Summarization of Band Selection Methods For Hyperspectral Images

Nowadays, Hyperspectral images are used for analysis of images taken from satellite. These images have a lot of information but processing these images is very difficult. The number of bands in normal images are ("Red", "Green", "Blue") but spectral images have a lot of bands. Dimension reduction is a necessary step in hyperspectral image analysis. So, in this paper we are reviewing current best dimension reduction techniques for hyperspectral image classification. The challenges of hyperspectral band selection are also discussed.