EMD-KURTOSIS: A New Classification Domain for Automated Greek Sign Language Gesture Recognition

Sign Language (SL), structured on a set of gestures, is the communication channel between the deaf; this, however, requires extra intervention, such as SL translator, to facilitate the communication of the deaf with the hearers. Efforts towards automated SL translation have already reported in the literature. In this paper, a new approach is proposed towards this direction. In particular, data from three-dimensional accelerometer and five-channel surface electromyogram of the user’s dominant forearm are analyzed using empirical mode decomposition combined with higher-order statistics (K4EMD) towards the automated recognition of Greek Sign Language (GSL) gestures. The fourth-order moment, i.e., kurtosis, which is a measure of the deviation from the Gaussianity, estimated within overlapping observation windows for each of the intrinsic mode functions of the measured data was used as feature set for the classification of the signed gestures. The proposed feature set is evaluated with principal component analysis based on the Mahalanobis distance criterion to identify the effective scales of the intrinsic mode functions for the calculation of the fourth-order moment for each channel of the measured data that best discriminate between signed gestures of the GSL. The classification results from the K4EMD analysis revealed that 90% of the inspected GSL gestures corresponding to ten words were correctly recognized, providing a promising solution to the automated GSL gesture recognition problem.

[1]  Thad Starner,et al.  Visual Recognition of American Sign Language Using Hidden Markov Models. , 1995 .

[2]  C. L. Nikias,et al.  Higher-order spectra analysis : a nonlinear signal processing framework , 1993 .

[3]  N. Huang,et al.  The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis , 1998, Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences.

[4]  Dimitris N. Metaxas,et al.  ASL recognition based on a coupling between HMMs and 3D motion analysis , 1998, Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271).

[5]  Dimitris N. Metaxas,et al.  Adapting hidden Markov models for ASL recognition by using three-dimensional computer vision methods , 1997, 1997 IEEE International Conference on Systems, Man, and Cybernetics. Computational Cybernetics and Simulation.

[6]  M. Vuskovic,et al.  Mahalanobis distance-based ARTMAP network , 2004, 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No.04CH37541).

[7]  Geoffrey E. Hinton,et al.  Glove-Talk: a neural network interface between a data-glove and a speech synthesizer , 1993, IEEE Trans. Neural Networks.

[8]  Dimitris N. Metaxas,et al.  Toward Scalability in ASL Recognition: Breaking Down Signs into Phonemes , 1999, Gesture Workshop.

[9]  Geoffrey E. Hinton,et al.  Glove-TalkII: Mapping Hand Gestures to Speech Using Neural Networks , 1994, NIPS.

[10]  David J. Hewson,et al.  Intrinsic Mode Entropy for Nonlinear Discriminant Analysis , 2007, IEEE Signal Processing Letters.

[11]  Leontios J. Hadjileontiadis,et al.  Intrinsic mode entropy: An enhanced classification means for automated Greek Sign Language gesture recognition , 2008, 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[12]  Robert W. Lindeman,et al.  A new instrumented approach for translating American Sign Language into sound and text , 2004, Sixth IEEE International Conference on Automatic Face and Gesture Recognition, 2004. Proceedings..

[13]  L.J. Hadjileontiadis,et al.  Evaluation of surface EMG features for the recognition of American Sign Language gestures , 2006, 2006 International Conference of the IEEE Engineering in Medicine and Biology Society.