Tracing the Motion of Finger Joints for Gesture Recognition via Sewing RGO-Coated Fibers onto a Textile Glove

In this study, we have designed and fabricated a flexible and low-cost data glove and then successfully realized the accurate gesture recognition. By sewing the reduced graphene oxide (RGO)-coated fiber that is prepared in a simple method onto a textile glove, we have manufactured a data glove that has been used to monitor the motion of ten finger joints from one hand. In particular, we accurately realize the static and dynamic gesture recognition via the data glove, respectively. Experimental results on the classification of ten static gestures representing digits from “0” to “9” show the good stability and repeatability of the data glove, in which the recognition accuracy achieves 98.5% in 2000 test samples. In addition, the recognition of nine Chinese Sign Language (CSL) words is successfully implemented, and the recognition accuracy reaches 98.3% in 180 test samples, which demonstrates the good real-time monitoring ability and applicability of the dynamic gesture scenarios of the as-prepared data glove.