Hand Gesture Detection using Deep Learning with YOLOv5

Hand gesture recognition has become a significant technological advancement in assistive communication, offering a reliable means of interaction for individuals with hearing and speech impairments. This research introduces an intelligent gesture detection system powered by YOLOv5, a leading object detection model, to enable accurate and real-time recognition of Indian Sign Language (ISL) gestures. The system effectively handles diverse environmental conditions and user-specific variations using an extensive and well-annotated dataset. The methodology encompasses essential stages such as image preprocessing, data augmentation, and feature extraction to optimize model performance. Furthermore, a user-friendly web interface allows users to upload images for gesture detection, with corresponding text and audio outputs generated using a text-to-speech module. Designed for seamless scalability, the system can accommodate additional gestures and languages, making it a versatile solution for educational institutions, healthcare facilities, and public service sectors. By fostering greater inclusivity and accessibility, this approach represents a step forward in empowering the hearing-impaired community through innovative deep-learning applications.