Rancang Bangun Sistem Virtual Mouse Dengan Hand Gesture Recognition Menggunakan Convolutional Neural Network

Abstract

The development of human-computer interaction technology continues to grow, especially in the context of hand gesture recognition for device control applications. One interesting application is designing a virtual mouse system based on hand gesture recognition using a Convolutional Neural Network (CNN). This research was conducted to overcome the limitations of conventional input devices and utilize modern technology to create a more sophisticated and adaptive control system. This research focuses on developing a system that is able to identify and interpret hand gestures accurately in a real-time environment. The proposed system utilizes an RGB camera to capture images of the user's hand gestures, which are then processed by a CNN that has been pre-trained to recognize different gestures. Experiments were carried out using a gesture dataset that included variations in pose and hand orientation to evaluate the system's performance in recognizing gestures with a sufficient level of accuracy. The research results show that the system can successfully control the movement of the cursor on the computer screen with high accuracy, validating the potential practical application of this technology in improving the user experience in computer interactions. The implication of this research is a contribution to the development of more intuitive and effective user interfaces through the integration of hand gesture technology in virtual mouse applications.