Creation of Image Segmentation Classifiers for Sign Language Processing for Deaf and Dumb

Recognition of sign language is an emerging area of research in the domain of gesture recognition. Research has been carried out around the world on sign language recognition, for many sign languages. The basic phase of sign language recognition systems is accurate hand segmentation. This paper used Otsu's technique of segmentation to create an improved vision-based recognition of sign language. There are about 466 million users suffering from hearing loss globally, of whom no. of kids is 34 million. Deaf people have very little or no capacity to hear. For communication, they use sign language. People in distinct areas of the globe use distinct sign languages which is very small in amount compared to spoken languages. Our goal is to create a static-gesture recognizer, a multi-class classifier that predicts the gestures of static sign language. In the proposed work, we identified the hand in the raw image and provided the static gesture recognizer (the multi-class classifier) with this section of the image. We build multi-class classifiers from the scikit learning library by first building the data set, each image being converted into a feature vector (X) and each has a label that matches the sign language alphabet denoted by (Y). Our predicted classifiers analyzed 65% of the said images with clarity.