Three Band Color Composite Imagery

Objectives

Introduction

In one single band from the Landsat 7 ETM sensor, the difference in energy levels between various land cover classifications may not be discernible. Since comparing the spectral characteristics of land features in multiple bands provides a better separation, or contrast, between different land surfaces, Landsat data from multiple bands can be combined to create a data product known as a composite image. Landsat composite images are often called three-band composite images since they are created using the measured energy level in each of three ETM+ spectral bands to control the amount of red, blue, and green in a color output image.

Mapping ETM data to an RGB display

Computers often use RGB (Red, Green, Blue) output to create color images. In an RGB display, all of the colors that make up an image are made up of a combination of red, green, and blue at varying levels of intensity, each ranging from 0-255. Each unique color has its own combination of red, green, and blue levels. With all of the possible combinations of red, green, and blue values, this provides for a display system capable of using millions of different colors. In the diagram below, each unique color is displayed with its red, green, and blue values.

The way the ETM data are mapped into three colors in the output image depends on the information that one wishes to be highlighted in the images. The spectral characteristics of the target being observed and the type of information a researcher hopes to extract from the raw data determine which bands will be used in the composite and which color (red, green, or blue) will be assigned to each band. For example, in some applications, it may be desirable that land cover classes be associated with familiar colors (e.g., grass is green). In other cases contrasting colors are preferred to highlight objects of interest from the background.

Regardless of the combination of bands used, the mapping of ETM+ sensor data to the RGB color display is the same. Three bands are selected, each is assigned to one of the three primary RGB colors, and the value of each color level is mapped to the measured value of each pixel in the appropriate band. For example, to create a composite image that maps the measured values of ETM+ bands 3, 2, and 1 to the colors red, green, and blue respectively, the color of each pixel would be calculated using the following logic:

For example, assume one pixel location in the Landsat 7 scene has the following measured energy levels:

This results is a RGB value of (18, 18,133 ) for that pixel location in the color composite image, which is a deep blue color (this color can be seen on the image above). This logic is repeated for every pixel in the scene being processed, until an entire image produced with the pixel values derived from a combination of each of the three bands.

Landsat three-band composite images are usually named using the three bands used to create the image in order from red to green to blue. Thus, the above example would be called a "321 Composite" image, since it was derived from bands 3, 2, and 1 and they were mapped to red, green, and blue, respectively. The following sections will discuss some of the three-band composites that are commonly derived from Landsat data.

True color composite images are created by combining the ETM spectral bands that most closely resemble the range of vision of the human eye. A true-color composite uses the visible red (band 3), visible green (band 2), and visible blue (band 1) channels to create an image that is very close to what a person would expect to see in a photograph of the same scene. The band to color mapping for a 321 Composite are:

[EXAMPLE IMAGE OF 321 COMPOSITE]

True color images are based entirely on reflected solar radiation in the visible portion of the electromagnetic spectrum. Haze in the atmosphere, shadows, clouds, and scattering all affect the quality and usefulness of a true-color composite. True-color images are often low in contrast and hazy in appearance since blue light is more easily scattered by the atmosphere.

True-color composite images can be very useful, especially when studying coastal regions, since energy in the visible bands can penetrate water surfaces. Particles in the water, such as sediment or algae, will reflect visible light and can therefore be detected by the visible sensors on the Landsat 7 ETM+ sensor. Using true-color composite imagery, we can observe and measure the amount of sediment flowing from rivers into larger bodies of water such as the Chesapeake Bay following storm events. We can also locate and measure large blooms of algae that threaten the water quality and fishery production in coastal waterways.

Near Infrared Composite (432)

A Near Infrared composite eliminates the visible blue band and uses a Near Infrared (NIR) band to produce the image. The resulting composite does not resemble what the human eye will see (for example, vegetation is red instead of green); however it is very useful to researchers. The mapping of color to band is:
· Band 4 (NIR) = red
· Band 3 (Visible red) = green
· Band 2 (Visible green) = blue

[EXAMPLE IMAGE OF 432 COMPOSITE]

Vegetation has a very high albedo in the NIR band since chlorophyll (the pigment in leaves that give plants their green color) reflects energy at this wavelength. Thus, in a 423 NIR composite image, vegetation is vividly depicted as varying shades of red. Since different types of vegetation have different levels of chlorophyll in their leaves, each type of plant has its own shade of red. This makes a 432 composite very useful in determining the extent of vegetation and in classifying different vegetation types as seen from space.

Water, which absorbs nearly all of the NIR energy that reaches its surface, appears very dark, nearly black, in a 432 NIR composite image. Therefore this type of imagery would not be useful for studying underwater features.

Short-wave Infrared Composite (743 or 742)

A Short-wave Infrared composite contains at least one band in the short-wave infrared (SWIR) portion of the electromagnetic spectrum. The other bands used can vary depending on the use of the composite data. Some examples of SWIR composite images would include the following bands mapped to RGB colors:

Or...

[EXAMPLE IMAGE OF 742 COMPOSITE]

The albedo of surface materials in the SWIR portion of the spectrum is determined primarily by the moisture content of the surfaces being measured. Vegetation that is under stress (due to drought, pests, climate change, pollution, etc) will generally have less moisture content than healthy vegetation. Therefore, in a SWIR composite image, vegetation stress can be detected and appropriate measures can be taken to protect vegetation in stressed areas. SWIR bands composites are also very useful in detecting soil types and soil disturbance since moisture is an important characteristic of soil structure.

© CGIS at Towson University