Hyperspectral unmixing is a powerful tool for the remote sensing image mining. Nonnegative matrix factorization (NMF) has been adopted to deal with this issue, while the precision of unmixing is closely related with the local minimizers of NMF. We present two novel initialization strategies that is based on CUR decomposition, which is physically meaningful. In the experimental test, NMF with the new initialization method is used to unmix the urban scene which was captured by airborne visible/infrared imaging spectrometer (AVIRIS) in 1997, numerical results show that the initialization methods work well.
 Pauca, V.P., Piper, J. and Plemmons, R.J. (2006) Nonnegative Matrix Factorization for Spectral Data Analysis. Linear Algebra and Its Applications, 416, 29-47. http://dx.doi.org/10.1016/j.laa.2005.06.025
 Lin, C.J. (2007) Projected Gradient Methods for Nonnegative Matrix Factorization. Neural Comput., 19, 2756-2779. http://dx.doi.org/10.1162/neco.2007.19.10.2756
 Pattero, P.A. and Tapper, U. (1994) Positive Matrix Factorization: A Non-Negative Factor Model with Optimal Utilization of Error Estimates of Data Values. Environmetrics, 5, 111-126. http://dx.doi.org/10.1002/env.3170050203
 Langville, A., Meyer, C., Albright, R., Cox, J. and Duling, D. (2006) Algorithms, Initializations, and Convergence for the Nonnegative Matrix Factorization. NCSU. Eprint Arxiv. http://langvillea.people.cofc.edu/NMFInitAlgConv.pdf
 Boutsidis, C. and Gallopoulos, E. (2008) SVD Based Initialization: A Head Start for Nonnegative Matrix Factorization. Pattern Recognition, 41, 1350-1362. http://dx.doi.org/10.1016/j.patcog.2007.09.010
 Kim, H. and Park, H. (2008) Non-Negative Matrix Factorization Based on Alternating Non-Negativity Constrained Least Squares and Active Set Method. SIAM J. Matrix Analysis Application, 30,713-730. http://dx.doi.org/10.1137/07069239X