Capturing Human Gestures for Sign Language Detection: An Advance Technique for Detection to Help Deprived Section of Society

This manuscript aims to build a machine learning model which helps people who are not able to speak and use sign language for communication. The proposed model is able to classify various hand gestures used in sign language communication. In this model, the author team has trained convolutional neural networks with image dataset of sign language and also testing is performed with completely different dataset. The authors have applied concept of deep learning convolutional Neural Network (CNN).To develop this model the team have also used the library like tensor flow, numpy, opencv, matplotlib, os etc. and also used tensorflow api keras which helps to develop this model. The research Team achieved accuracy of their model at least between 95- 99%.