The Deep Learning Approach For American Sign Language Detection

Sign language has importance rule to deal with communication process especially with impairments hearing people. Sign language detection also attract lot of researchers to join the challenge of research to detect and recognize the sign language in the field of Computer Science. Hence, there is still no any standard approach and method to recognize the meaning in every pose of sign language. This research proposed a mechanism to detect Alphabet American Sign Language by utilizing Convolutional Neural Network (CNN) process. The CNN approach was chosen based on the ability and capability to recognize image. In this research, MNIST dataset is used for traning and testing process. The proposed CNN architecture produced 97% of accuracy that outperform the previous research using the same dataset which made this architecture promising.