GEOG 372: Introduction to Remote Sensing
Fall 2007
September 26th & 27th
Laboratory 4
Visual Analysis of Very High Resolution Remotely Sensed Imagery
No late labs can be accepted. Please type the answers just below the questions and hand the printouts at the beginning of the next lab: October 3rd (0102) and 4th (0101)
Part 1
Background: Remotely sensed imagery allows a synoptic point of view that can provide valuable information to those studying features on the earth's surface or other locations. This aerial perspective provides new and different information that can be added to data gathered on the ground. This unique way of collecting spatial information allows researchers to gather information about much broader areas. In this lab you will learn the basic visual cues that are used to interpret images. You have been developing these skills all of your life but this lab will help you to recognize the phenomena that you use to identify objects visually.
Images are based on recording the reflectance, scattering or emissions from different portions of the electromagnetic (EM) spectrum (e.g., visible, infrared, thermal infrared, microwave). These different portions of the EM spectrum will produce different visual patterns of the same features. However, seven visual characteristics can be used to identify objects or phenomena in remote sensing imagery no matter what type of images are being considered. Aerial panchromatic photography will be used in this exercise as an introduction to these visual characteristics. The underlying principle of interpretation is that every feature on the earth has its own unique "signature." This signature permits consistent recognition of the features in the image over space and time. The characteristics are:
1. Shape: distinctive form of a physical feature
2. Size: length, width, volume, height, perimeter, or area of a feature (size may be judged relative to some well known feature)
3a. Tone: (in black and white images) as represented by shades of gray; this tone is a function of the amount of light reflected in the portion of the electromagnetic spectrum that is being observed
3b. Color: (in color images) as represented by hue, saturation, and brightness
4. Texture: apparent roughness or smoothness of a surface
5. Pattern: regular or characteristic arrangement of color, tone, texture, and shape
6. Shadow: the lack of solar illumination; aids in the evaluation of texture, shape, and size of features; but may also obscure other features as well
7a. Site/situation: relative or absolute postion (latitude and longitude) of a feature or the location of features with respect to terrain, water, or landmarks
7b. Association: identification on the basis of other objects or features commonly associated with them.
Refer to the 2 photographs and answer the following questions
As the pictures below demonstrate, things look very different from the ground and from the air.
Questions:1. Why do objects on the ground look different in remotely sensed images?
2. What area does the aerial photo seen above display? Locate the ground-based photo on the aerial photography image seen above (describe the location using terms like top, bottom, middle, left, and/or right side of the image. You can also take a screen shot of this image and use arrows to describe the location).
3. What are the white lines on the image? Why are they white (what are they made of)? Does this material have high or low reflectance?
4. If you assume that north is at the top of this image, where was the sun in the sky? How do you know this?
5. What is the approximate size of the pixels in this image? How do you know this?
6. Guess the time of year that this image was collected and tell me why you decided on this time of year (fall, winter, spring, or summer).
Part 2
During this exercise we will work with a Digital Ortho Quarter Quadrangle (DOQQ) image located in the directory U:\\g372\372-fall07\lab4data
A digital orthophoto quadrangle (DOQ) is a computer-generated image of an aerial photograph in which image displacement caused by terrain relief and camera tilts has been removed. It combines the image characteristics of a photograph with the geometric qualities of a map. The standard DOQ’s produced by the U.S. Geological Survey (USGS) are either grayscale or color-infrared (CIR) images with a 1-meter ground resolution; they cover an area measuring 3.75- minutes longitude by 3.75-minutes latitude, approximately 5 miles on each side or a quarter of a USGS 7.5 minute quad. Each DOQ has between 50 and 300 meters of overedge image beyond the latitude and longitude corner crosses embedded in the image. This overedge facilitates tonal matching and mosaicking of adjacent images. All DOQ’s are referenced to the North American Datum of 1983 (NAD83) and cast on the Universal Transverse Mercator (UTM) projection. Primary (NAD83) and secondary (NAD27) datum coordinates for the upper left pixel are included in the header to allow users to spatially reference other digital data with the DOQ. (http://www.usgsquads.com/prod_doqq.htm)
Open ENVI 4.2. On the ENVI toolbar (located in the upper left portion of your screen) go to File – Open Image File and navigate to the directory where the class images are stored. Open TAYLO1NE.TIF. Do not assign any projection when it opens, but in the box labeled Zone, enter the number 18. Then just click OK in the pop-up window. Load image TAYLO1NE in Display 1 in RGB combination (Load band R in red channel, band G in green channel, band B in blue channel).
7. Is the image a true or a false color composite? Why do you think so?
8. Examine the header of the image. What is its pixel size (=spatial resolution)?
As stated before, there are 7 main components for visual image interpretation: size, shape, texture, shadow, tone/color, pattern, and site/situation/association. Find pixels located at the x/y coordinates below and examine the surrounding locations. Identify the object in this location and describe how each of the visual interpretation components contributed to image interpretation (note: not every component will play an important role every time). For example, navigate to pixel (1493, 7567). To do that, go to Tools – Pixel locator. In the “Pixel locator” pop-up window type in “Samples” 1493 and “Lines” 7567. Click Apply. In the zoom display window you will see a red cross centered on the pixel of interest. Examine the entire area within the zoom display window and also look at the surroundings in scroll and image display windows. Your pixel is centered on a pier. The reasons for this interpretation are (in order of their importance): 1) tone/color – bright white object in the middle of a water body; 2) site – surrounded by water and attached to land on one side; 3) shape – long narrow object; 4) shadow – cast a shadow over the water which implies that it is positioned above the water or it is 3-dimensional with significant height; 5) texture - it appears to be smooth; 6) it’s size relative to other objects is consistent with that of a pier; 7) the pattern it creates relative to the shoreline is also consistent with that of a pier.
9. Identify the objects located at these x/y coordinates and describe the contribution of various visual interpretation components:
5458, 1301
1662, 7135
5953, 2340
2727, 6533
1269, 3177
1792, 3670
10. What season do you think this image was acquired in? Why?
11. Look at the agriculture fields in the image. Why are they varying shades of red, pink, and white? What material on the ground is red in the image? What material on the ground is white in the image? Why is there a striped pattern?
12. Look at the water in the image. Why are some areas brighter (lighter blue) than others?
13. Navigate to 3935, 4980 and examine the general area. Look at the blue areas of land bordering the river, bay, agricultural fields, and forested areas. What type of land cover is this? Why?
Log off your computer and turn off the monitor when you are finished.