Dynamic Fingerspelling Recognition using Geometric and Motion Features

Author

Goh, P. ; Holden, E. -J.

Author_Institution

Sch. of Comput. Sci. & Software Eng., Western Australia Univ., Crawley, WA, Australia

fYear

2006

Firstpage

2741

Lastpage

2744

Abstract

This paper presents the Australian sign language (Auslan) fingerspelling recognizer (APR): a system capable of recognizing signs consisting of Auslan manual alphabet letters from video sequences. The APR system uses a combination of geometric features and motion features based on optical flow which are extracted from video sequences. The sequence of features are then classified using hidden Markov models (HMMs). Tests using a vocabulary of twenty signed words showed the system could achieve 97% accuracy at the letter level and 88% at the word level by using a finite state grammar network and embedded training.

Keywords

feature extraction; hidden Markov models; image classification; image motion analysis; image recognition; video signal processing; Auslan fingerspelling recognizer; Auslan manual alphabet letters; Australian sign language; dynamic fingerspelling recognition; feature extraction; features sequence classification; finite state grammar network; geometric features; hidden Markov models; image segmentation; motion features; optical flow; signs recognition; video sequences; vocabulary; Australia; Feature extraction; Geometrical optics; Handicapped aids; Hidden Markov models; Image motion analysis; Shape; Skin; Testing; Video sequences; Feature Extraction; Hidden Markov Models; Image Segmentation; Motion Analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Image Processing, 2006 IEEE International Conference on

Conference_Location

Atlanta, GA

ISSN

1522-4880

Print_ISBN

1-4244-0480-0

Type

conf

DOI

10.1109/ICIP.2006.313114

Filename

4107136