Advanced search
1 file | 1.22 MB

Non-negative matrix factorization with mixture of Itakura-Saito divergence for SAR images

Chi Liu (UGent) , Wenzhi Liao (UGent) , Heng-Chao Li and Wilfried Philips (UGent)
Author
Organization
Abstract
Synthetic aperture radar (SAR) data are becoming more and more accessible and have been widely used in many applications. To effectively and efficiently represent multiple SAR images, we propose the mixture of Itakura-Saito (IS) divergence for non-negative matrix factorization (NMF) to perform the dimension reduction. Our proposed method incorporates the unit-mean Gamma mixture model into the NMF to model the multiplicative noise. To obtain the closed-form update equations as much as possible, we approximate the log-likelihood function with its lower bound. Finally, we apply Expectation-Maximization (EM) algorithm to estimate the parameters, resulting in the closed-form multiplicative update rules for the two matrix factors. Experimental results on real SAR dataset demonstrate the effectiveness of the proposed method and its applicability to post applications (e.g., classification) with improved performances over the conventional dimension reduction methods.
Keywords
Non-negative matrix factorization (NMF), synthetic aperture radar (SAR), dimension reduction, Itakura-Saito divergence, mixture model.

Downloads

  • Chi IGARSS2017.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 1.22 MB

Citation

Please use this url to cite or link to this publication:

Chicago
Liu, Chi, Wenzhi Liao, Heng-Chao Li, and Wilfried Philips. 2017. “Non-negative Matrix Factorization with Mixture of Itakura-Saito Divergence for SAR Images.” In 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS) , 779–782. IEEE.
APA
Liu, Chi, Liao, W., Li, H.-C., & Philips, W. (2017). Non-negative matrix factorization with mixture of Itakura-Saito divergence for SAR images. 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS) (pp. 779–782). Presented at the 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS 2017), IEEE.
Vancouver
1.
Liu C, Liao W, Li H-C, Philips W. Non-negative matrix factorization with mixture of Itakura-Saito divergence for SAR images. 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS) . IEEE; 2017. p. 779–82.
MLA
Liu, Chi, Wenzhi Liao, Heng-Chao Li, et al. “Non-negative Matrix Factorization with Mixture of Itakura-Saito Divergence for SAR Images.” 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS) . IEEE, 2017. 779–782. Print.
@inproceedings{8567913,
  abstract     = {Synthetic aperture radar (SAR) data are becoming more and more accessible and have been widely used in many applications. To effectively and efficiently represent multiple SAR images, we propose the mixture of Itakura-Saito (IS) divergence for non-negative matrix factorization (NMF) to perform the dimension reduction. Our proposed method incorporates the unit-mean Gamma mixture model into the NMF to model the multiplicative noise. To obtain the closed-form update equations as much as possible, we approximate the log-likelihood function with its lower bound. Finally, we apply Expectation-Maximization (EM) algorithm to estimate the parameters, resulting in the closed-form multiplicative update rules for the two matrix factors. Experimental results on real SAR dataset demonstrate the effectiveness of the proposed method and its applicability to post applications (e.g., classification) with improved performances over the conventional dimension reduction methods.},
  author       = {Liu, Chi and Liao, Wenzhi and Li, Heng-Chao and Philips, Wilfried},
  booktitle    = {2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS) },
  isbn         = {978-1-5090-4951-6},
  issn         = {2153-6996},
  language     = {eng},
  location     = {Fort Worth, TX, USA},
  pages        = {779--782},
  publisher    = {IEEE},
  title        = {Non-negative matrix factorization with mixture of Itakura-Saito divergence for SAR images},
  url          = {http://dx.doi.org/10.1109/IGARSS.2017.8127068},
  year         = {2017},
}

Altmetric
View in Altmetric
Web of Science
Times cited: