CNN Approach for Static Hand Gesture Recognition in Indian Sign Language

Abstract

Indian Sign Language (ISL) plays a crucial role in bridging the communication gap between individuals who are hearing-impaired and the broader society. However, limited research and technological solutions exist for recognising ISL, especially in regional contexts. This paper presents a deep learning-based approach for recognising static hand gestures that represent the ISL alphabet (A–Z). A Convolutional Neural Network (CNN) model is trained on a publicly available dataset containing labelled hand sign images. The system classifies input images into corresponding alphabetic characters with high accuracy, providing a real-time, low-cost, and accessible solution. The aim is to support inclusive human-computer interaction and assistive technology for the hearing-impaired community. The experimental results demonstrate the effectiveness of the proposed model, making it suitable for educational tools, basic communication aids, and future integration into mobile or web applications.