A simulated annealing feature extraction approach for hyperspectral images

摘要: In this paper, a novel study of the simulated annealing feature extraction (SAFE) for high-dimensional remote sensing images is proposed. The approach based on greedy modular eigenspace (GME) scheme. GME was developed by clustering highly correlated bands into smaller subset algorithm. Unfortunately, doesn't guarantee to reach global optimal solution algorithm except exhaustive search method. Accordingly, finding an (or near-optimal) very expensive. order overcome disadvantage, SAFE scheme introduced improve performance optimally modifying correlation coefficient operations and taking sets non-correlated hyperspectral heuristic optimization It presents framework, which consists two algorithms, referred as scale uniformity transformation (FSUT). designed extract features new defined three-dimensional (SAME) optimize eigenspace, while FSUT performed fuse most from different spectrums associated with data sources. proposed method evaluated applying it airborne synthetic aperture radar (SAR) images. experimental results demonstrated that not only effective but also alternative existing dimensionality reduction methods.