Hyperspectral image deblurring with PCA and total variation

Wenzhi Liao, Bart Goossens, Jan Aelterman, Hiep Luong, Aleksandra Pizurica, Niels Wouters, Wouter Saeys, Wilfried Philips

Research output: Contribution to conferencePaper

9 Citations (Scopus)

Abstract

In this paper, we propose a novel algorithm for hyper-spectral (HS) image deblurring with principal component analysis (PCA) and total variation (TV). We first decorrelate the HS images and separate the information content from the noise by means of PCA. Then, we employ the TV method to jointly denoise and deblur the first principal components (PCs). Subsequently, noise in the last principal components is suppressed using a simple soft-thresholding scheme, for computational efficiency. Experimental results on simulated and real HS images are very encouraging.
Original languageEnglish
Number of pages4
DOIs
Publication statusPublished - 2013
Event2013 5th Workshop on hyperspectral image and signal processing : evolution in remote sensing (WHISPERS) - Gainesville, United States
Duration: 25 Jun 201328 Jun 2013

Conference

Conference2013 5th Workshop on hyperspectral image and signal processing : evolution in remote sensing (WHISPERS)
CountryUnited States
CityGainesville
Period25/06/1328/06/13

Keywords

  • hyper-spectral images
  • principal
  • component analysis
  • deblurring
  • total variation
  • principal component analysis
  • noise reduction
  • image restoration
  • geophysical image processing
  • hyperspectral imaging
  • image segmentation

Fingerprint Dive into the research topics of 'Hyperspectral image deblurring with PCA and total variation'. Together they form a unique fingerprint.

  • Cite this

    Liao, W., Goossens, B., Aelterman, J., Luong, H., Pizurica, A., Wouters, N., ... Philips, W. (2013). Hyperspectral image deblurring with PCA and total variation. Paper presented at 2013 5th Workshop on hyperspectral image and signal processing : evolution in remote sensing (WHISPERS), Gainesville, United States. https://doi.org/10.1109/WHISPERS.2013.8080664