MAP model order selection rule for 2-D sinusoids in white noise

Mark Kliger, Joseph M. Francos

Research output: Contribution to journalArticlepeer-review

10 Scopus citations

Abstract

We consider the problem of jointly estimating the number as well as the parameters of two-dimensional (2-D) sinusoidal signals, observed in the presence of an additive white Gaussian noise field. Existing solutions to this problem are based on model order selection rules and are derived for the parallel one-dimensional (1-D) problem. These criteria are then adapted to the 2-D problem using heuristic arguments. Employing asymptotic considerations, we derive a maximum a posteriori (MAP) model order selection criterion for jointly estimating the parameters of the 2-D sinusoids and their number. The proposed model order selection rule is strongly consistent. As an example, the model order selection criterion is applied as a component in an algorithm for parametric estimation and synthesis of textured images.

Original languageEnglish
Pages (from-to)2563-2575
Number of pages13
JournalIEEE Transactions on Signal Processing
Volume53
Issue number7
DOIs
StatePublished - 1 Jul 2005

Keywords

  • 2-D parameter estimation
  • 2-D sinusoids
  • Maximum a posteriori estimation
  • Model order selection
  • Random fields
  • Texture parametric model

Fingerprint

Dive into the research topics of 'MAP model order selection rule for 2-D sinusoids in white noise'. Together they form a unique fingerprint.

Cite this