Real Time Sign Language Recognition Using Image Classification

Image classification using Convolutional Neural Networks is a key component of Computer Vision. This paper proposes a real-time sign language translation system based on American Sign Language. The system aims at effective communication between vocally and hearing-impaired people and common people in society by translating sign language gestures into text and speech in English. Built-in webcam using OpenCV is the mode of video capturing used to track the hand gestures by the system. Convolutional Neural Networks (CNNs), a powerful deep learning tool, are used to train the system for recognizing signs. The proposed methodology in the paper aims to give effective outputs with minimum delay in real-time.