Continuous Arm Gesture Recognition Based On Natural Features And Logistic Regression Using Arduino

This project presents a continuous arm gesture recognition system using natural hand movements, implemented through five flex sensors embedded in a glove. The system utilizes an Arduino microcontroller to process the analog values from the flex sensors, which capture finger bending patterns corresponding to specific hand gestures. A logistic regression model is employed to classify these gestures into predefined commands. Upon recognizing a gesture, the system triggers corresponding voice outputs using a voice module and speaker, providing audio feedback. Additionally, an LCD display connected to the Arduino visually shows the command being executed. This system enables intuitive, hands-on control and can be applied in assistive technologies, human-computer interaction, and automation control scenarios.