Investigating hidden Markov models´ capabilities in 2D shape classification

Author

Bicego, Manuele ; Murino, Vittorio

Author_Institution

Dip. di Inf., Univ. di Verona, Italy

Volume

26

Issue

2

fYear

2004

Firstpage

281

Lastpage

286

Abstract

In this paper, Hidden Markov Models (HMMs) are investigated for the purpose of classifying planar shapes represented by their curvature coefficients. In the training phase, special attention is devoted to the initialization and model selection issues, which make the learning phase particularly effective. The results of tests on different data sets show that the proposed system is able to accurately classify objects that were translated, rotated, occluded, or deformed by shearing, also in the presence of noise.

Keywords

hidden Markov models; image classification; learning (artificial intelligence); object recognition; probability; 2D shape classification; HMM investigation; hidden Markov model investigation; learning phase; model selection; noise; object classification; planar shapes; shearing; training phase; two dimensional shape classification; Algorithm design and analysis; Computer vision; Face detection; Hidden Markov models; Noise shaping; Object recognition; Shape; Shearing; System testing; Topology; Algorithms; Artificial Intelligence; Computer Graphics; Computer Simulation; Image Enhancement; Image Interpretation, Computer-Assisted; Information Storage and Retrieval; Markov Chains; Models, Statistical; Numerical Analysis, Computer-Assisted; Pattern Recognition, Automated; Reproducibility of Results; Sensitivity and Specificity; Signal Processing, Computer-Assisted; Subtraction Technique;

fLanguage

English

Journal_Title

Pattern Analysis and Machine Intelligence, IEEE Transactions on

Publisher

ieee

ISSN

0162-8828

Type

jour

DOI

10.1109/TPAMI.2004.1262200

Filename

1262200