Combined FATEMD‐based band selection method for hyperspectral images

Guest Editorial: Hyperspectral Imaging and Applications

“…Four papers are presented for band selection of HSI [13–16]. Specifically, Zhang et al propose a crossover‐based gravitational search algorithm to identify and select the most informative subset of band [13].…”

“…Specifically, Zhang et al propose a crossover‐based gravitational search algorithm to identify and select the most informative subset of band [13]. Yu et al present an approach to select band subset based on combined fast and adaptive tridimensional empirical mode decomposition [14]. Chen et al propose an approach to combine band selection and feature extraction for dimension reduction of HSI [15], where determinantal point process is exploited for band selection followed by two‐dimensional singular spectrum analysis for feature extraction on the selected bands.…”

“…For effective feature extraction, one trend is the feature transforms. Followed by the success of folded‐principal analysis [26], wavelet [27] and discrete cosine transform [28], empirical mode decomposition [14], and singular spectral analysis [15] are used for band selection and QR factorisation [19] for spectral unmixing.…”