Hyperspectral subpixel target detection based on extended mathematical morphology

A hyperspectral subpixel target detection algorithm was proposed based on extended mathematical morphology and spectral angle mapping. The spectral and spatial information had been used to locate and detect targets under the condition that prior knowledge of targets and background was unknown. Then hyperspectral subpixel targets was detected and recognized. The extended mathematical morphological erosion and dilation operations were performed respectively to extract endmembers. The spectral angle mapping method was used to detect and recognize interested targets. The hyperspectral image collected by AVIRIS was applied to evaluate the proposed algorithm. The proposed algorithm was compared with SAM algorithm and RX algorithm by a specifically designed experiment. From the results of the experiments, it is illuminated that the proposed algorithm can detect subpixel targets with low false alarm rate and its performance is better than that of the other algorithms under the same condition.