HMM-Based Online Handwriting Recognition System for Telugu Symbols

Author

Babu, V.J. ; Prasanth, L. ; Sharma, R.R. ; Bharath, A.

Author_Institution

Sri Sathya Sai Inst. of Higher Learning, Puttaparthy

Volume

1

fYear

2007

fDate

23-26 Sept. 2007

Firstpage

63

Lastpage

67

Abstract

In this paper we present an online handwritten symbol recognition system for Telugu, a widely spoken language in India. The system is based on hidden Markov models (HMM) and uses a combination of time-domain and frequency-domain features. The system gives top-1 accuracy of 91.6% and top-5 accuracy of 98.7% on a dataset containing 29,158 train samples and 9,235 test samples. We also introduce a cost-effective and natural data collection procedure based on ACECADreg Digimemoreg and describe its usage in building a Telugu handwriting dataset.

Keywords

frequency-domain analysis; handwriting recognition; hidden Markov models; time-domain analysis; Telugu symbols; frequency-domain analysis; hidden Markov models; online handwriting recognition system; time-domain analysis; Handwriting recognition; Hidden Markov models; Keyboards; Natural languages; Personal digital assistants; Shape; Support vector machines; System testing; Time domain analysis; User interfaces;

fLanguage

English

Publisher

ieee

Conference_Titel

Document Analysis and Recognition, 2007. ICDAR 2007. Ninth International Conference on

Conference_Location

Parana

ISSN

1520-5363

Print_ISBN

978-0-7695-2822-9

Type

conf

DOI

10.1109/ICDAR.2007.4378676

Filename

4378676