期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：1980
卷号：XXIII Part B10
页码：590-595
出版社：Copernicus Publications
摘要：The spectral separability studies are generally carriedout on the basis of histogr3m outputs to select suitable bandsto achieve maximum classification accuracy. The objective ofthis study is to select suitable statistical tests to test theh y po 1te s is w h e t h e r t h e d is t r i b u t i o n s o f s p e c t r a l v a 1 u e s indifferent classes represent the same papulation and the consequentimpact of rejectiun of such a hypothesis on the separability.In testing, normality is assumed in these samples butthe histogram outputs showed deviation from normality. Theapplication of non-parametric tests which have no valid assumptionhave been suggested. The Ko lmo gro v-Smir ina v two tailedtest is discussed and the maximum difference statistic S(x)-F (x) = Max/D/ indicated the degree uf separability
