Feature extraction of hyperspectral images based on reformulate computation of between class scatter matrixes

This paper proposes a method called Adjusted Linear Discriminant Analysis (ALDA) for feature extraction of hyperspectral remote sensing imagery. In this method the variances of the classes are considered as a weight to estimate between-class scattering matrix appropriately. Experimental results on well known hyperspectral dataset demonstrate that compared to conventional LDA based feature extraction algorithms the overall accuracy of the classification increased.