Hyperspectral Image Denoising Using First Order Spectral Roughness Penalty in Wavelet Domain

Author

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

Author_Institution

Dept. of Electr. & Comput. Eng., Univ. of Iceland, Reykjavik, Iceland

Volume

7

Issue

6

fYear

2014

fDate

Jun-14

Firstpage

2458

Lastpage

2467

Abstract

In this paper, a new denoising method for hyperspectral images is proposed using First Order Roughness Penalty (FORP). FORP is applied in the wavelet domain to exploit the Multi-Resolution Analysis (MRA) property of wavelets. Stein´s Unbiased Risk Estimator (SURE) is used to choose the tuning parameters automatically. The simulation results show that the penalized least squares using FORP can improve the Signal to Noise Ratio (SNR) compared to other denoising methods. The proposed method is also applied to a corrupted hyperspectral data set and it is shown that certain classification indices improve significantly.

Keywords

geophysical image processing; hyperspectral imaging; image denoising; remote sensing; MRA property; SURE; Stein Unbiased Risk Estimator; classification index; first order spectral roughness penalty; hyperspectral data set; hyperspectral image denoising; multiresolution analysis; remote sensing; signal-to-noise ratio; wavelet domain; Educational institutions; Hyperspectral imaging; Multiresolution analysis; Noise reduction; Signal to noise ratio; Classification; Stein´s unbiased risk estimator; denoising; first order roughness penalty; hyperspectral image; multiresolution analysis; penalized least squares; wavelets;

fLanguage

English

Journal_Title

Selected Topics in Applied Earth Observations and Remote Sensing, IEEE Journal of

Publisher

ieee

ISSN

1939-1404

Type

jour

DOI

10.1109/JSTARS.2013.2272879

Filename

6570741