Hand Gesture Recognition System Using Thermal Images: A Systematic Literature Review

Hand gesture detection is a pivotal technology in advancing human-computer interaction, offering intuitive and touch-free control across various applications. This paper presents the development of a robust hand gesture detection system utilizing Convolutional Neural Networks (CNNs), leveraging their ability to automatically extract and learn spatial hierarchies of features from input images. The proposed system processes hand images and accurately classifies various gestures, addressing challenges associated with traditional recognition methods that require extensive feature engineering and complex pre-processing. The primary objective of this research is to enable seamless interaction between users and devices through hand gestures, enhancing accessibility in domains such as gaming, assistive technology, and virtual reality. A labeled dataset of hand gesture images is used to train and optimize the CNN model to achieve high accuracy and low latency in real-time predictions. The performance of the system is further enhanced by implementing efficient CNN architectures and optimizing the model for low- power devices, thereby expanding its practical applications. The proposed system demonstrates the potential to provide a more natural, flexible, and immersive interaction experience across diverse digital environments.