DocumentCode :
478276
Title :
Self-Defined Gesture Recognition on Keyless Handheld Devices using MEMS 3D Accelerometer
Author :
Zhang, Shiqi ; Yuan, Chun ; Zhang, Yan
Author_Institution :
Grad. Sch. at Shenzhen, Tsinghua Univ., Shenzhen
Volume :
4
fYear :
2008
fDate :
18-20 Oct. 2008
Firstpage :
237
Lastpage :
241
Abstract :
A novel gesture recognition method is introduced in this paper, and it is designed for keyless handheld devices specially. This terminal contains a three dimensional MEMS accelerometer and a Single Chip Microcontroller, which enable kinds of gesture recognitions, such as inclination, swaying and handwritten Arabic numerals. In this way, we could operate a handheld device without any key just by gesture recognition. It is noted that the same track of a numeral may come from different strokes, for example, Arabic numeral "0" could be written clockwise or anti-clockwise, so self-defined gesture recognition is adopted in Arabic numeral recognition, by which every user could "write" in their own way. Using keyless handheld devices, young guys could get more fun and old people or blind persons could get more convenience. A large number of experiments have been done to verify its reliability and the correct rates (C.R.) prove this terminal could satisfy users\´ basic requirements.
Keywords :
accelerometers; gesture recognition; microsensors; notebook computers; MEMS 3D accelerometer; correct rates; handwritten Arabic numerals; inclination; keyless handheld devices; self-defined gesture recognition; single chip microcontroller; swaying; Accelerometers; Cameras; Clocks; Handheld computers; Handwriting recognition; Man machine systems; Microcontrollers; Micromechanical devices; Personal digital assistants; Space technology; Accelerometer; Arabic Numeral Recognition; Human-Machine Interface; Keyless Handheld Devices; MEMS; Self-Defined Gesture Recognition;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Natural Computation, 2008. ICNC '08. Fourth International Conference on
Conference_Location :
Jinan
Print_ISBN :
978-0-7695-3304-9
Type :
conf
DOI :
10.1109/ICNC.2008.387
Filename :
4667282
Link To Document :
بازگشت