Masthead

Lab 7 : Image Processing I

Introduction

Arcata 1956

This lab covers how to download aerial images and other basic image processing techniques, specifically pansharpening datasets and georeferencing aerial images. Part I of this lab will show you how to obtain recent Landsat imagery from EarthExplorer and pansharpen the image to produce a higher spatial resolution dataset. The second exercise goes over how to search and obtain historical aerial imagery and how to georeference this imagery in ArcGIS Pro. EarthExplorer has a great deal of aerial photography available for download, but the majority is not georeferenced.

In this lab you have your choice of completing one of the two activities. Each activity is worth 15 points. If you complete both activities you will receive 15 extra points.

Option 1: Obtaining and Pan-sharpened Landsat Imagery

Option 2: Obtaining and Georeferencing Aerial Photographs

You will digitize the extent of development in the greater Las Vegas area in both of these options and calculate the area in acres.

Learning Outcomes

Set up your Workspace

First we need to set up our workspace.

  1. Create a folder on the Desktop named “Lab_7”.
    1. Within the folder named “Lab_7”, create two subfolders as below: Folder Structure (within Lab_7 folder):
      1. Originals
      2. Final
Continue on to Option 1 or Option 2.

Option 1: Obtain Landsat Imagery, Pan-sharpen

Acquire and Download Data

  1. Open up your favorite web browsers and navigate to http://earthexplorer.usgs.gov and log into to your EarthExplorer account.
  2. Now we are ready to select a location to search for data sets. Search for Las Vegas, Nevada and select Las Vegas, NV, USA or use the map to navigate to Las Vegas and drop a marker.
  3. For this part of the exercise we will be searching for Landsat imagery. We will be looking for recent imagery, so you can leave the date range blank, this will return the most recent images first. Click the Cloud Cover tab and set to Cloud Cover Range: to 0% to 10%. This will only return imagery with less than 10% cloud cover.
  4. The next step is to select the type of data you are interested in. Click on the “Data Sets” button to proceed to the next screen and navigate to the “Landsat” heading and click the “+” beside the name to expand the category. You will now see all of the available Landsat imagery options. Click the box beside "Landsat Collection 2 Level-1". Check the Landsat 8-9 OLI/TIRS box. This will search for recent Landsat 8 or Landsat 9 level-1 images. Note that you must download Level-1 data if you are planning on performing pansharpening operations.
  5. Click the "Additional Criteria" tab and select "Day" under the Day/Night Indicator. Click “Results” and you should see thumbnails showing the data sets that match your search criteria.
  6. Click the foot icon to see the footprint of each scene in the Google Map interface, click the Show Metadata and Browse icon to see a preview of the image and the metadata.
  7. Select a recent, cloud free image and download the Level-1 data product bundle. Click the Download Data Icon and select “Landsat Collection 2 Level 1 Product Bundle”. A new window will pop-up, in this window select the top button, Landsat Collection 2 Level 1 Product Bundle. It may take a few minutes to download the file.
  8. Locate the file (.tar) you just downloaded and move it or extract the files directly into your “Originals” folder. Use 7-Zip to extract the files (See Lab 6).

Create Region of Interest and Pansharpen Landsat Image

Pan-sharpening algorithms are used to sharpen multispectral data using high spatial resolution panchromatic data. Landsat 8 and 9 data includes a 15 meter spatial resolution pan-chromatic band (Band 8).

  1. Start “ENVI” and open the Landsat 8/9 image you downloaded, by selecting the metadata text file ending in "_MTL.txt". You should now see the Landsat image featuring Las Vegas, Lake Mead and the surrounding landscape.
  2. Create a new ROI by right clicking on the mulispectral Landsat image and selecting New Region of Interest. Rename the ROI "LasVegas".
  3. Create a large rectangular ROI that encompasses the greater Las Vegas area (this includes Henderson, North Las Vegas etc). Use the reference map link tool in ENVI to verify that you are focused on Las Vegas region. We will use this ROI to subset the data when we pansharpen.
  4. From the Toolbox, select Image Sharpening → Color Normalized (Brovey) Sharpening. This sharpening methods only sharpens the red, green and blue bands.
  5. In the first window select the associated red, green and blue bands from the Landsat imagery and click OK.
  6. In the second window select the High Resolution file. Select the panchromatic band from the Landsat image.Pansharpened
  7. In the same window click the Spatial Subset button. In the Spatial Subset window click the ROI/EVF button under the Subset Usingheading. Select the LasVegas ROI you previously created. Click OK after selecting the ROI and click OK one more time to proceed to the final step.
  8. In the final parameters windows save the output file as pansharpened.dat in your Final folder. Click OK to start the process.
  9. Be patient, the pan sharpening process can take awhile depending. Using an ROI to subset the images as part of the workflow can speed up processing time. You should see a blue bar that indicates the progress of the process. Once complete the pan sharpened image should appear in your viewer. Use the Flicker, Blend and Swipe tools to compare the resolution of the sharpened image to the original Landsat image.
  10. Close ENVI when you are done comparing the imagery.

ArcGIS Pro: Create a Shapefile and Digitize Development

The next thing we are going to create a shapefile showing the current extent of developed land in the greater Las Vegas area and create a map displaying the results. The greater Las Vegas area includes Las Vegas, North Las Vegas and many of the surrounding communities (use your best judgment on which areas to include).

  1. Start ArcGIS Pro and create a new map project. Add the subsetted pansharpened Landsat image of Las Vegas and remove the basemap layers.
  2. Open the Catalog pane and navigate to your Final folder. Right click on the folder and select File → New → Shapefile. In ArcGIS Pro this launches the Create Feature Class tool. This tool can also be launched from the Geoprocessing Tool Pane (Data Management Tools → Feature Class → Create Feature Class).
  3. Name your feature class (shapefile) "current_area" and select polygon as the geometry type. Set the coordinate system to match the pansharpened.dat Landsat image. Then click Run to create the new shapefile.Create FEature Class
  4. The new shapefile you just created should now appear in the Table of Contents. Note that while you will see the layer in the Contents you won't see anything on your map until you start digitizing.
  5. Change the symbology of the shapefile so there is only an outline (make it high contrast color) and no fill. Close or hide the Symbology pane when you are done.
  6. Select the new shapefile in the Contents pane and in the main tool bar Edit tab, click Features → Create. The Create Features pane will pop up.Editor Window
  7. In the “Create Features” pane, select your shapefile icon ("current_area") and the “Construction Tools” window will appear below, the Polygon tool should be selected.Create Features
  8. Begin digitizing the developed area by clicking the map to insert a vertex. Digitize the extent of current development in the greater Las Vegas area by creating a polygon. Your polygon should be fairly detailed, as the goal it to obtain an accurate estimate of the extent of developed area. When you are done, right-click and select "Finish Sketch".
  9. In the Edit menu click the Save icon to save the edits. You can also start editing again if you want to adjust your shapefile, just remember to save any changes once you are done editing the feature.Save Edits
  10. Now we will calculate the area of the polygon. From the Geoprocessing pane search for the Calculate Geometry Attributes tool.
  11. In the Calculate Geometry Attributes tool select the shapefile as the Input Features. Under Geometry Attribute type "area" and select Area from the dropdown next to the field. This will create a new field in the attribute table. Select US Survey Acres for the Area Unit. You can leave the Coordinate System blank in this case - this will use the coordinate system of the spaefile to calculate the area.
  12. Right click on the shapefile and select Attribute Table. The attribute table for the file should appear in a new pane. Look for the "area" field, in the field will be the area that was calculated in the previous step.
  13. Add a layout to your project (Insert>Layout). Create a map (any size) that features your pan-sharpened Landsat image with the shapefile you created showing the extent of development, a scale bar, title, north arrow, data source, date of imagery and map author. Also include the developed area (in acres) that you calculated above. You may include this in a legend, title or with text. See Lab 5 and Lab 6 for references on how to create a map layout in ArcGIS Pro and see Canvas assignment page for an example map. TIPS
    • To create a map frame exactly the dimensions of your data, bring up the Map Frame Properties > Display Options. Under Constraint select the layer that you want the map to be focused on (either the shapefile or the pansharpened satellite image). This will automatically set the map frame size to match the data.
  14. Export your completed Map as a PDF file (minimum 300 dpi resolution) and upload to Canvas.

Option 2: Obtain and Georeference Aerial Imagery

Download Data

  1. Open up your favorite web browsers and navigate to http://earthexplorer.usgs.gov and log into to you EarthExplorer account.
  2. Now we are ready to select a location to search for data sets. We will be searching using known coordinates. Click the Add Coordinate button, the below window will appear.
  3. Add Coordinates
    Coordinate Latitude Longitude
    Coordinate 1 36°09'14"N 115°09'26"W
  4. Using the information in the table to the right to enter the latitude and longitude for the coordinate and click Add.
  5. When you are finished you should have a coordinate on your map marking Las Vegas, Nevada.
  6. For this part of the exercise we will be searching for historical aerial photography. Specify a date range of 01/01/1952 to 12/31/1952.
  7. The next step is to select the type of data you are interested in. Click on the “Data Sets” button to proceed to the next screen.
  8. Navigate to the “Aerial Imagery” heading and click the “+” beside the name to expand the category. You will now see all of the available aerial imagery options. Click the box beside "Aerial Photo Single Frames". This will only return a variety of historical aerial images.
  9. Click “Results” and you should see thumbnails showing the data sets that match your search criteria.
  10. Click the foot icon to see the footprint of each scene in the Google Map interface. Look for the image the best covers the area of interest.
  11. This should be entity ID: ARB031250020006.
  12. Click on the "Show Metadata and Browse" icon to bring up the window with the metadata. Notice there are coordinates showing the four corners of the image and center. These can sometimes be useful if there are no landmarks to use as control points.
  13. Click the download icon to download the image, and select "High Resolution Product".
  14. Locate the files you just downloaded and move them by right clicking on the files and copying them and pasting the files into your “Originals” folder with within Lab 10 folder.
  15. In your working folder, right-click on the file "B031250020006.tif.gz" and select 7-Zip → Extract here. The file has now been extracted and is ready to use.

Georeference the Image in ArcGIS Pro

 

  1. Start ArcGIS Pro and create a new map project. Add the aerial image to be georeferenced, "B031250020006.tif", click yes if you are asked to generate pyramid layers. The image won't be in the correct geographic location since we haven't georeferenced the image yet.
  2. In the Contents pane, click the on the aerial image that you want to georeference. Click the Imagery tab and click Georeference to open the Georeference tab.
  3. In the Prepare group of the Georeferencing tab, click Set SRS (Spatial Reference System) SRS. This will set the spatial reference system that will be assigned to the georeferenced raster. Set the coordinate system to Projected Coordinate System → UTM → WGS 1984 → Northern Hemisphere → WGS 1984 UTM Zone 11N. Click OK.
  4. In the Map tab click the Basemap button and select Imagery. This will be our reference imagery we will use to georeference the aerial photo.
  5. Use the navigation tool to zoom and pan to Las Vegas on the map. Zoom in to approximately the area covered by the aerial photograph. Tip: Use the footprint from EarthExplorer as a guide to determine the approximate location and orientation of the aerial photograph.
  6. Click back to the Georeference tab. Click Fit To Display Fit to Display. Then use the Move, Scale Scale, and Rotate tools to place the photo in the approximate location.
  7. In the Adjust group, click the Add Control Points tool. To add a control point, first click a location on the aerial image you are georeferencing (the source layer); then click the same location on the target layer on the map (the imagery base map). Press the L key to turn the transparency of your source raster on or off. Look for unique landmarks or other easily identifieable features that can be located in both the aerial image and the basemap imagery.
  8. Continue adding control points, try to distribut the control points throughout the image. See the below diagram of well distributed control points. Add a minimum of 5 control points. As you add control points the image will automatically update. If you add control points too close together or an incorrect control point it can warp the image. If possible, try to add a control point in each corner of the image and one in the center of the image.
  9. Control Points
  10. In the Review group, click the Control Point Table button to open Control Point Table to evaluate the residual error for each control point. Press the L key to switch the transparency of your source raster on and off. Delete any unwanted control points using the Control Point table. Alternatively, edit the inaccurate points by selecting them and moving the vertices. When you're satisfied with the current alignment, stop entering control points.
  11. In the Save group, choose how you want to save your georeferencing information. First click Save, this saves the control points in the current session. Then click Save as New, this will save the georeferenced image to a new GeoTIFF file. The GeoTIFF will include the coordinate system and referencing information.
  12. The Export Raster pane will open on the side. Save as “LasVegas1952.tif” in your finals folder. Double check the coordinate system and format (should be UTM and “TIFF"), then hit the Export button.

Create a Shapefile and Digitize Development

The last thing we are going to create a shapefile showing the extent of development in Las Vegas in 1952.

  1. Open the Catalog pane and navigate to your Final folder. Right click on the folder and select File → New → Shapefile. In ArcGIS Pro this launches the Create Feature Class tool. This tool can also be launched from the Geoprocessing Tool Pane (Data Management Tools → Feature Class → Create Feature Class).
  2. Name your feature class (shapefile) "1952_area" and select polygon as the geometry type. Set the coordinate system to match the aerial photograph. Then click Run to create the new shapefile.Create FEature Class
  3. The new shapefile you just created should now appear in the Table of Contents. Note that while you will see the layer in the Contents you won't see anything on your map until you start digitizing.
  4. Change the symbology of the shapefile so there is only an outline (make it high contrast color) and no fill. Close or hide the Symbology pane when you are done.
  5. Select the new shapefile in the Contents pane and in the main tool bar Edit tab, click Features → Create. The Create Features pane will pop up.Editor Window
  6. In the “Create Features” pane, select your shapefile icon and the “Construction Tools” window will appear below, the Polygon tool should be selected.Create Features
  7. Begin digitizing the developed area in the historic aerial photograph by clicking the map to insert a vertex. Digitize the extent of development in the Las Vegas area by creating a polygon (or mulitple polygons if you would like). Your polygon should be fairly detailed, as the goal it to obtain an accurate estimate of the extent of developed area at this point in time. When you are done, right-click and select "Finish Sketch".
  8. In the Edit menu click the Save icon to save the edits. You can also start editing again if you want to adjust your shapefile, just remember to save any changes.
  9. Now we will calculate the area of the polygon. From the Geoprocessing pane search for the Calculate Geometry Attributes tool.
  10. In the Calculate Geometry Attributes tool select the shapefile as the Input Features. Under Geometry Attribute type "area" and select Area from the dropdown next to the field. This will create a new field in the attribute table. Select Acres for the Area Unit. You can leave the Coordinate System blank in this case - this will use the coordinate system of the shapefile to calculate the area.Calculate Area Attribute
  11. Add a layout to your project. Create a map (any size) that features your georeferenced historic aerial photograph with the shapefile you created showing the extent of development, a scale bar, title, north arrow, data source and map author. Also include the developed area in 1952 (in acres) that you calculated above. You may include this in a legend, title or with text.
  12. Export your completed Map as a PDF file (minimum 300 dpi resolution) and upload to Canvas.



Contact Info

Humboldt State University
1 Harpst Street Arcata, CA 95521
skh28@humboldt.edu

© Copyright 2020 HSU - All rights reserved.