TY - GEN
T1 - Smart Hand Device Gesture Recognition with Dynamic Time-Warping Method
AU - Lee, Boon Giin
AU - Tran, Viet Cuong
AU - Chong, Teak Wei
N1 - Publisher Copyright:
� 2017 Association for Computing Machinery.
PY - 2017/12/20
Y1 - 2017/12/20
N2 - In this paper, we present a smart wearable hand-gesture recognition system based on the movement of the hand and fingers. The proposed smart wearable system is built using the fewest sensors necessary for gesture recognition. Thus, motion sensors are placed on the thumb and index finger to detect finger motions. Another sensor is placed on the back of the hand to measure hand movement. A total of six gestures are analyzed via hand and finger movement using a dynamic time-warping method. Gestures include swipe right, swipe left, zoom in, zoom out, rotate left, and rotate right. An Android-based mobile device application simulator measures gesture recognition effectiveness. Gestures are analyzed using a trained recognition model. Once a gesture is detected, it is transmitted to the mobile application via Bluetooth low energy communication. Received gestures then trigger corresponding commands, as specified in the mobile application. The proposed smart wearable system can detect gestures at mean accuracy of 93.19 %.
AB - In this paper, we present a smart wearable hand-gesture recognition system based on the movement of the hand and fingers. The proposed smart wearable system is built using the fewest sensors necessary for gesture recognition. Thus, motion sensors are placed on the thumb and index finger to detect finger motions. Another sensor is placed on the back of the hand to measure hand movement. A total of six gestures are analyzed via hand and finger movement using a dynamic time-warping method. Gestures include swipe right, swipe left, zoom in, zoom out, rotate left, and rotate right. An Android-based mobile device application simulator measures gesture recognition effectiveness. Gestures are analyzed using a trained recognition model. Once a gesture is detected, it is transmitted to the mobile application via Bluetooth low energy communication. Received gestures then trigger corresponding commands, as specified in the mobile application. The proposed smart wearable system can detect gestures at mean accuracy of 93.19 %.
KW - Dynamic time warping
KW - Gesture recognition
KW - Mobile application
KW - Motion sensors
KW - Wearable system
UR - http://www.scopus.com/inward/record.url?scp=85046664569&partnerID=8YFLogxK
U2 - 10.1145/3175684.3175697
DO - 10.1145/3175684.3175697
M3 - Conference contribution
AN - SCOPUS:85046664569
T3 - ACM International Conference Proceeding Series
SP - 216
EP - 219
BT - BDIOT 2017 - Proceedings of the International Conference on Big Data and Internet of Thing
PB - Association for Computing Machinery
T2 - 2017 International Conference on Big Data and Internet of Thing, BDIOT 2017
Y2 - 20 December 2017 through 22 December 2017
ER -