The Tasseled Cap Transformation is a technique commonly used in land cover mapping or other classification projects. It takes the linear combination of satellite imagery bands and a specialized coefficient matrix to create a n-band image with the first 3 bands containing the majority of the useful information, similar to Principal Component Analysis. The first three bands created are generally held to represent Brightness, Greenness and Wetness. The coefficient matrix, which is unique to each imaging sensor, is based on image statistics and empirical observations.
|Sentinel-2 on the left (Truecolor 4,3,2), Tasselled Cap on the right (Brightness, Greenness, Wetness)|
The paper by Nedkov provides the coefficients specific to the Sentinel-2 sensors.
13-band Sentinel-2 data in the standard order of Bands 1, 2, 3, 4, 5, 6, 7, 8, 8A, 9, 10, 11, 12
The input data is assumed to be unsigned integer (such that pixels where all bands are 0 are assumed to be NoData). If the input is a different datatype (such as float), alternative approaches may have to be taken to identify NoData.
Nedkov, R. (2017). ORTHOGONAL TRANSFORMATION OF SEGMENTED IMAGES FROM THE SATELLITE SENTINEL-2. Comptes rendus de l'Académie bulgare des sciences. 70. 687-692.
|Sentinel-2 truecolor (4,3,2) on the left, Brightness on the right|
|Sentinel-2 truecolor (4,3,2) on the left, Greenness on the right|
|Sentinel-2 truecolor (4,3,2) on the left, Wetness on the right|
Sentinel-2 Filename In: Enter the filename of the Sentinel-2 input, with the following channels: 1,
2, 3, 4, 5, 6, 7, 8, 8A, 9, 10, 11, 12
Tasseled Cap Filename Out: Enter the filename of the tasseled cap output. Band 1 will be the Brightness, Band 2 will be Greenness and Band 3 will be Wetness.