Classification Of Indian Sign Language Characters Utilizing Convolutional Neural Networks And Transfer Learning Models With Different Image Processing Techniques

Abstract

In terms of visual-spatial modality, the sign language is considered to be a natural as well as a full-fledged language. It has all of the linguistic characteristics of spoken language (from phonology to syntax). Sign language is a form of communication in which the hands are used instead of words. It uses a variety of signs to convey thoughts and concepts. For ISL static character recognition, we propose a Convolutional Neural Network (CNN) architecture in this paper. Comparison of different feature extraction techniques tested on CNN architecture is done in this particular paper. We hand-crafted the dataset used to train the CNN model in order to come as near to the real-life scenario in which the model’s viability would be assessed as possible. The proposed method was successfully implemented with a 99.90 percent accuracy, which is better than the majority of currently available methods.