Indian Sign Language Classification using Advancement of CNN

In our increasingly digital world, addressing the communication needs of Indian Sign Language (ISL) users is a pressing challenge. This research focuses on developing an advanced Convolutional Neural Network (CNN) model for real-time ISL gesture classification. The paper aims to design a CNN-based system capable of recognizing and classifying a wide range of ISL gestures, involving convolutional layers, pooling layers, and fully connected layers. The primary objective is to provide an innovative and efficient solution for real-time ISL gesture classification, empowering the hearing-impaired community. The novelty lies in leveraging CNNs to decode the complexity of ISL gestures in various media, a milestone in the field. Our analysis indicates that the proposed methodology outperforms existing techniques with an accuracy of 91%, a precision of 92%, a recall of 90%, and an F-measure of 91%, demonstrating the robustness and effectiveness of the CNN-based approach. In summary, this research presents a pioneering solution for real-time ISL gesture recognition, utilizing CNNs to offer efficient communication tools, enhancing accessibility and inclusivity for the hearing-impaired.