SinoGreenTech Academic Portal
Official PDF TranslationSCIENCE CHINA Materials

Wearable Interactive System with Uncoded Gesture Recognition Logic Enabled by Deep Learning

Authors: LIU Liangjie; LI Lingxiao; LIN Yibin; CHEN Wandi; WENG Shuchen; SUN Lei; ZHOU Xiongtu; GUO Tailiang; WU Chaoxing; ZHANG Yongai

DOI: 10.1007/s40843-025-3553-7Status: Verified Translated Edition
Sponsored AdvertisementAd Placement Area
reCAPTCHA Bot Shield Active

Preparing Secure Academic Download

Verifying human reader & generating high-resolution document...

Verifying Document Integrity15s remaining
← Back to Article
Protected by Google reCAPTCHA v3.PrivacyTerms
Sponsored ContentAdSense In-Feed Ad Slot

Key Findings in This Report

• • The PS-TENG-based finger ring achieves recognition accuracies of 98.21% for English letters, 96.87% for Arabic numerals, and 96.44% for Chinese characters, demonstrating high fidelity for text input without predefined gesture encoding. • • Open-circuit voltage scales linearly with finger bending angle: 0.45 V at 30° to 3.15 V at 90°, enabling precise gesture differentiation and reliable threshold-based control. • • Four distinct gestures are defined by unique sensor activation patterns (e.g., Gesture 1: high-amplitude impulse from Sensor-I1 and low-amplitude from Sensor-I2), enabling robust wireless gaming control with directional commands. • • The system integrates a signal conditioning circuit stabilizing outputs to 0–3.3 V, ensuring compatibility with standard microcontrollers and Bluetooth modules for wireless data transmission.