Handwriting to Digital Text Translation Using a Self-Powered Triboelectricity-Induced Piezoelectric Writing Pad...

Area of Science:

• Materials Science
• Electrical Engineering
• Computer Science

Background:

• Human-machine interfaces require integrated energy, communication, and computing.
• Traditional writing methods lack digital efficiency.
• Need for intuitive yet powerful input systems.

Purpose of the Study:

• Introduce a triboelectricity-induced piezoelectric writing pad (TPWP) for handwriting to digital text conversion.
• Evaluate the TPWP's effectiveness in advanced letter detection.
• Demonstrate a novel human-machine interface for enhanced productivity.

Main Methods:

• Developed a micropatterned TPWP for signal acquisition.
• Collected handwritten signals for English letters A, B, C, and D from multiple individuals.
• Applied deep learning algorithms, including convolutional neural networks (CNNs), for signal processing and classification.
• Filtered and trained handwritten patterns over 250 epochs, transforming time-varying signals into spectrograms.

Main Results:

• The TPWP successfully captured unique handwritten signals, differentiating between individuals and recognizing letter patterns.
• Achieved high classification accuracy: 100% for letters A and B, 99% for letters C and D.
• The overall model precision reached 99% for handwriting recognition.
• Demonstrated the potential of 3D printed textured TPWP for practical applications.

Conclusions:

• The developed TPWP offers an effective solution for converting handwritten input into digital text.
• The integration of triboelectricity, piezoelectricity, and deep learning (CNNs) significantly enhances letter detection accuracy.
• The TPWP technology holds promise for future human-machine interfaces, improving information management and user productivity.