Hyperspectral image denoising using a new linear model and Sparse Regularization

Behnood Rasti; Johannes R. Sveinsson; Magnus O. Ulfarsson; Jon Atli Benediktsson

doi:10.1109/IGARSS.2013.6721191

Hyperspectral image denoising using a new linear model and Sparse Regularization

Behnood Rasti, Johannes R. Sveinsson, Magnus O. Ulfarsson, Jon Atli Benediktsson

University of Iceland

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

This paper deals with hyperspectral image reconstruction using a new linear model and Sparse Regularization (SR). The new model is based on Principal Components (PCs) and wavelets. Since the hyperspectral PCs are not spatially sparse, wavelet is applied to get spatially sparse representation. Sparse regularization is used to recover the corrupted signal. The regularization parameter is chosen by Stein's Unbiased Risk Estimator (SURE). The results show improvements for simulated data sets compare to other denoising methods based on Signal to Noise Ratio (SNR). In addition, the methods are applied on a real noisy data set, and the results of the new method demonstrate visual improvement. The proposed approach is automatic, fast and has the ability to be applied on very large data sets.

Original language	English
Title of host publication	2013 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2013 - Proceedings
Pages	457-460
Number of pages	4
DOIs	https://doi.org/10.1109/IGARSS.2013.6721191
Publication status	Published - 2013
Event	2013 33rd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2013 - Melbourne, VIC, Australia Duration: 21 Jul 2013 → 26 Jul 2013

Publication series

Name	International Geoscience and Remote Sensing Symposium (IGARSS)

Conference

Conference	2013 33rd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2013
Country/Territory	Australia
City	Melbourne, VIC
Period	21/07/13 → 26/07/13

Other keywords

Hyperspectral image
Stein's unbiased risk estimator
denoising
principal components
singular value decomposition
sparse regularization
wavelet

Access to Document

10.1109/IGARSS.2013.6721191

Cite this

Rasti, B., Sveinsson, J. R., Ulfarsson, M. O., & Benediktsson, J. A. (2013). Hyperspectral image denoising using a new linear model and Sparse Regularization. In 2013 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2013 - Proceedings (pp. 457-460). Article 6721191 (International Geoscience and Remote Sensing Symposium (IGARSS)). https://doi.org/10.1109/IGARSS.2013.6721191

@inproceedings{aa4fb8b01f604ea2a4523751e88a790a,

title = "Hyperspectral image denoising using a new linear model and Sparse Regularization",

abstract = "This paper deals with hyperspectral image reconstruction using a new linear model and Sparse Regularization (SR). The new model is based on Principal Components (PCs) and wavelets. Since the hyperspectral PCs are not spatially sparse, wavelet is applied to get spatially sparse representation. Sparse regularization is used to recover the corrupted signal. The regularization parameter is chosen by Stein's Unbiased Risk Estimator (SURE). The results show improvements for simulated data sets compare to other denoising methods based on Signal to Noise Ratio (SNR). In addition, the methods are applied on a real noisy data set, and the results of the new method demonstrate visual improvement. The proposed approach is automatic, fast and has the ability to be applied on very large data sets.",

keywords = "Hyperspectral image, Stein's unbiased risk estimator, denoising, principal components, singular value decomposition, sparse regularization, wavelet",

author = "Behnood Rasti and Sveinsson, \{Johannes R.\} and Ulfarsson, \{Magnus O.\} and Benediktsson, \{Jon Atli\}",

year = "2013",

doi = "10.1109/IGARSS.2013.6721191",

language = "English",

isbn = "9781479911141",

series = "International Geoscience and Remote Sensing Symposium (IGARSS)",

pages = "457--460",

booktitle = "2013 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2013 - Proceedings",

note = "2013 33rd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2013 ; Conference date: 21-07-2013 Through 26-07-2013",

}

Rasti, B, Sveinsson, JR, Ulfarsson, MO & Benediktsson, JA 2013, Hyperspectral image denoising using a new linear model and Sparse Regularization. in 2013 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2013 - Proceedings., 6721191, International Geoscience and Remote Sensing Symposium (IGARSS), pp. 457-460, 2013 33rd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2013, Melbourne, VIC, Australia, 21/07/13. https://doi.org/10.1109/IGARSS.2013.6721191

Hyperspectral image denoising using a new linear model and Sparse Regularization. / Rasti, Behnood; Sveinsson, Johannes R.; Ulfarsson, Magnus O. et al.
2013 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2013 - Proceedings. 2013. p. 457-460 6721191 (International Geoscience and Remote Sensing Symposium (IGARSS)).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Hyperspectral image denoising using a new linear model and Sparse Regularization

AU - Rasti, Behnood

AU - Sveinsson, Johannes R.

AU - Ulfarsson, Magnus O.

AU - Benediktsson, Jon Atli

PY - 2013

Y1 - 2013

N2 - This paper deals with hyperspectral image reconstruction using a new linear model and Sparse Regularization (SR). The new model is based on Principal Components (PCs) and wavelets. Since the hyperspectral PCs are not spatially sparse, wavelet is applied to get spatially sparse representation. Sparse regularization is used to recover the corrupted signal. The regularization parameter is chosen by Stein's Unbiased Risk Estimator (SURE). The results show improvements for simulated data sets compare to other denoising methods based on Signal to Noise Ratio (SNR). In addition, the methods are applied on a real noisy data set, and the results of the new method demonstrate visual improvement. The proposed approach is automatic, fast and has the ability to be applied on very large data sets.

AB - This paper deals with hyperspectral image reconstruction using a new linear model and Sparse Regularization (SR). The new model is based on Principal Components (PCs) and wavelets. Since the hyperspectral PCs are not spatially sparse, wavelet is applied to get spatially sparse representation. Sparse regularization is used to recover the corrupted signal. The regularization parameter is chosen by Stein's Unbiased Risk Estimator (SURE). The results show improvements for simulated data sets compare to other denoising methods based on Signal to Noise Ratio (SNR). In addition, the methods are applied on a real noisy data set, and the results of the new method demonstrate visual improvement. The proposed approach is automatic, fast and has the ability to be applied on very large data sets.

KW - Hyperspectral image

KW - Stein's unbiased risk estimator

KW - denoising

KW - principal components

KW - singular value decomposition

KW - sparse regularization

KW - wavelet

UR - https://www.scopus.com/pages/publications/84894241628

U2 - 10.1109/IGARSS.2013.6721191

DO - 10.1109/IGARSS.2013.6721191

M3 - Conference contribution

SN - 9781479911141

T3 - International Geoscience and Remote Sensing Symposium (IGARSS)

SP - 457

EP - 460

BT - 2013 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2013 - Proceedings

T2 - 2013 33rd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2013

Y2 - 21 July 2013 through 26 July 2013

ER -

Hyperspectral image denoising using a new linear model and Sparse Regularization

Abstract

Publication series

Conference

Other keywords

Access to Document

Fingerprint

Cite this