Spectral segmentation based dimension reduction for hyperspectral image classification

Hyperspectral images (HSI) contain a wide range of information, the most prominent technology for observing earth. However, using an original HSI high-dimensional datacube, classification task faces significant challenges since it has high computational cost. As result, dimensionality reduction is indispensable. A dimension method been introduced in this paper, including feature extraction and selection to obtain subsets. Minimum Noise Fraction (MNF) popular HSI, requiring capability. We propose segmented MNF that divides complete into groups utilising normalised cross-cumulative residual entropy (nCCRE). An nCCRE-based also employed improve quality chosen features max-relevancy min-redundancy measure. The support vector machine (SVM) classifier used on two real evaluate efficiency extracted