Wearable System for Frequency-Selective Surface-Based Sign Language Gesture Recognition
Wearables antennas, gesture recognition, dual polarized antenna.
This thesis proposes the development of a radiofrequency (RF)-based device for the recognition of static hand gestures, aiming to support the inclusion of deaf and hardof-hearing individuals. Based on a comprehensive review of the state of the art and the identification of existing research gaps, an experimental method was designed using variations in the S21 parameter, which is highly sensitive to subtle environmental changes. The adopted structure—a frequency selective surface (FSS)—was initially analyzed in isolation and subsequently integrated into a glove, allowing users to perform gestures from the Brazilian Sign Language (LIBRAS). Experimental results demonstrated the system’s effectiveness in distinguishing gestures with similar physical characteristics, a conclusion supported by statistical analysis. A robustness analysis further validated the reliability of the proposed method and device. The results highlight the potential of the approach for gesture recognition applications, human–machine interaction, gaming interfaces, and augmented reality.