Method for Estimating the Number of Mixed Spectral Endmembers Based on Feature-Enhanced Spatial Spectral Matching Algorithm

Based on the characteristics of space target components, this paper proposes a spectral in-degree distribution curvematching algorithm. Simulation experiments showed the average accuracy of the proposed algorithm is improved in comparison with the Harsanyi–Farrand–Chang and HySime algorithms. The simulation experiments were performed for a case in which the number of mixed spectral bands was reasonably small. The results showed the number of endmembers could be estimated accurately when the number of endmembers was three to eight. The algorithm proposed in this paper is suitable for estimating the number of endmembers of a space target.