A non-negative matrix factorization approach for hyperspectral unmixing with partial known endmembers

In this paper, the ground truth information is introduced to improve the accuracy of hyperspectral unmixing based on nonnegative matrix factorization. Specifically, the partial known endmembers which could be surveyed is introduced in NMF model. The relationship between the known and unknown endmembers are explored. The distance function is designed to describe the relationship and combined with NMF model. In this way, the new proposed NMF approach, called PENMF, could use the known endmembers to help estimating the unknown endmembers, so that accurate and robust results can be obtained. The proposed algorithm was compared with NMFupk, which also considered partial known endmembers, using extensive synthetic data and real hyperspectral data. The experiments show that the proposed algorithm can give a better performance.

[1]  P. Paatero,et al.  Positive matrix factorization: A non-negative factor model with optimal utilization of error estimates of data values† , 1994 .

[2]  José M. Bioucas-Dias,et al.  Vertex component analysis: a fast algorithm to unmix hyperspectral data , 2005, IEEE Transactions on Geoscience and Remote Sensing.

[3]  Wei Xia,et al.  An approach based on constrained nonnegative matrix factorization to unmix hyperspectral data , 2011, IEEE Transactions on Geoscience and Remote Sensing.

[4]  Antonio J. Plaza,et al.  A quantitative and comparative analysis of endmember extraction algorithms from hyperspectral data , 2004, IEEE Transactions on Geoscience and Remote Sensing.

[5]  Antonio J. Plaza,et al.  Collaborative nonnegative matrix factorization for remotely sensed hyperspectral unmixing , 2012, 2012 IEEE International Geoscience and Remote Sensing Symposium.

[6]  Hairong Qi,et al.  Endmember Extraction From Highly Mixed Data Using Minimum Volume Constrained Nonnegative Matrix Factorization , 2007, IEEE Transactions on Geoscience and Remote Sensing.

[7]  H. Sebastian Seung,et al.  Algorithms for Non-negative Matrix Factorization , 2000, NIPS.

[8]  Jun Zhou,et al.  Hyperspectral Unmixing via $L_{1/2}$ Sparsity-Constrained Nonnegative Matrix Factorization , 2011, IEEE Transactions on Geoscience and Remote Sensing.

[9]  Paul D. Gader,et al.  Sparsity Promoting Iterated Constrained Endmember Detection in Hyperspectral Imagery , 2007, IEEE Geoscience and Remote Sensing Letters.

[10]  Mireille Guillaume,et al.  Minimum Dispersion Constrained Nonnegative Matrix Factorization to Unmix Hyperspectral Data , 2010, IEEE Transactions on Geoscience and Remote Sensing.

[11]  Bo Du,et al.  An Endmember Dissimilarity Constrained Non-Negative Matrix Factorization Method for Hyperspectral Unmixing , 2013, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[12]  Hassan Ghassemian,et al.  Spectral Unmixing of Hyperspectral Imagery Using Multilayer NMF , 2014, IEEE Geoscience and Remote Sensing Letters.

[13]  Sen Jia,et al.  Constrained Nonnegative Matrix Factorization for Hyperspectral Unmixing , 2009, IEEE Transactions on Geoscience and Remote Sensing.

[14]  Chein-I. Chang.  Spectral information divergence for hyperspectral image analysis , 1999, IEEE 1999 International Geoscience and Remote Sensing Symposium. IGARSS'99 (Cat. No.99CH36293).

[15]  Antonio J. Plaza,et al.  Impact of Initialization on Design of Endmember Extraction Algorithms , 2006, IEEE Transactions on Geoscience and Remote Sensing.

[16]  Mark L. G. Althouse,et al.  Least squares subspace projection approach to mixed pixel classification for hyperspectral images , 1998, IEEE Trans. Geosci. Remote. Sens..

[17]  Zhenwei Shi,et al.  Nonnegative matrix factorization for hyperspectral unmixing using prior knowledge of spectral signatures , 2012 .

[18]  Chein-I. Chang.  Hyperspectral Imaging: Techniques for Spectral Detection and Classification , 2003 .