Title: Gesture recognition system using 2D-invariant moment feature and Elman neural network

Authors: M.P. Paulraj; C.R. Hema; Sazali Bin Yaacob; Mohd Shuhanaz Zanar Azalan; Rajkumar Palaniappan

Addresses: School of Mechatronic Engineering, University Malaysia Perlis, Perlis, 02600, Malaysia ' Faculty of Engineering, Karpagam University, Coimbatore, 641 021, India ' School of Mechatronic Engineering, University Malaysia Perlis, Perlis, 02600, Malaysia ' School of Mechatronic Engineering, University Malaysia Perlis, Perlis, 02600, Malaysia ' School of Mechatronic Engineering, University Malaysia Perlis, Perlis, 02600, Malaysia

Abstract: This paper presents a simple sign language recognition system that has been developed using skin colour segmentation and Elman neural network. A simple segmentation process is carried out to separate the right and left hand. The 2D-invariant moments of the right and left hand segmented image are obtained as features. Using the 2D-invariant moment features, an Elman neural network model was developed. The system has been implemented and tested for its validity. Experimental results show that the system has a recognition rate of 90.63%.

Keywords: sign language recognition; hand gestures; 2D invariant moment features; gesture recognition; neural networks; skin colour segmentation; image segmentation.

DOI: 10.1504/IJAISC.2013.056826

International Journal of Artificial Intelligence and Soft Computing, 2013 Vol.3 No.4, pp.298 - 309

Received: 06 Jul 2012
Accepted: 03 Mar 2013
Published online: 12 Jul 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article