Euclidean Distance-Based Hand Gesture Recognition for Virtual Mouse Control in Human-Computer Interaction

This study proposes a virtual mouse system that employs hand gesture recognition, utilizing the Euclidean distance algorithm to achieve precise gesture detection. The system was developed using the Rapid Application Development (RAD) methodology, with Python, MediaPipe Hands, and OpenCV facilitating real-time hand tracking and image processing. It recognizes seven distinct hand gestures mapped to essential mouse functions, including left click, right click, hold/drag, cursor movement, double click, and scrolling. Extensive testing was conducted to evaluate system performance across different devices, webcam resolutions, and environmental conditions. The results indicate that high-resolution webcams (480p and 1080p) significantly enhance accuracy, while low-resolution cameras (144p) fail to provide reliable gesture detection. Furthermore, the system demonstrates optimal functionality under well-lit conditions but struggles in low-light environments. Background interference had minimal impact, as the system effectively prioritized and tracked the user’s hand. Findings suggest that the effectiveness of the virtual mouse is highly dependent on hardware quality and environmental factors. High-end devices with superior webcams ensure improved responsiveness and accuracy. These insights underscore the necessity of using optimal conditions to achieve reliable gesture-based human-computer interaction. The proposed system offers a promising alternative input method for accessibility and hands-free computing, paving the way for future enhancements in gesture-controlled interfaces.