An optimal distinguishing band selection method was presented in terms of the spectrum distinguishable degree.

Ancient study indicates optimal bands combination,texture and vegetation indices can advance classification accuracy in a certain extent.

The main results are as following:1、Based on TM remote sensing image,using the valuation Index of optimal bands .

A method of the best band combination selection based on spectral characteristics of surface features;

the intrinsical relationships of the “entropy”methods on optimal bands selection in multispectral remote sensing data interpretation are analysed,and,limitations of using these methods in hyperspectral remote sensing data interpretation explained.

According to reflective properties of major rocks on the lunar surface to different electromagnetic waves,this article starts with the theory of traditional multispectrum datasets bands selection,and analyses the methods of optimal bands selection adapting to multispectrum datasets of the lunar surface.

This paper takes the multitemporal remote sensing images as the data source, by attaching importance to the analysis of the selection of the optimum fusion bands and the methods of image processing which the water information is extracted, which offer a theory base for remote sensing technology applied in the research of water environment.

To detect the foreign fibers effectively,a method for selecting optimal detection bands was proposed based on the extrema distribution of reflectance differences between cotton fiber and foreign fibers.