Two-Handed Bangla Sign Language Recognition Using Principal Component Analysis (PCA) And KNN Algorithm
Promila Haque, Badhon Das, Nazmun Nahar Kaspy · 2019
Communication is the basic necessity for the existence in the society. Hearing-impaired people connect among themselves utilizing sign language, but typical people find it challenging to recognize it. The purpose of our paper is to decrease the communication gap between ordinary people and people with hearing and speech disabilities. This paper shows the sign language recognition framework prepared for recognizing 26 gestures from the Bangla Sign Language by utilizing PYTHON. In our proposed structure there are three phases: 1. Database formation phase 2. Training phase 3. Classification phase Preprocessing, segmentation, Eigenvector, and Eigenvalues are utilized as a part of the recognition. The Principal Component Analysis (PCA) was being used for recognizing images by extracting their principal component, and K-Nearest Neighbor algorithm is used for the classification phase. This method is very efficient with various backgrounds and effects.