Machine learning-enhanced gesture recognition through impedance signal analysis.

Area of Science:

• Human-computer interaction
• Biomedical engineering
• Machine learning

Background:

• Gesture recognition is vital for virtual reality, healthcare, and human-computer interaction.
• Current methods require enhanced precision to meet growing demands.
• Innovative approaches are needed to improve accuracy and robustness.

Purpose of the Study:

• To present a novel approach combining Impedance Signal Spectrum Analysis (ISSA) with machine learning for improved gesture recognition precision.
• To evaluate the performance of various machine learning algorithms using ISSA features.
• To demonstrate the robustness and adaptability of the proposed methodology.

Main Methods:

• A diverse dataset of predefined gestures was collected from five participants.
• Impedance Signal Spectrum Analysis (ISSA) was employed to extract relevant features.
• Machine learning algorithms including KNN, GBM, NB, LR, RF, and SVM were utilized for classification.

Main Results:

• The machine learning models demonstrated notable precision in gesture recognition.
• Logistic Regression (LR) achieved the highest accuracy at 89%.
• Other algorithms like KNN and GBM reached 86% accuracy, RF and SVM 87%, and NB 84%.

Conclusions:

• Impedance features are crucial for refining gesture recognition accuracy.
• The ISSA-enhanced machine learning model shows significant potential for various applications.
• The model's adaptability under different conditions highlights its broad applicability.