TY - GEN
T1 - Dual LMCs fusion for recognition of isolated Arabic sign language words
AU - Aliyu, S.
AU - Mohandes, M.
AU - Deriche, M.
N1 - Publisher Copyright:
© 2017 IEEE.
PY - 2017/12/4
Y1 - 2017/12/4
N2 - In this paper, we propose a Dual-Leap Motion Controllers (DLMC) based Arabic Sign Language recognition system. More particularly, we propose to use both front and side controllers to cater for the challenges of finger occlusions and missing data. For feature extraction, we select an optimum set of geometric features extracted from both controllers, while for classification, we used both a Bayesian approach with a Gaussian Mixture Model (GMM) and a simple Linear Discriminant Analysis (LDA) approach. Though this paper focused only on the GMM approach. Data was collected from a native adult signer, for 100 isolated Arabic words. Ten observations were collected for each of the signs. The proposed framework uses an intelligent strategy to handle the case of missing data from one or both controllers. A recognition accuracy of 94.63% was achieved, with the proposed system. The proposed system outperforms glove-based systems and a single-LMC based techniques.
AB - In this paper, we propose a Dual-Leap Motion Controllers (DLMC) based Arabic Sign Language recognition system. More particularly, we propose to use both front and side controllers to cater for the challenges of finger occlusions and missing data. For feature extraction, we select an optimum set of geometric features extracted from both controllers, while for classification, we used both a Bayesian approach with a Gaussian Mixture Model (GMM) and a simple Linear Discriminant Analysis (LDA) approach. Though this paper focused only on the GMM approach. Data was collected from a native adult signer, for 100 isolated Arabic words. Ten observations were collected for each of the signs. The proposed framework uses an intelligent strategy to handle the case of missing data from one or both controllers. A recognition accuracy of 94.63% was achieved, with the proposed system. The proposed system outperforms glove-based systems and a single-LMC based techniques.
KW - Arabic sign langauge recognition
KW - electronic glove
KW - image-based system
KW - leap motion controller
KW - machine learning
UR - https://www.scopus.com/pages/publications/85046649103
U2 - 10.1109/SSD.2017.8167010
DO - 10.1109/SSD.2017.8167010
M3 - Conference contribution
AN - SCOPUS:85046649103
T3 - 2017 14th International Multi-Conference on Systems, Signals and Devices, SSD 2017
SP - 611
EP - 614
BT - 2017 14th International Multi-Conference on Systems, Signals and Devices, SSD 2017
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 14th International Multi-Conference on Systems, Signals and Devices, SSD 2017
Y2 - 28 March 2017 through 31 March 2017
ER -