Development of a Low-Cost Wearable Data Glove for Capturing Finger Joint Angles

Area of Science:

• Wearable technology
• Biomedical engineering
• Human-computer interaction

Background:

• Accurate finger joint angle capture is crucial for human-computer interaction and hand function assessment.
• Existing methods may lack precision or wearability for real-time applications.

Purpose of the Study:

• To develop and evaluate a novel wearable data glove for capturing finger joint angles.
• To assess the precision and accuracy of the glove in measuring joint angles and recognizing gestures.

Main Methods:

• A data glove equipped with 14 custom-designed optical sensing units (grating strip and detector) was developed.
• Sensing unit response, calibration experiments, grasping tests, and static digital gesture recognition were performed.
• A generalized regression neural network (GRNN) was employed for gesture recognition.

Main Results:

• Calibration experiments demonstrated a comprehensive precision of 1.67% for measuring joint angles on a wooden finger model.
• The data glove achieved a 99% recognition accuracy for static digital gestures when combined with a GRNN.

Conclusions:

• The proposed wearable data glove effectively captures finger joint angles with high precision.
• The system shows significant potential for applications in human-computer interaction and hand function evaluation.