Lab 4, Displaying and understanding a digital image

Geog 418/518, Fundamentals of Remote Sensing

 

Name 1                                                       Name 2                                            

 

Lab Goals

 

The goals of this lab are to:

 

a)      have you use some of the same software commands you learned in last week’s virtual tour of ERDAS Imagine;

b)      familiarize you with basic image information that you can find out about an image (e.g., pixel resolution, pixel values);

c)      begin thinking about the meaning of the image in terms of spectral reflectance, absorption, scattering and transmission;

d)      speculate on how remote sensing raster data might or might not be the best option available; and

e)      develop skills in being self sufficient in using remote sensing software.

 

With respect to the last goal, if you want to become a capable user of remote sensing imagery, it is critical that you develop the ability to search and find information or commands that you need without always having someone guide you.  In the following lab, some of the commands will be provided to you, but other commands or information are not provided.  If you have questions about how to do a certain task, I or the GTF will help, but we will always ask you first what approaches you have taken to trying to work it out.   I expect you to have tried searching the Help menu and to have tried some exploration on you own. 

 

General background information on remote sensing and data storage topics can be found in the ERDAS Imagine Field Guide.  Specific help on the software and which buttons to push to access certain pieces of information or commands can be found in the ERDAS Imagine Tour Guides.  Both guides can be accessed via the Help button on the ERDAS menu.

 

Lab Procedures:

 

I want you to work in pairs on this lab and turn in only one lab per team.  Many studies and my own personal experience have shown that working in pairs maximizes the amount of learning that occurs.  Even if you feel you understand it perfectly, the act of explaining it to your partner or comparing answers with your partner will improve your understanding and your ability to articulate your knowledge to another user – a key skill in remote sensing.  As a remote sensing professional, you will often find yourself in the position of having to explain the imagery and the analysis to novices.  This is a good starting point for acquiring those verbal presentation skills.

 

Your team is also encouraged to talk to other teams if you get stuck.  You must, however, fill in the answers with your own words.

 

Write answers in complete sentences, except where the question specifies otherwise.  Please be legible.

Lab Activities and Questions:

 

Go into the Lab 4 folder and open the Landsat Thematic Mapper image lanier.img from Georgia. 

 

  1. The lanier.img image has seven bands (“layers” in ERDAS) that cover the spectral ranges shown in the table below.  Based on what you have learned in class, label the name of the spectra covered by each band (be as precise as possible, e.g., state “red” rather than “visible”).  Check out the Field Guide if you cannot remember which spectral ranges correspond to different names (You will note that spectral ranges for certain IR band widths differ slightly from what I gave you in class.  This is typical – there is not a single widely accepted definition of the spectral range for these different IR bands.)

 

             

Band

Sensitivity (μm)

Name of this spectral range

Minimum value

Maximum value

Mode

value

1

0.45-0.52

 

 

 

 

2

0.52-0.60

 

 

 

 

3

0.63-0.69

 

 

 

 

4

0.76-0.90

 

 

 

 

5

1.55-1.75

 

 

 

 

6

10.4-12.5

 

 

 

 

7

2.08-2.35

 

 

 

 

 

  1. Fill in the minimum value, the maximum value, and the mode value for each band using the Utility/Layer Info command.

 

Note the way in which minimum values change from band 1 through band 7.

 

  1. Why do the minimum values from band 1 through band 7 (excluding band 6) vary in this way?  Be specific in describing which bands are most affected by certain processes.

 

 

 

 

 

 

 

 

 

 

  1. Why does band 6 not fit this pattern?  Specifically, why is its minimum value higher than those for bands 5 and 7.

 

 

 

 

  1. Considering the entire image, why do all values range between 0 and 255?

 

 

 

 

 

 

 

 

  1. If you were trying to detect subtle differences in reflectance from the surface, would you rather have higher radiometric resolution (e.g., 10 bit) or lower (e.g. 6 bit) resolution?  Why?

 

 

 

 

 

 

 

  1. How can you tell the image is a raster image rather than a vector image? (There are many possible ways to determine this).

 

 

 

 

 

 

 

 

  1. In terms of the precision of boundaries and point locations, what differences are there between a raster image and a vector or paper map?  Which is preferable for this purpose?

 

 

 

 

 

 

 

 

  1. In terms of showing subtle variations in landcover (e.g., variations within a forest), is a raster or a vector image preferable?  Why?

 

 

 

 

 

 

  1. What is the size of one pixel on the image?  What are two ways to determine this?

 

 

 

 

 

 

 

 

 

  1. What geographic coordinate system is the image projected onto? 

 

 

 

The image has a geographic coordinate system, which you determined in 11 above.  Every raster image also has a “file” coordinate system, which simply states how many pixels to the right or left, or up or down, one is within the image.  In other words, the file coordinate system tells you which row and column you are at within the image.  Knowing which way the columns and rows are numbered (e.g. do they get higher as one goes “up” the image, or as one goes “down” the image) is crucial to assigned geographic coordinates to the image.

 

  1. What is the 0,0 point (the origin) for the pixels (NOT for the geographic coordinate system) and in what direction do the pixel counts get higher?  For example, does the x file coordinate get higher as one moves up or down the image?  What about the y coordinate value for the pixels?

 

 

 

 

 

 

 

  1. How did you determine your answer for number 12?

 

 

 

 

 

 

 

Okay, okay – you have probably gotten the point that I am big on having you look at the metadata to try to understand the numbers behind the image on your screen.  Checking out these data can save you a lot of head ache when you start doing analysis with multiple images. 

 

Let’s now extract some different information from this feature and do some interpretation of that information. 

 

 

  1.  Use the software to determine the approximate distance (to the nearest 100 meters) along the major freeway cutting from SW to NE across the image?  How did you determine this?

 

 

 

 

 

 

 

  1. Use the Spectral Profile Tool to fill in the table below.  The numbers refer to the pixel file coordinates. 

 

When you bring up the Spectral Profile Tool, click on Spectral, then click on View and open up the Tabular data option.  This will show you the actual data values for the pixels while also keeping the graphical portrayal of spectral values.

 

To collect spectral reflectance values you must click on the cross hairs symbol, then click on the Viewer window.  You can then enter the pixel coordinates directly into the Spectral Profile Tool window.  Make sure it is set on “File” so that the x,y coordinates you give it work correctly.

 

Make sure to re-click the cross hair symbol each time you collect data for a new point.  Otherwise it will simply write over the last data.

 

Band

Pixel Reflectance Values

Lake (137,178)

Forest

(407, 208)

Field

(74,437)

Urban

(227,285)

Mystery

(455,44)

1

 

 

 

 

 

2

 

 

 

 

 

3

 

 

 

 

 

4

 

 

 

 

 

5

 

 

 

 

 

6

 

 

 

 

 

7

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  1. Using the data above, fill in the graph below, using different colors or symbols for the different lines.  Label each line.  If you wish, you may print out your graph from the Spectral Profile window and attach it with this lab, but be sure to edit the graph so that the different lines are clearly identifiable.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


Please answer the following questions with respect to absorption, scattering, and reflection.

 

17.  Why are the values for water so consistently low?

 

 

 

 

 

 

 

 

18.  Why is it that the forest values are relatively low in bands 1, 2 and 3, but abruptly increase in bands 4 and 5?

 

 

 

 

 

 

 

19.    Why do you think the fields have higher values in all bands than the forest?

 

 

 

 

 

 

 

 

 

 

20.  Which of the various surfaces is the warmest?  Why is this?

 

 

 

 

 

 

 

 

 

 

 

21.  Speculate on the nature of the mystery area based on a comparison of its spectral graph relative to other materials.

 

 

 

 

 

 

 

 

 

 

22.  How might you use the kind of spectral information you have gathered to find or classify similar features in the rest of the image?