The Tasseled-Cap Transformation is a conversion of the original bands of an image into a new set of bands with defined interpretations that are useful for vegetation mapping. A tasseled-cap transform is performed by taking “linear combinations” of the original image bands - similar in concept to principal components analysis. So each tasseled-cap band is created by the sum of image band 1 times a constant plus image band 2 times a constant, etc… The coefficients used to create the tasseled-cap bands are derived statistically from images and empirical observations and are specific to each imaging sensor.
The first tasseled-cap band corresponds to the overall brightness of the image. The second tasseled-cap band corresponds to “greenness” and is typically used as an index of photosynthetically-active vegetation. The third tasseled-cap band is often interpreted as an index of “wetness” (e.g., soil or surface moisture) or “yellowness” (e.g., amount of dead/dried vegetation).
A tasseled-cap transformation produces the same number of output bands as input bands, but not all of the tasseled-cap output bands will be useful. Because of the similarity between tasseled-cap and principal components, the first tasseled cap contains the most information from the image and each subsequent band contains less of the original image information. Generally, the first three tasseled-cap bands contain useful information and the rest contain much of the image's “noise” and are not used.
The tasseled-cap transformation was originally defined by Kauth and Thomas (1976) based on a spectral analysis of the growth of wheat in fields. The transformation got it's name from the way that a graph looked when the red band values of pixels were plotted against the near infra-red pixel values. The tasseled-cap transformation coefficients were defined against this graph to maximize the separation of the different growth stages of wheat.
Image source: Kauth and Thomas (1976), axis labels added.
The tasseled cap is a type of data transformation. Another commonly applied transformation similar to the tasseled cap is Principal Components Analysis. Other data transformations are used to calculate indices that are related to specific ecosystem attributes. Examples of these are the Normalized Difference Vegetation Index, the Enhanced Vegetation Index, and the Normalized Burn Ratio.
The output of a Tasseled-cap transformation is a new set of image bands that have specific interpretations. You will get the same number of bands out of a Tasseled-cap transformation as you put into it, but not all of them will be useful. In general, the majority of the information in an image will be contained in the first three Tasseled Cap bands. Band 4 is typically a “noise” band and contains little useful information. With a Tasseled-cap transformation on Landsat data, Tasseled-cap Band 5 may contain useful information, but Tasseled-Cap Band 6 will be another noise band.
One of the biggest limitations to the Tasseled-cap Transformation is that it can only be applied to sensors for which transformation coefficients have already been developed. Currently, this includes Landsat,MODIS, ASTER, Ikonos, and SPOT. There are several instances of Tasseled-cap coefficients that were derived from one sensor being applied to another one (e.g., Landsat coefficients applied to Quickbird data), but this is generally not recommended.
Performing a Tasseled-cap transformation requires two things: 1) an input satellite image, and 2) a set of transformation coefficients specific to the sensor that acquired the image. Transformation coefficients can be defined to work with either radiance or reflectance, and it is important to know which the transformation coefficients have been defined for (and which your image is using!).
There are no special hardware requirements to perform a tasseled-cap transformation. In terms of software requirements, some sort of program that can perform image/raster operations on a cell-by-cell basis is needed. Most remote sensing packages have the ability to do simple image arithmetic (i.e., multiply bands by a constant value and add the results together on a cell-by-cell basis). Some remote sensing packages, like ERDAS Imagine, have Tasseled-cap functions. Tasseled-cap transformation, however, can be done in other programs like ArcGIS, R, SAS, or even some photo-manipulation programs (e.g., Photoshop, GIMP).
You must have an account and be logged in to post or reply to the discussion topics below. Click here to login or register for the site.