Educating and communicating with deaf learner's using CNN based Sign Language Prediction System

Talking to the deaf has always been an important issue. Sign language has become a panacea and a very powerful tool by which deaf and hard of hearing people can communicate their feelings and opinions to the teachers, which helps to improve their education. This makes the referral process between themselves and teachers smoother and less complicated. Here, visible body movements are used to convey an important message. Sign language involves movements of different parts of the body, such as arms, legs, and face. Nonverbal physical communication, such as pure expressiveness, closeness, or shared interest, is different from gestures that convey a specific message. Gestures are very specific and have different meanings depending on your social or cultural background. However, just inventing sign language is not enough. Many conditions are attached to this blessing, the communication gap that has existed for many years now can be bridged by the advent of various gesture recognition automation technologies. This project involves developing a deep learning algorithm that classifies different sign language images such as alphabets and numbers, which helps deaf students to communicate with teachers and understand them better. Comparing the proposed algorithm with the current algorithm, it can be seen that the accuracy of hand gesture type classification based on CNN is higher than that of other algorithms. The success rate of the obtained results is expected to increase if the CNN method is facilitated by the addition of additional feature extraction methods.