Source identification and separation of number of active muscles during a complex action is useful information to identify the action, and to determine pathologies. Biosignals such as surface electromyogram are a result of the summation of electrical activity of a number of sources. The complexity of the anatomy and actions results difficulty in identifying the number of active sources from the multiple channel recordings. ICA has been applied to sEMG to separate the signals originating from different sources. But it is often difficult to determine the number of active sources that may vary between different actions and gestures. This paper reports research conducted to evaluate the use of sub-band ICA for the separation of bioelectric signals when the number of active sources may not be known. The paper proposes the use of value of the determinant of the global matrix generated using sub-band ICA for identifying the number of active sources. The results indicate that the technique is successful in identifying the number of active muscles for complex hand gestures.
[1]
Marimuthu Palaniswami,et al.
Hand gestures for HCI using ICA of EMG
,
2006
.
[2]
Andrzej Cichocki,et al.
Adaptive blind signal and image processing
,
2002
.
[3]
Carlo J. De Luca,et al.
Physiology and Mathematics of Myoelectric Signals
,
1979
.
[4]
Terrence J. Sejnowski,et al.
An Information-Maximization Approach to Blind Separation and Blind Deconvolution
,
1995,
Neural Computation.
[5]
Francesco Carlo Morabito,et al.
A Morlet wavelet classification technique for ICA filtered sEMG experimental data
,
2003,
Proceedings of the International Joint Conference on Neural Networks, 2003..
[6]
Christopher J James,et al.
Independent component analysis for biomedical signals
,
2005,
Physiological measurement.
[7]
Erkki Oja,et al.
Independent Component Analysis
,
2001
.
[8]
Pierre Comon,et al.
Independent component analysis, A new concept?
,
1994,
Signal Process..