Because of the multispectral or vector character of remote sensing image data we can devise transformations that generate new pixel vector descriptions from the old, and thereby create synthetic image components, or bands, as shown in Fig. 6.1. They represent an alternative description of the data. The new components are related to the old brightness values of the pixels in the original set of spectral bands via a mathematical operation, which is usually linear. Such a transformed image may make evident features that are not easily seen in the original data; alternatively it might be possible to preserve most of the essential information content of the original image using a reduced number of the transformed dimensions. The last point has significance for displaying data in the three colour primaries, and for compressed transmission and storage of data. Fig. 6.1 The principle of spectral domain transformation
CITATION STYLE
Richards, J. A. (2013). Spatial Domain Image Transforms. In Remote Sensing Digital Image Analysis (pp. 203–246). Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-642-30062-2_7
Mendeley helps you to discover research relevant for your work.