Two-stage denoising method for hyperspectral images combining KPCA and total variation

Wenzhi Liao; Jan Aelterman; Hiep Luong; Aleksandra Pizurica; Wilfried Philips; Peter Hobson; Gennaro Percannella; Mario Vento; Arnold Wiliem

doi:10.1109/ICIP.2013.6738422

Two-stage denoising method for hyperspectral images combining KPCA and total variation

Wenzhi Liao, Jan Aelterman, Hiep Luong, Aleksandra Pizurica, Wilfried Philips, Peter Hobson (Editor), Gennaro Percannella (Editor), Mario Vento (Editor), Arnold Wiliem (Editor)

Electronic And Electrical Engineering

Research output: Contribution to conference › Paper › peer-review

3 Citations (Scopus)

Abstract

This paper presents a two-stage denoising method for hyperspectral image (HSI) by combining kernel principal component analysis (KPCA) and total variation (TV). In the first stage, we use KPCA denoising to reduce spectrally uncorrelated noise. In the second stage, the information content is largely separated from the remaining noise by means of principal component analysis (PCA). The remaining noise is then efficiently removed by fast primal-dual TV denoising in lowenergy PCA channels. Experimental results on simulated and real HSIs are very encouraging.

Original language	English
Pages	2048-2052
Number of pages	5
DOIs	https://doi.org/10.1109/ICIP.2013.6738422
Publication status	Published - 2013
Event	2013 20th IEEE International Conference on Image Processing, ICIP 2013 - Melbourne, VIC, United Kingdom Duration: 15 Sept 2013 → 18 Sept 2013

Conference

Conference	2013 20th IEEE International Conference on Image Processing, ICIP 2013
Country/Territory	United Kingdom
City	Melbourne, VIC
Period	15/09/13 → 18/09/13

Keywords

total variation
classification
kernel principal component analysis
denoising
hyperspectral images
component analysi

Access to Document

10.1109/ICIP.2013.6738422

Cite this

Liao, W., Aelterman, J., Luong, H., Pizurica, A., Philips, W., Hobson, P. (Ed.), Percannella, G. (Ed.), Vento, M. (Ed.), & Wiliem, A. (Ed.) (2013). Two-stage denoising method for hyperspectral images combining KPCA and total variation. 2048-2052. Paper presented at 2013 20th IEEE International Conference on Image Processing, ICIP 2013, Melbourne, VIC, United Kingdom. https://doi.org/10.1109/ICIP.2013.6738422

@conference{c884e8bf176a435491d6661b0e4ad2dc,

title = "Two-stage denoising method for hyperspectral images combining KPCA and total variation",

abstract = "This paper presents a two-stage denoising method for hyperspectral image (HSI) by combining kernel principal component analysis (KPCA) and total variation (TV). In the first stage, we use KPCA denoising to reduce spectrally uncorrelated noise. In the second stage, the information content is largely separated from the remaining noise by means of principal component analysis (PCA). The remaining noise is then efficiently removed by fast primal-dual TV denoising in lowenergy PCA channels. Experimental results on simulated and real HSIs are very encouraging.",

keywords = "total variation, classification, kernel principal component analysis, denoising, hyperspectral images, component analysi",

author = "Wenzhi Liao and Jan Aelterman and Hiep Luong and Aleksandra Pizurica and Wilfried Philips and Peter Hobson and Gennaro Percannella and Mario Vento and Arnold Wiliem",

year = "2013",

doi = "10.1109/ICIP.2013.6738422",

language = "English",

pages = "2048--2052",

note = "2013 20th IEEE International Conference on Image Processing, ICIP 2013 ; Conference date: 15-09-2013 Through 18-09-2013",

}

Liao, W, Aelterman, J, Luong, H, Pizurica, A, Philips, W, Hobson, P (ed.), Percannella, G (ed.), Vento, M (ed.) & Wiliem, A (ed.) 2013, 'Two-stage denoising method for hyperspectral images combining KPCA and total variation', Paper presented at 2013 20th IEEE International Conference on Image Processing, ICIP 2013, Melbourne, VIC, United Kingdom, 15/09/13 - 18/09/13 pp. 2048-2052. https://doi.org/10.1109/ICIP.2013.6738422

TY - CONF

T1 - Two-stage denoising method for hyperspectral images combining KPCA and total variation

AU - Liao, Wenzhi

AU - Aelterman, Jan

AU - Luong, Hiep

AU - Pizurica, Aleksandra

AU - Philips, Wilfried

A2 - Hobson, Peter

A2 - Percannella, Gennaro

A2 - Vento, Mario

A2 - Wiliem, Arnold

PY - 2013

Y1 - 2013

N2 - This paper presents a two-stage denoising method for hyperspectral image (HSI) by combining kernel principal component analysis (KPCA) and total variation (TV). In the first stage, we use KPCA denoising to reduce spectrally uncorrelated noise. In the second stage, the information content is largely separated from the remaining noise by means of principal component analysis (PCA). The remaining noise is then efficiently removed by fast primal-dual TV denoising in lowenergy PCA channels. Experimental results on simulated and real HSIs are very encouraging.

AB - This paper presents a two-stage denoising method for hyperspectral image (HSI) by combining kernel principal component analysis (KPCA) and total variation (TV). In the first stage, we use KPCA denoising to reduce spectrally uncorrelated noise. In the second stage, the information content is largely separated from the remaining noise by means of principal component analysis (PCA). The remaining noise is then efficiently removed by fast primal-dual TV denoising in lowenergy PCA channels. Experimental results on simulated and real HSIs are very encouraging.

KW - total variation

KW - classification

KW - kernel principal component analysis

KW - denoising

KW - hyperspectral images

KW - component analysi

UR - http://hdl.handle.net/1854/LU-4143226

U2 - 10.1109/ICIP.2013.6738422

DO - 10.1109/ICIP.2013.6738422

M3 - Paper

SP - 2048

EP - 2052

T2 - 2013 20th IEEE International Conference on Image Processing, ICIP 2013

Y2 - 15 September 2013 through 18 September 2013

ER -

Two-stage denoising method for hyperspectral images combining KPCA and total variation

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this