Finger Gesture Recognition Based on 3D-Accelerometer and 3D-Gyroscope

Gesture-based interaction, as a natural way for human-computer interaction, has a wide range of applications in the ubiquitous computing environment. The latest research findings reveal that user’s arm and hand gestures are likely to be identified with ease using the motion sensors worn on the wrist, but it is not clear how much of user’s finger gestures can be recognized. This paper presents a method, which is capable of recognizing the bending of fingers, based on input signals from the 3D-accelerometer and 3D-gyroscope worn on the wrist. Features from Time-domain and Frequency-domain are extracted. Gestures are recognized by five classifiers, and the recognition results were then compared with each other. In this paper, maximal information coefficient is adopted for examining the effect of features on the gesture classification. Besides, we work out a faster calculation method, which is based on the features of top 30 maximal information coefficient. Our present results can be widely applied for medical rehabilitation and consumer electronics control based on gesture interaction.