to include areas of each land use classification and also to include all areas of the study area.  Once the points were converted to graphics and the attribute table edited Google Maps and its "Street View" were used to verify each of the determined land classifications.  Using "street view" allows the user to see first hand the area and make a more accurate classification or determination.  Here the wetlands were the most difficult to accurately classify simply because there is no "street view" available.  Residential and Commercial sites were the most accurately depicted and I came up with a 57% true rate on my classifications from module 3.  I was unable to get the "true" results to display as green despite using the select by attributes and export data functions.  The symbology tab in the properties window was showing correct; however, that did not carry over to the layout view.  The new codes on the "false" classifications are seen in yellow and the original codes are in black.

I enjoyed using the Google Maps Street View, this makes the lab "real" as you look at the actual addresses that you have selected.  The lab made me realize how complex land use and classification can be.

Friday, September 22, 2017

MTR Prepare

This week begins Module 2 and looking at Mountaintop Removal.  This issue hits (literally) home for me because it is such a large factor in the area in which I grew up and still reside.  I've seen the ill effects first hand and therefore further analysis of this practice is important to me and my home.  First, a basemap of the study area was constructed, as seen below.



The study area is shown in basic form and streams are illustrated.  This was accomplished through the use of the hydrology toolset in ArcMap.  After various tools were implemented (ex: fill, flow direction, flow accumulation, con, basin, stream to feature and raster to polygon) the resulting image shows the significant streams that will be used in this analysis.

The next portion of the lab involved creating a story tour through GIS online.  The link to my story map is here: http://arcg.is/2xmpU9T .
The story map outlines the mountaintop removal process through pictures and visual aids.  Finally a story map journal was started.  This is a work in progress over the next couple of weeks as myself and others analyze the study area.  The link to this journal is here: http://arcg.is/1q0Gyn .
As this is a work in progress, several placeholders are used to represent the final product.

I enjoyed this lab and particularly the online sections and story maps as they are intuitive and make for getting a point across in a simple fashion.

Tuesday, September 19, 2017

Land Use and Classification

The lab this week takes a look at land use and land classification.  The lab was done entirely in ArcMap using an aerial image only.  The object was to identify and classify all elements of the photo. My finished classifications and land usage for this photo are seen in the result below:



The classification was achieved by studying the level II listing of landscape characteristics and applying them to the aerial photograph.  Once each element was classified, polygons were drawn in editing sessions and converted to graphics which were then given a code to identify the type of classification and these were also given a code description.  I enjoyed learning about the levels of classification and seeing how practical this can be in every day use.  My map above shows areas of forest, wetland, water, commercial, industrial and residential.  The areas not shaded represent the residential areas.  I determined that the commercial areas contain larger buildings and parking lots, close to transportation networks and residential areas.  Industrial, I determined, are areas of clustered commercial buildings generally near to a natural resource.  Forest areas show a large amount of dense trees and wetlands contain sparse trees with minor water features.  Obviously the water areas are larger areas of hydrology.

I enjoyed this lab although it did get repetitive; however, this aided in reinforcing the concepts and establishing how these methods can be used most practically in every day work.