HMM-based on-line multi-stroke sketch recognition

Author

Jiang, Wei ; SUN, ZHENG-XING

Author_Institution

State Key Lab. for Novel Software Technol., Nanjing Univ., China

Volume

7

fYear

2005

fDate

18-21 Aug. 2005

Firstpage

4564

Abstract

This paper describes a new approach for on-line multi-stroke sketch recognition. The approach is based on hidden Markov model (HMM). Sketches are modeled to HMM chains, and strokes are mapped to different HMM states. The proposed approach introduces a new method to determine HMM state-number, based on which an adaptive HMM sketch recognizer is constructed. A combined feature based on curvature, velocity and geometrical character of stroke for sketch recognition is also proposed to improve recognition accuracy. Finally, the experiments prove the effectiveness and efficiency of the proposed approach.

Keywords

character recognition; feature extraction; hidden Markov models; image recognition; HMM state-number; Markov chains; adaptive hidden Markov model sketch recognition; feature extraction; geometrical character; online multistroke sketch recognition; sketch modeling; stroke curvature; stroke mapping; stroke velocity; Character recognition; Engines; Handwriting recognition; Hidden Markov models; Shape; Speech recognition; Sun; Support vector machine classification; Support vector machines; Topology; Adaptive Hidden Markov model; Multi-stroke; Sketch Recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on

Conference_Location

Guangzhou, China

Print_ISBN

0-7803-9091-1

Type

conf

DOI

10.1109/ICMLC.2005.1527743

Filename

1527743