Smart Hand Device Gesture Recognition with Dynamic Time-Warping Method

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 %.