A non-probabilistic recognizer of stochastic signals based on KLT

This paper presents an efficient algorithm which is able to accurately recognize non-deterministic signals generated by synthetic non-chaotic and chaotic stochastic processes (SPs), as well as by natural phenomena (that are inherently stochastic) such as speech, image, and electroencephalographic signals. This recognition algorithm exploits a Karhunen-Loeve transform (KLT)-based model able to characterize signals in terms of non-deterministic trajectories and consists of the concatenation of (i) a training stage, which iteratively extracts suitable parameter collections by means of the KLT and (ii) a recognition procedure based on ad hoc metric that measures the trajectory-proximities, in order to associate the unknown signal to the SP which this signal can be considered a realization of. The proposed methodology is able to recognize SPs without estimating their probability density function (pdf), thus requiring a low computational complexity to be implemented. Exhaustive experimentation on specific case-studies was performed and some experimental results were compared to other existing techniques such as hidden Markov model (HMM), vector quantization (VQ), and dynamic time warping (DTW). Recognition performance is similar to current best practices for non-chaotic signals and higher for chaotic ones. A better noise rejection was also achieved, and a reduction of two orders of magnitude in training-times compared with HMM was obtained, thus making the proposed methodology one of the current best practices in this field. Finally, the experimental results obtained by three different applications of the recognizer (an automatic speech recognition system, an automatic facial recognition system, and an automatic diagnosis system of the ictal and interictal epilepsy) clearly show excellent classification performance, and it is worth noting as complex filters are not needed unlike other current best practices.

[1]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[2]  Mark A. Clements,et al.  Digital Signal Processing and Statistical Classification , 2002 .

[3]  Trevor Darrell,et al.  Nearest-Neighbor Methods in Learning and Vision , 2008, IEEE Trans. Neural Networks.

[4]  K Lehnertz,et al.  Indications of nonlinear deterministic and finite-dimensional structures in time series of brain electrical activity: dependence on recording region and brain state. , 2001, Physical review. E, Statistical, nonlinear, and soft matter physics.

[5]  S. Hahn.  Hilbert Transforms in Signal Processing , 1996 .

[6]  T. Poggio,et al.  General conditions for predictivity in learning theory , 2004, Nature.

[7]  Erkki Oja,et al.  Subspace methods of pattern recognition , 1983 .

[8]  Sergios Theodoridis,et al.  Recognition of isolated musical patterns using Context Dependent Dynamic Time Warping , 2002, 2002 11th European Signal Processing Conference.

[9]  Jean Véronis,et al.  A multilingual prosodic database , 1998, ICSLP.

[10]  Anil K. Jain,et al.  Statistical Pattern Recognition: A Review , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[11]  F. Gianfelici,et al.  Nearest-Neighbor Methods in Learning and Vision (Shakhnarovich, G. et al., Eds.; 2006) [Book review] , 2008 .

[12]  Arnt-Børre Salberg,et al.  Robust Multidimensional Matched Subspace Classifiers Based on Weighted Least-Squares , 2007, IEEE Transactions on Signal Processing.

[13]  M. Markus,et al.  Oscillations and turbulence induced by an activating agent in an active medium. , 2001, Physical review. E, Statistical, nonlinear, and soft matter physics.

[14]  Edward R. Dougherty,et al.  Random processes for image and signal processing , 1998, SPIE / IEEE series on imaging science and engineering.

[15]  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.

[16]  Jesper Ø. Olsen.  ICSLP'98 : Proceedings of the 5th International Conference on Spoken Language Processing, November 30-December 4, 1998, Sydney, Australia , 1998 .

[17]  S. Elaydi.  An introduction to difference equations , 1995 .

[18]  S. Ramamohan,et al.  Sinusoidal model-based analysis and classification of stressed speech , 2006, IEEE Transactions on Audio, Speech, and Language Processing.

[19]  Giorgio Biagetti,et al.  A novel KLT algorithm optimized for small signal sets [speech processing applications] , 2005, Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005..

[20]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[21]  Petros Maragos,et al.  AM-FM energy detection and separation in noise using multiband energy operators , 1993, IEEE Trans. Signal Process..

[22]  P. Zweifel.  Advanced Mathematical Methods for Scientists and Engineers , 1980 .

[23]  Dafydd Gibbon,et al.  EUROM - a spoken language resource for the EU - the SAM projects , 1995, EUROSPEECH.

[24]  Steven Kay,et al.  Asymptotically Optimal Approximation of Multidimensional pdf's by Lower Dimensional pdf's , 2007, IEEE Transactions on Signal Processing.

[25]  Petros Maragos,et al.  Multicomponent AM-FM demodulation via periodicity-based algebraic separation and energy-based demodulation , 2000, IEEE Trans. Commun..

[26]  B. Picinbono.  Random Signals and Systems , 1993 .

[27]  Paul M. Baggenstoss.  The PDF projection theorem and the class-specific method , 2003, IEEE Trans. Signal Process..

[28]  Adam Krzyzak,et al.  Learning and Design of Principal Curves , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[29]  Giorgio Biagetti,et al.  Multicomponent AM–FM Representations: An Asymptotically Exact Approach , 2007, IEEE Transactions on Audio, Speech, and Language Processing.