Statistics and Its Interface

Volume 3 (2010)

Number 3

Singular spectrum analysis for image processing

Pages: 419 – 426

DOI: http://dx.doi.org/10.4310/SII.2010.v3.n3.a14

Authors

Licesio J. Rodríguez-Aragón (Department of Mathematics, University of Castilla-La Mancha, Ciudad Real, Spain)

Anatoly Zhigljavsky (School of Mathematics, Cardiff University, Cardiff, United Kingdom)

Abstract

A technique of image processing based on application of the Singular Spectrum Analysis (SSA) is discussed and illustrated on the problem of denoising the celebrated ‘Lena’ image corrupted with noise. Also, SSA-based distances between two images are introduced and suggested for a possible use in the face verification problem.

Keywords

singular spectrum analysis, time series analysis, distance between images, face verification

2010 Mathematics Subject Classification

Primary 37M10, 62M10. Secondary 94A08.

Full Text (PDF format)