Asymptotic performance analysis of Bayesian target recognition

Author

Grenander, Ulf ; Srivastava, Anuj ; Miller, Michael I.

Author_Institution

Div. of Appl. Math., Brown Univ., Providence, RI, USA

Volume

46

Issue

4

fYear

2000

fDate

7/1/2000 12:00:00 AM

Firstpage

1658

Lastpage

1665

Abstract

This article investigates the asymptotic performance of Bayesian target recognition algorithms using deformable-template representations. Rigid computer-aided design (CAD) models represent the underlying targets; low-dimensional matrix Lie-groups (rotation and translation) extend them to particular instances. Remote sensors observing the targets are modeled as projective transformations, converting three-dimensional scenes into random images. Bayesian target recognition corresponds to hypothesis selection in the presence of nuisance parameters; its performance is quantified as the Bayes´ error. Analytical expressions for this error probability in small noise situations are derived, yielding asymptotic error rates for exponential error probability decay

Keywords

Bayes methods; CAD; Lie groups; clutter; error statistics; image recognition; image representation; matrix algebra; random processes; remote sensing; 3D scenes; Bayes error; Bayesian target recognition algorithms; asymptotic error rates; asymptotic performance analysis; clutter; computer-aided design models; deformable-template representations; exponential error probability decay; hypothesis selection; low-dimensional matrix Lie-groups; noise; nuisance parameters; performance; projective transformations; random images; remote sensors; rotation; translation; Bayesian methods; Design automation; Error analysis; Error probability; Image converters; Layout; Matrix converters; Performance analysis; Remote sensing; Target recognition;

fLanguage

English

Journal_Title

Information Theory, IEEE Transactions on

Publisher

ieee

ISSN

0018-9448

Type

jour

DOI

10.1109/18.850712

Filename

850712