Performance Analysis of Sign Language Detection Using Deep Neural Networks and Computer Vision

Authors

  • Abha Bansal Research Scholar M. Tech, College of Engineering, Roorkee, UTU, U.K, INDIA

Keywords:

Convolutional, Neural, Classifier.

Abstract

This paper related to the Method of Training a Deep Learning Model and how we have used it for the American Sign Language detection. We have trained a Convolution Neural Nets (CNN) using Keras and TensorFlow as a backend. There is multiple image manipulation done in between using Computer Vision like resizing, thresholding, RGB2GRAY and the most important is histogram analysis which helps to identify the difference in background and image. The main aim of this project is to track the gestures made by the hand in American Sign Language and translate it into English. The entire project has been coded in Python language for its versatility. Using our Convolutional Neural Network and Keras, we were able to obtain 97.07% accuracy.

Downloads

Published

2020-09-30

How to Cite

Abha Bansal. (2020). Performance Analysis of Sign Language Detection Using Deep Neural Networks and Computer Vision. AlQalam Journal of Medical and Applied Sciences, 3(2), 78–81. Retrieved from https://journal.utripoli.edu.ly/index.php/Alqalam/article/view/90

Issue

Section

Articles