A novel electromyography (EMG) based classification approach for Arabic handwriting

Author

Lansari, Azzedine ; Bouslama, Faouzi ; Khasawneh, Mohammed ; Al-Rawi, Ali

Author_Institution

Coll. of Inf. Syst., Zayed Univ., Abu Dhabi, United Arab Emirates

Volume

3

fYear

2003

fDate

20-24 July 2003

Firstpage

2193

Abstract

In this paper, a novel classification approach for handwritten Arabic characters is proposed. Features for classification are extracted from electromyographic (EMG) signals detected on two forearm muscles. Noise cancellations in conjunction with a process parameter estimator for feature identification are proposed. Neural networks using a potentially damped least mean squared algorithm is used at the classification stage. The proposed new classification technique is used on handwritten Arabic characters.

Keywords

electromyography; feature extraction; handwritten character recognition; least mean squares methods; neural nets; parameter estimation; pattern classification; EMG; EMG signals detection; damped least mean squared algorithm; electromyography based classification; feature extraction; feature identification; forearm muscles; handwritten Arabic characters; handwritten character recognition; neural networks; noise cancellations; process parameter estimator; Biological neural networks; Biological system modeling; Educational institutions; Electromyography; Humans; Muscles; Noise cancellation; Signal detection; Signal processing; Wrist;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 2003. Proceedings of the International Joint Conference on

ISSN

1098-7576

Print_ISBN

0-7803-7898-9

Type

conf

DOI

10.1109/IJCNN.2003.1223748

Filename

1223748