Robust L1 PCA and Its Application in Image Denoising

Junbin Gao, Paul Kwan, Yi Guo

Research output: Book chapter/Published conference paperConference paper

17 Downloads (Pure)

Abstract

The so-called robust L1 PCA was introduced in our recent work1 based on the L1 noise assumption. Due tothe heavy tail characteristics of the L1 distribution, the proposed model has been proved much more robust against data outliers. In this paper, we further demonstrate how the learned robust L1 PCA model can be used to denoise image data.
Original languageEnglish
Title of host publicationMIPPR 2007
Subtitle of host publicationAutomatic target recognition and image analysis; and multispectral image acquisition
EditorsTianxu Zhang, Carl A Nardell, Duane D Smith, Hangqing Lu
Place of PublicationWashington, USA
PublisherSPIE
Pages67860T
Volume6786
DOIs
Publication statusPublished - 2007
EventInternational Symposium on Multispectrum Image Processing and Pattern Recognition (MIPPR) - Wuhan, China, China
Duration: 15 Nov 200717 Nov 2007

Conference

ConferenceInternational Symposium on Multispectrum Image Processing and Pattern Recognition (MIPPR)
CountryChina
Period15/11/0717/11/07

Fingerprint Dive into the research topics of 'Robust L1 PCA and Its Application in Image Denoising'. Together they form a unique fingerprint.

  • Cite this

    Gao, J., Kwan, P., & Guo, Y. (2007). Robust L1 PCA and Its Application in Image Denoising. In T. Zhang, C. A. Nardell, D. D. Smith, & H. Lu (Eds.), MIPPR 2007: Automatic target recognition and image analysis; and multispectral image acquisition (Vol. 6786, pp. 67860T). SPIE. https://doi.org/10.1117/12.774719