Title :
An improved decorrelation method for the efficient display of multispectral data
Author :
Durand, Jean M. ; Kerr, Yann H.
Author_Institution :
Centre Nat. d´´Etudes Spatiales, Lab. d´´Etudes et de Recherches en Teledetection Spatiale, Toulouse, France
fDate :
9/1/1989 12:00:00 AM
Abstract :
A technique to enhance multispectral image displays and to improve their visual interpretations is proposed. The decorrelation approach is tested with a SPOT simulation and thematic mapper image. The technique is considered to be of great help for the interpretation of remotely sensed images. Furthermore, the choice between two-, three-, and N -dimensional decorrelation allows the user to adjust to specific cases. However, the main drawback of this technique is its strong link to the statistics of the image being displayed, which makes the analysis of the displayed image data (in terms of spectral properties of the surface, for example) more complex. For a specific application conducted with a given satellite, this inconvenience could be overcome by using a set of fixed linear combinations
Keywords :
correlation methods; geophysical techniques; picture processing; remote sensing; SPOT simulation; decorrelation method; image displays; multispectral data; remotely sensed images; statistics; technique; thematic mapper image; visual interpretations; Color; Colored noise; Crops; Decorrelation; Displays; Multispectral imaging; Remote sensing; Signal synthesis; Signal to noise ratio; Vegetation mapping;
Journal_Title :
Geoscience and Remote Sensing, IEEE Transactions on
DOI :
10.1109/TGRS.1989.35944