Empirical Automatic Estimation of the Number of Endmembers in Hyperspectral Images

Bin Luo (1) , Jocelyn Chanussot (2) , Sylvain Douté (3) , Liangpei Zhang (1)

1 Key Laboratory of Western China's Environmental Systems (Ministry of Education)
2 GIPSA-SIGMAPHY - GIPSA - Signal Images Physique
3 IPAG - Institut de Planétologie et d'Astrophysique de Grenoble

In this letter, an eigenvalue-based empirical method is proposed in order to estimate the number of endmembers in hyperspectral data. This method is based on the distribution of the differences of the eigenvalues from the correlation and the covariance matrices, respectively. The eigenvalues corresponding to the noise are identical in the covariance and the correlation matrices, while the eigenvalues corresponding to the signal (the endmembers) are larger in the correlation matrix than in the covariance matrix. The proposed method is totally parameter free and very fast. It is validated by experiments carried on both synthetic and real data sets.

Mots clés: Imaging; spectral analysis

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Bin Luo, Jocelyn Chanussot, Sylvain Douté, Liangpei Zhang. Empirical Automatic Estimation of the Number of Endmembers in Hyperspectral Images. IEEE Geoscience and Remote Sensing Letters, 2013, 10 (1), pp.24-28. ⟨10.1109/LGRS.2012.2189934⟩. ⟨hal-00798401⟩