A Low cost learning based sign language recognition system

Download
2018
Akış, Abdullah Hakan
Sign Language Recognition (SLR) is an active area of research due to its important role in Human Computer Interaction (HCI). The aim of this work is to automatically recognize hand gestures consisting of the movement of hand, arm and fingers. To achieve this, we studied two different approaches, namely feature based recognition and Convolutional Neural Networks (CNN) based recognition. The first approach is based on segmentation, feature extraction and classification whereas the second one is based on segmentation and CNN which learns the signs from the image itself. In order to calculate the recognition rate of the systems, tests are conducted using eNTERFACE dataset of 8 American Sign Language (ASL) signs. Detailed analysis is done to evaluate each step of both approaches. Experimental results show that the feature based SLR system and CNN based SLR system achieved recognition rate of 95.31% and 93.12%, respectively. Experimental results also show that CNN based SLR system achieved recognition rate of 94.29% when data augmentation is used to increase the training dataset.
Citation Formats
A. H. Akış, “A Low cost learning based sign language recognition system,” M.S. - Master of Science, Middle East Technical University, 2018.