Hidden Markov models with spectral features for 2D shape recognition

Author

Cai, Jinhai ; Liu, Zhi-Qiang

Author_Institution

Sch. of Comput. Sci. & Software Eng., Queensland Univ. of Technol., Brisbane, Qld., Australia

Volume

23

Issue

12

fYear

2001

fDate

12/1/2001 12:00:00 AM

Firstpage

1454

Lastpage

1458

Abstract

We present a technique using Markov models with spectral features for recognizing 2D shapes. We analyze the properties of Fourier spectral features derived from closed contours of 2D shapes and use these features for 2D pattern recognition. We develop algorithms for reestimating parameters of hidden Markov models. To demonstrate the effectiveness of our models, we have tested our methods on two image databases: hand-tools and unconstrained handwritten numerals. We are able to achieve high recognition rates of 99.4 percent and 96.7 percent without rejection on these two sets of image data, respectively

Keywords

feature extraction; hidden Markov models; image recognition; parameter estimation; probability; 2D pattern recognition; 2D shape recognition; Fourier spectral features; closed contours; hand-tools; hidden Markov models; image databases; spectral features; unconstrained handwritten numerals; Deformable models; Handwriting recognition; Hidden Markov models; Image databases; Image recognition; Pattern analysis; Pattern recognition; Shape; Testing; Uncertainty;

fLanguage

English

Journal_Title

Pattern Analysis and Machine Intelligence, IEEE Transactions on

Publisher

ieee

ISSN

0162-8828

Type

jour

DOI

10.1109/34.977569

Filename

977569