Kinect-based mid-air handwritten digit recognition using multiple segments and scaled coding

Author

Fu-An Huang ; Chung-Yen Su ; Tsai-Te Chu

Author_Institution

Dept. of Appl. Electron. Technol., Nat. Taiwan Normal Univ., Taipei, Taiwan

fYear

2013

fDate

12-15 Nov. 2013

Firstpage

694

Lastpage

697

Abstract

Human-computer interaction has attracted much attention in recent years, especially in the field of entertainment. In our previous study, we presented an effective method of Kinect-based mid-air handwritten digit recognition for a potential application to TV remote controllers. However, its recognition accuracy is only about 86.7%. In this study, we propose an improved method, based on multiple segments and scaled coding. Experimental results show that the proposed method can elevate the accuracy up to 94.6%.

Keywords

feature extraction; gesture recognition; handwriting recognition; human computer interaction; image coding; support vector machines; Kinect-based midair handwritten digit recognition; TV remote controllers; human-computer interaction; multiple segments; scaled coding; Accuracy; Cameras; Conferences; Encoding; Handwriting recognition; Shape; Training; Kinect; hand gesture recognition; handwritten digit recognition; human-computer interaction; support vector machine;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Signal Processing and Communications Systems (ISPACS), 2013 International Symposium on

Conference_Location

Naha

Print_ISBN

978-1-4673-6360-0

Type

conf

DOI

10.1109/ISPACS.2013.6704638

Filename

6704638