Lab 8: Spectral Signature Analysis & Resource Monitoring

Goal 

The main goal of this lab is to gain experience on the measurement and interpretation of spectral signatures of various Earth surface and near surface materials captured by satellite images and also perform basic monitoring of Earth resources using remote sensing band ratio techniques. Multiple spectral signatures will be collected from an image and analyzed.  Techniques for monitoring soil and vegetation health will also be explored.

Methods

Collecting Spectral Signatures from a Satellite Image

The Landsat ETM+ image eau_claire_2000.img was brought into ERDAS Imagine and 12 measurements of spectral reflectance were taken.

1. Standing Water
2. Moving water
3. Forest
4. Riparian vegetation.
5. Crops
6. Urban Grass
7. Dry soil (uncultivated)
8. Moist soil (uncultivated)
9. Rock
10. Asphalt highway
11. Airport runway
12. Concrete surface (bridge, parking lot, or any type of concrete surface)

To do this the polygon draw tool was used to create a polygon in the area of interest.  The Raster - Signature Editor tool was used to create a list of the 12 signatures.  The signatures were then plotted on a signature mean plot to show the reflectance in each of the 6 bands.

Figure 1

Multiple spectral signatures could be put on one mean plot to compare and analyze the signatures.

Figure 2

Resource Monitoring

To monitor vegetation health a simple band ratio was performed by implementing the normalized difference vegetation index (NDVI) on an image of the Eau Claire and Chippewa area. To do this the Raster - Unsupervised - NDVI tool was used.  The satellite was specified and NDVI function was selected.  The tool then created a separate NDVI image. This image was imported into ArcMap and the data was classified into five groups to create a map of vegetation health.

To monitor ferrous minerals, a similar process was used as the vegetation monitoring but this time a different band ratio was used.  The Raster - Unsupervised - Indices tool was used this time.  The satellite and function were selected and an image was created.  This image was imported into ArcMap and classified into 5 classes to create a map of Ferrous Minerals in the Eau Claire, Chippewa area.

Results

The signature mean plot with all signatures included allows for analysis between the different materials on earth (Figure 2).  

After the NDVI image was created, a cartographically pleasing map was made in ArcMap to show areas that have healthy, to no, vegetation. A map like this could be useful in many studies. 

  

A similar process was used to create a map of ferrous minerals in Eau Claire. Ferrous minerals are present in many common metal used for construction.  The white areas on the map below are mostly populated areas.

Conclusion

Having the ability to extract spectral signatures of different materials is a useful skill that has many real world applications.  It is a simple process that can save time and money.  Instead of having to send multiple people out into the field with expensive equipment, a simple analysis can be done in house and accurate results can be found.  Monitoring techniques of vegetation and soil done on a computer can also save time and money.  It can raise questions about an area or answer questions.

Sources

Satellite image is from Earth Resources Observation and Science Center, United States Geological Survey

Lab 7: Photogrammetry

Goal:

The main objective of this lab was to develop skills of performing photogrammetric tasks on aerial photographs and satellite images. More specifically, this lab was to help understand the math behind the calculation of photographic scales, measurement of areas and perimeters of features, calculating relief displacement, and introduce stereoscopy and performing orthorectification on satellite images.

Methods

The first section of the lab was to calculate scale of a photo when measuring the distance between two objects on the screen and using that distance and the distance in real life to come up with a scale.  This resulted in a scale of 1/39,393.  The distance measured was from point A to point B.

The next thing done was to measure perimeters and areas from a image.  The measure perimeters and areas tool was used in Erdas to measure the area and perimeter of the lake in the middle left portion of the image above.  The pond has a area of 94.5 Acres and a perimeter of 2.54 Miles.

The next section was to calculate relief displacement from object height from an image. The camera height and scale of the image were known.  The relief displacement was 4.58 in.

The next part of the lab was to generate a 3D image using an elevation model.  To do this the Terrain-Anaglyph tool was used in Erdas to combine a digital elevation model with an image of the study area with a 1 meter spatial resolution. This image below is 3d when viewed with Polaroid glasses.  

This same process was done with a digital surface model.  The result is seen below. 

The last part of the lab was to orthorectify images and create a planimetrically true othoimage. Erdas Imagine Lecia Photogrammetric Suits (LPS) was used to perform this task.  A new block file was created in LPS and certain parameters were set.  This geometric model interface was set to polynomial-bases pushbroom and the geometric model catagory was set to SPOT pushbroom.  After this the coordinate system and projection type were chosen along with other parameters.  This puts the images that are going to be used in the process to the same coordinate system and projection.  

The next step was to bring in satellite imagery of Palm Spring, CA (spot_pan.img) and add it to the block in LPS.  No changes were done to the image but the verification process was completed to specify the sensor.  Next, ground control points were collected using the Classic Point Measurement Tool in LPS using a reference image.  The corresponding points were collected on the spot_pan.img using the point measurement tool.  Using the automatic drive function, the remaining GCP's were collected. The last two points were collected from a different image with a different horizontal reference source.  Elevation information was collected from a DEM file using the Reset Vertical Reference Source button in the point measurement tool palette. 

The next step was to bring in another image (spot_panb) and GCP's were collected for this image.  The GCP's for this image were collected using the already rectified spot_pan image.  Tie points were then collected for both images using Automatic Tie Point Generation Properties in the Point Measurement Tool Palette.  Parameters were set and the accuracy of the tie points were confirmed to make sure the image was properly orthorectified. 

LPS project manager was then used to perform Triangulation.  Parameters were set and the resulting text file was saved.

The final orthorectified image was created by using the Start Ortho Resampling Process in LPS.  Parameters were set and images were created.  The images were then brought in and overlapped to compare the spatial accuracy of the images.  Seen Below.

Results:

Anaglyph using DEM

Anaglyph using DSM

Orthorectifed Images

Conclusion:

When using satellite images, scale, perimeter, area, and relief displacement can be calculated and used for further analysis.  Anaglyphs can be created from elevation and surface models to make 3D models that would be useful in many situations.  Lastly, Erdas makes orthorectifying images relatively simple.  This makes a planimetrically correct image that has a constant scale that represents objects where they truly are. 

Sources:

Hexagon Geospatial. (2009). Erdas Imagine [computer software]. Georgia: Norcross. 

United States Department of Agriculture. (2005). [Satellite images in img. format]. National Agriculutre Imagery Program. Retrieved from: https://gdg.sc.egov.usda.gov/.

United States Department of Agriculture Natural Resources Conservation Service. (2010). [Digital elevation model for Eau Claire in dbf. format]. Retrieved from: http://www.nrcs.usda.gov/wps/portal/nrcs/site/national/home/.

Lab 6: Geometric Correction

Goal and Background: 

The goal of this lab is to display the technique and skills to geometrically correct a satellite image.  Two types of geometric correction will be displayed, image-to-map and image-to-image.     Geometric correction is necessary because an image may be distorted or not collected in its proper planimetric position.  This results in an image that is skewed or objects appear different than they actually are.

Methods: 

In order to geometrically correct an image there must be a reference image that is already accurate.  The first step is to bring the correct reference image and the image to be corrected into Erdas.  Once both images are in Erdas in seperate viewers the process is ready to begin.  To correct the image the user must add control points to both images.  This is done by using the control points tool under the multispectral raster processing tools.  A number of windows then pop up to set the parameters.  the reference image must be selected again and the order of polynomial must be selected.  In the image-to-map example, first order polynomial was selected.  This means only three ground control points are necessary to correct the image.  When the parameters are set, a new window will open and the user is ready to add control points. Control points 1-3 must be manually added to both images in the same exact spot.  After 3 are added it will say "model solution is current" and control points will automatically be added on the second image when inserted in the first.  After all the points are added, some fine tuning of the points is required to lower the root mean square error (RMS error).  The lower the RMS value the more accurate the image will be.
For the image-to-image geometric correction, the same process was used but this time was performed using a third order polynomial.  This mean ten control points are necessary to correct the image.

Results:

Image-to-Map Geometric Correction Control Points

When adding control points it is important to zoom in when adding each point to match the exact pixel that the point is being placed in both images. The Image on the left is the image that is being corrected and it was corrected to within .4 of a pixel to the reference map.

Image-to-Image Geometric Correction Control Points

A total of 12 ground control points were placed in the above photo.  10 GCP's are necessary in order to correct a third order polynomial correction, but 12 were added to ensure even more accuracy.  The RMS error value in the bottom right of the image is .001 and is extremely accurate but not perfect.  

Conclusion:

It is necessary to geometrically correct an image before using it for analysis.  Erdas makes this process user friendly and easy for everyone to do in a timely fashion.  The resulting image will then be ready to take place in research or further analysis.  This is a simple tool that should be used every time a new image is being used in a workplace.  

Sources:

Satellite images

Earth Resources Observation and Science Center, United States Geological Survey

Digital raster graphic (DRG)

Illinois Geospatial Data Clearing House

Lab 5: LiDAR Remote Sensing

Goals and Background:

The goal of this lab is to be introduced to working with LiDAR data structure and processing.  Surface and terrain models along with an intensity image will be created from LiDAR point clouds.  Working with LiDAR data and the LAS file format is important because it is a growing field in the industry and is an important skill set to have.  

Methods:

To begin working with LiDAR data, a LAS dataset was created in Arccatalog from a file of point cloud data.  This dataset was then imported into Arcmap.  Using the metadata from the point cloud data the LAS dataset was assigned a horizontal and a vertical coordinate system.  Using the LAS toolbar in Arcmap, filters, and other options can be manipulated to view the data in different ways.  Elevation, aspect, slope and contour are all options that the data can be viewed in.  An interesting tool that is available in Arcmap is the LAS dataset profile view tool.  This allows the user to select a area of interest and view it like they were standing on the ground next to it.  This example is looking at a bridge. 

LAS Dataset Profile View Tool

Next a Digital Surface Model and a Digital Terrain Model were created from point clouds.  To do this the LAS dataset was converted into a raster in Arcmap by using the LAS dataset to raster tool and using the maximum value as the input value.  This raster image was then turned into a digital surface model by running it through the hillshade tool.  This created a surface model that includes buildings and vegetation. 

Digital Surface Model

To create the digital terrain model the same thing was done with the point cloud data except this time minimum was used as the input value to make the raster image.  This made the image use the data that was closest to the surface of the earth. This digital terrain model is of the ground in the area of interest. 

Digital Terrain Model

The last image that was created from the LiDAR data was an intensity image.  This was done using the same LAS dataset to raster tool but this time using intensity as the value.  This image is like a optical image but without the visible bands. 

Intensity Image

Results:

 The resulting images that were created can be used in a variety of different ways.  The digital terrain and surface model can be overlaid on each other and a swipe tool can be used to look at them at the same time.

Swipe tool in Arcmap

This can be useful if the effects of water drainage and water flow wanted to be examined.  The user could look at the terrain model and determine where the water would flow without any vegetation or buildings.  The surface model could then be looked at to see how it would influence the water after the general flow from the terrain was already determined.  This could be very useful when trying to plan emergency water drains for the city.  

Sources:

Eau Claire County. (2013).

Price, M. (2014). Mastering ArcGIS 6th Edition. Mastering ArcGIS 6th Edition Dataset [shapefile]. New York: McGraw Hill

Lab 4

Goal and Background:

The goal of this lab is to show how to delineate a study area from a larger satellite image, demonstrate how spatial resolution of images can be optimized for interpretation, use radiometric enhancement techniques in optical images, linking a satellite image to Google Earth, explore various methods of resampling satellite images, image mosaicking, and use simple geographical modeling to detect binary change.

Methods:

Image Subsetting:

Using ERDAS Imagine 2016, subsetting an image is simple.  After the image is brought in, simply right click, use the inquire box tool, set the box over the desired area of the image, then use the subset and chip set of tools to create subset image.  This opens a new window that lets the user save the subsetted image to a new file.  Simply click "from inquire box" and okay and a new image is created from what subset. If a square is not the right shape for the area that is being subset, a shape file can be imported and used as the area of interest.


Image Fusion:

This can be used to pan sharpen images and give them higher resolution.  To pan sharpen the resolution merge function can be utilized to combine a reflective band image with a panchromatic band image to create a higher resolution image.  Both images are input to the function, the user names the output file and selects the method and resampling technique and then runs the function.  The resulting image will have a better resolution.


Simple Radiometric Enhancement Techniques:

Haze reduction helps an image look crisper and makes it easier to view. Using the haze reduction tool under the raster settings, the user inputs the image that they want the haze to be reduces, name an output file name and click run and the tool reduces the haze in the image.

Linking Image Viewer to Google Earth:

Linking the ERDAS viewer to google earth is a good way to get another view of the image being examined.  ERDAS has a google earth function that makes this very easy.  All the user has to do is click the Google Earth tab at the top and click connect to Google Earth.  This will open a new window and then can be open side by side.  ERDAS and Google Earth can be matched and linked by clicking, match GE to view and then syncing them together.  Google Earth has very high resolution and labels making it useful to identifying objects being examined.

Resampling:

Resampling refers to changing the size of pixels in the image.  This can be useful if the user wants to run a function where two images need the same pixel size.  To do this the user goes to spatial under the raster tab, then clicks on the resample pixel size tool.  This opens up a new window where the input and output and new pixel size can be entered.  The user then clicks ok and the program is ran, changing the pixel size.

Image Mosaicking:

Image mosaicking is used when the area of interest cannot be contained in one satellite image.  Mosaicking is used to bring together two separate pictures into one image.  Mosaic pro is a program in ERDAS that makes mosaicking easy and does the job well.

Binary Change Detection:

Binary change detection is used to determine if two images have changed.  This is useful when comparing an area from 1991 to the same area in 2011.  In this section of the lab the model maker function was utilized to create an image from two images and determine what changed in that time.  The new image was then overlaid on a whole image in ArcMap and the changes in the images were then clearly visible.




Results:

Subset Image

Pansharpened Image

Google Earth Sync View

Mosaicked Image

Binary Change Detection Overlay