Impact of Colour Image and Skeleton Plotting on Sign Language Recognition Using Convolutional Neural Networks (CNN)

The significance of sign language lies in its role in communicating and expressing the views of the deaf and hard-of-hearing community. In a world where verbal communication is one of the highest regards of communication among each other, sign language serves as a powerful and inclusive means of expression, allowing individuals to connect and converse with other individuals. In this paper, we introduce a Convolutional Neural Network (CNN) based model, using American Sign Language. Users must capture images of the hand gestures, and the model must predict the sign made using the captured image. The paper also defines the difference in model accuracy depending upon the type of preprocessing done on the data. The comparison of the model accuracy based on defiantly preprocessed data is used to achieve an accuracy of about 99% using a colored, skeleton-mapped sign image dataset.