Alphabet Handwriting Recognition: From Wood-Framed Hydrogel Arrays Design to Machine Learning Decoding

Area of Science:

• Materials Science
• Computer Science
• Biotechnology

Background:

• Handwriting recognition systems require integrated hardware and software, posing challenges for sustainable design.
• Developing accessible computing networks with sustainable materials for input devices is a significant research gap.

Purpose of the Study:

• To create a flexible, electrically conductive handwriting input panel using wood-derived hydrogel.
• To enable alphabet handwriting recognition via machine learning on an accessible computing network.

Main Methods:

• Sequential modification of wood: lignin extraction-refill, polypyrrole coating, and polyacrylic acid filling.
• Fabrication of a 5x5 wood-derived hydrogel array for signal matrix generation.
• Implementation of expertise-based feature engineering and low-complexity algorithms for handwritten recognition.

Main Results:

• A functional wood-derived hydrogel array capable of capturing handwriting signals was successfully developed.
• Efficient alphabet handwriting recognition was achieved using machine learning with simplified algorithms.
• The system demonstrated boosted model performance through strategic feature engineering.

Conclusions:

• Wood-derived hydrogels offer a sustainable and flexible material for handwriting input panels.
• Expertise-based feature engineering and simplified algorithms enhance machine learning-based handwriting recognition.
• The developed technology holds potential for wearable devices and healthcare applications.