Title :
Using Suitable Neighbors to Augment the Training Set in Hyperspectral Maximum Likelihood Classification
Author :
Richards, John A. ; Jia, Xiuping
Author_Institution :
Coll. of Eng. & Comput. Sci., Australian Nat. Univ., Canberra, ACT
Abstract :
A method is presented for supplementing the training set in maximum likelihood classification of hyperspectral data to mitigate the Hughes phenomenon. Based on the idea that the near neighbors of training pixels are likely to come from the same class, measures are proposed to assess neighbors as potential candidates so that those selected give improved class statistics and classification accuracy.
Keywords :
geophysical techniques; image classification; remote sensing; AVIRIS Northwestern Indiana data set; Hughes phenomenon mitigation; USA; hyperspectral maximum likelihood classification; labeled training pixels; neighbor selection method; training set; user-identified training pixels; Classification; generalization; hyperspectral; training;
Journal_Title :
Geoscience and Remote Sensing Letters, IEEE
DOI :
10.1109/LGRS.2008.2005512