Un algorithme rapide d'extraction d'arêtes dans le scalogramme et son utilisation dans la recherche de zones stationnaires

Un algorithme rapide d'extraction d'arêtes dans le scalogramme et son utilisation dans la recherche de zones stationnaires

A fast ridge extraction algorithm from the scalogram, applied to search of stationary areas

Hélène Leman Catherine Marque 

Université de Technologie de Compiègne, UMR CNRS 6600, Compiègne

Corresponding Author Email: 
elene.leman@utc.fr
Page: 
577-581
|
Received: 
8 March 1998
| |
Accepted: 
N/A
| | Citation

OPEN ACCESS

Abstract: 

We propose to extract the ridges from the scalogram, with the help of a rapid and simple algorithm, based on the local maxima detection, their linkage and their interpolation. The signal reconstruction is possible thanks to the Marseille's algorithm, proposed in 1990. Moreover, the reconstruction, from a limited number of ridges of greatest energy, makes the stationarity research easier. Indeed, the reconstructed version of the signal is, in someway, a simplified version of the original signal . We apply these methods to the uterine electromyogram signal, in order to characterize the contractions during pregnancy.

Résumé

Nous proposons d'extraire les arêtes dans le scalogramme, grâce à un algorithme rapide et simple, basé sur la détection des maxima locaux, leur lien et leur interpolation . La reconstruction du signal est possible grâce à l'algorithme proposé par Marseille en 1990 . La recherche de zones stationnaires est alors facilitée par la reconstruction du signal, à partir d'un certain nombre d'arêtes . En effet, la version reconstruite est, en quelque sorte, une version «simplifiée» du signal original. Nous appliquons ces méthodes au signal électromyographique utérin, en vue d'une caractérisation des contractions utérines pendant la grossesse.

Keywords: 

Electrohysterogram, ridges, scalogram, stationnarity

Mots clés

Électrohystérogramme, arêtes, scalogramme, stationnarité

1. Introduction
2. Principe De L'algorithme
3. Détection De Zones De Stationnarité
4. Résultats : Application À L'EMG Utérin
5. Conclusion
  References

[1] S. Adak, «Time-Dependent Spectral Analysis of Nonstationarity Time Series», tech . rep ., Standford University, 1995 .

[21 U . Appel and A . Brandt, «Adaptive Sequential Segmentation of Piecewise Stationary Time Series», Information Sciences, vol . 29, pp . 27—56, 1983 .

[3] U . Appel and A. Brandt, «A Comparative Study of Three Sequential Time Series Segmentation Algorithms», Signal Processing, vol . 6, pp . 45—60, 1984 .

[4] R . Carmona, W. Hwang, and B . Torrésani, «Characterization of Signals by the Ridges of their Wavelet Transform», tech . rep ., Standford University, 1995 .

[5] R . Carmona, W. Hwang, and B . Torrésani, «Identification of Chirps with Continuous Wavelet Transform», tech . rep., Standford University, 1995 .

[6] R . Carmona, W. Hwang, and B. Torrésani, «Multi-Ridge Detection and Time - Frequency Reconstruction», tech . rep., Standford University, 1995 .

[7] P. Carré and C . Fernandez, «Research of Stationary Partitions in Nonstationary Processes by Measurement of Spectral Distance with the Help on non Dyadic Malvar's Decomposition», in TFTS, (Pittsburgh, USA), 1998 .

[8] B . Dahanayake and A . Upton, «On-Line Methodology for Detection of Suspected Epileptic Seizures for Electroencephalogram EEG», in IFAC, 12th Triennal World Congress, (Sydney, Australia), 1993 .

[9] N. Delprat, B . Escudié, P. Guillemain, R . Kronland-Martinet, P. Tchamitchian , and B .Torrésani, «Asymptotic Wavelet and Gabor Analysis : Extraction of Instantaneous Frequencies», IEEE Trans. Warm. Theor., vol . 38, pp . 644—664, 1992 .

[10] D . Donoho, «Wavelet Shrinkage and w.v .d . : A 10-minute tour», tech. rep. , Standford University, 1992 .

[11] M . Grewal and A. Andrews, Kalman filtering . Englewood Cliffs : Prentice - Hall Inc, 1993 .

[12] S . Kay, Modern Spectral Estimation, Theory and Application . Englewood Cliffs : Prentice-Hall Inc, 1988 .

[13] R. Kronland-Martinet and P. Guillemain, «Ridges Associated to Continuous Linear Time-Frequency Representations of Asymptotic and Transient Signals», in TFTS, (Paris, France), 1996 .

[14] H. Leman and C . Marque, «Optimum Denoising Method for the Uterine EHG», in IEEE EMBS, 19th Annual Conference, (Chicago, USA), 1997.

[15] H . Leman and C. Marque, «Ridge Extraction from the Scalogram of the Uterine Electromyogram», in TFTS, (Pittsburgh, USA), 1998 .

[16] H. Leman, C . Marque, and J . Gondry, «Use of the Electrohysterogram Signal for Characterization of Contractions during Pregnancy» . Accepted for publication in IEEE Trans . on Biomedical Engineering .

[17] J. Pardey, S . Roberts, and L . Tarassenko, «A Review of Parametric Modeling Techniques for EEG Analysis», Med. Eng. Phys., vol. 18, pp . 2-7, 1995

[18] P. Tchamitchian and B . Torrésani, Ridge and Skeleton Extraction from the Wavelet transform. Boston : in Wavelets and Applications, Ruskai MB et al. Ed, 1992 .

[19] G . Tognola, P. Ravazzani, F. Minicucci, T.Locatelli, F. Grandori, J . Ruohonen , and G . Comi, «Analysis of Temporal Non-Stationnarities in EEG Signals by Means of Parametric Modelling», Technology and Health Care, vol . 4 , pp . 169-185, 1996 .

[20] B . Torrésani, Analyse Continue par Ondelettes . Paris : InterEditions et CNRS Editions, 1995 .

[21] M . Wickerhauser, Adapted Wavelet Analysis from Theory to Software . Massachusetts : A K Peters, 1994 .

[22] B . Widrow and S . Stearns, Adaptive Signal Processing . Englewood Cliffs : Prentice-Hall Inc, 1985.