Approche bayésienne pour la décomposition conjointe d’une séquence de spectres de photo-électrons

Approche bayésienne pour la décomposition conjointe d’une séquence de spectres de photo-électrons

Vincent Mazet Sylvain Faisan  Antoine Masson  Marc-André Gaveau  Lionel Poisson  Jean-Michel Mestdagh 

ICube, Université de Strasbourg, CNRS, UMR 7357 300 boulevard Sébastien Brant, BP 10413, F-67412 Illkirch cedex

CNRS, IRAMIS, SPAM, Laboratoire Francis Perrin, URA 2453 F-91191 Gif-sur-Yvette

CEA, IRAMIS, SPAM, Laboratoire Francis Perrin, URA 2453 F-91191 Gif-sur-Yvette

Corresponding Author Email: 
{vincent.mazet,faisan}@unistra.fr
Page: 
9-34
|
DOI: 
https://doi.org/10.3166/TS.30.9-34
Received: 
N/A
| |
Accepted: 
N/A
| | Citation

OPEN ACCESS

Abstract: 

This work deals with the decomposition of a temporal sequence of photoelectron spectra into a sum of peaks whose positions, amplitudes and widths are estimated. Since the peaks exhibit a (slow) evolution with time, the decomposition is performed jointly on the whole sequence to take this temporal information into account. To this end, we have developed a Bayesian model where a Markov random field favors a smooth evolution of peaks. The approach is unsupervised and a Gibbs sampler within a simulated annealing scheme enables to estimate the maximum a posteriori. We show the relevance of this approach compared with a method in which the spectra are decomposed separately and present an application on real photoelectron data.

RÉSUMÉ

Ce travail traite de la décomposition d’une séquence temporelle de spectres de photo-électrons en une somme de raies dont on estime les positions, amplitudes et largeurs. Comme les raies évoluent (lentement) dans le temps, la décomposition est effectuée conjointement sur toute la séquence afin de prendre en compte cette information temporelle. À cette fin, nous avons développé un modèle bayésien où un champ de Markov favorise une évolution douce des raies. L’approche est non supervisée et un échantillonneur de Gibbs couplé à un schéma de recuit simulé permet d’estimer le maximum a posteriori. Nous montrons la pertinence de cette approche par rapport à une méthode dans laquelle les spectres sont décomposés séparément et présentons une application sur données réelles de photo-électrons.

Keywords: 

spectroscopic signal sequence decomposition, Bayesian inference, Markov chain Monte Carlo (MCMC) methods, simulated annealing, photoelectron spectroscopy

MOTS-CLÉS

décomposition d’une séquence de signaux spectroscopiques, inférence bayésienne, méthodes de Monte Carlo par chaînes de Markov (MCMC), recuit simulé, spectroscopie de photo-électrons

1. Introduction
2. Modèle Bayésien
3. Algorithme MCMC Et Recuit Simulé
4. Résultats Sur Données Simulées
5. Résultats Sur Spectres De Photo-Électrons
6. Conclusion
  References

Allen G., McMeeking R. (1978). Deconvolution of spectra by least-squares fitting. Analytica Chimica Acta, vol. 103, no 1, p. 73-108.

Caruana R., Searle R., Heller T., Shupack S. (1986). Fast algorithm for the resolution of spectra. Analytical Chemistry, vol. 58, p. 1162-1167.

Chen S., Donoho D., Saunders M. (1995). Atomic decomposition by basis pursuit. Rapport technique. Standford University.

Davy M., Godsill S., Idier J. (2006, avril). Bayesian analysis of polyphonic western tonal music. Journal of the Acoustical Society of America, vol. 119, no 4, p. 2498-2517.

Devroye L. (1986). Non-uniform random variate generation. Springer-Verlag. http://luc.devroye.org/rnbookindex.html

Donoho D., Tsaig Y. (2006). Fast solution of l1-norm minimization problems when the solution may be sparse. Rapport technique. Standford University.

Eppink A., Parker D. (1997). Velocity map imaging of ions and electrons using electrostatic lenses: Application in photoelectron and photofragment ion imaging of molecular oxygen. Review of Scientific Instruments, vol. 68, no 9, p. 3477-3484.

Fischer R., Dose V. (2001). Analysis of mixtures in physical spectra. Bayesian methods, p. 145-154.

Fraser R., Suzuki E. (1966). Resolution of overlapping absorption bands by least squares procedures. Analytical Chemistry, vol. 38, no 12, p. 1770-1773.

Garcia G., Nahon L., Powis I. (2004). Two-dimensional charged particle image inversion using a polar basis function expansion. Review of Scientific Instruments, vol. 75, no 11, p. 4989-4996.

Ge D., Idier J., Le Carpentier E. (2011). Enhanced sampling schemes for MCMC based blind Bernoulli–Gaussian deconvolution. Signal Processing, vol. 91, no 4, p. 759-772.

Geman D., Yang C. (1995). Nonlinear image recovery with half-quadratic regularization. IEEE Transactions on Image Processing, vol. 4, no 7, p. 932–946.

Geman S., Geman D. (1984). Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images. IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 6, p. 721-741.

Gloaguen E., Mestdagh J.-M., Poisson L., Lepetit F., Visticot J.-P., et al. (2005). Experimental evidence for ultrafast electronic relaxation in molecules, mediated by diffuse states. Journal of the American Chemical Society, vol. 127, no 47, p. 16529-16534.

Gobinet C., Vrabie V., Manfait M., Piot O. (2009, mai). Preprocessing methods of Raman spectra for source extraction on biomedical samples: Application on paraffin-embedded skin biopsies. IEEE Transactions on Biomedical Engineering, vol. 56, no 5, p. 1371-1382.

Green P. (1995). Reversible jump Markov chain Monte Carlo computation and Bayesian model determination. Biometrika.

Gulam Razul S., Fitzgerald W., Andrieu C. (2003). Bayesian model selection and parameter estimation of nuclear emission spectra using RJMCMC. Nuclear Instruments and Methods in Physics Research A, vol. 497, p. 492-510.

Haan N., Godsill S. (2002). Bayesian models for DNA sequencing. In Ieee international conference on acoustics, speech, and signal processing, p. IV–4020–IV–4023. Idier J., Goussard Y. (1993, juillet). Markov modeling for Bayesian restoration of twodimensional layered structures. IEEE Transactions on Information Theory, vol. 39, no 4, p. 1356-1373.

Katajamaa M., Orešic M. (2005). Processing methods for differential analysis o ˇ f LC/MS profile data. BMC Bioinformatics, vol. 6, no 1, p. 179-191. http://www.biomedcentral.com/1471-2105/6/179

Keshava N., Mustard J. (2002, janvier). Spectral unmixing. IEEE Signal Processing Magazine, vol. 19, no 1, p. 44-57.

Kirkpatrick S., Gelatt Jr. C., Vecchi M. (1983). Optimization by simulated annealing. Science, vol. 220, p. 671-680.

Mallat S., Zhang Z. (1993, décembre). Matching pursuits with time-frequency dictionaries. IEEE Transactions on Signal Processing, vol. 41, no 12, p. 3397-3415.

Masson A., Poisson L., Gaveau M.-A., Soep B., Mestdagh J.-M., et al. (2010). Dynamics of highly excited barium atoms deposited on large argon clusters. I. General trends. The Journal of Chemical Physics, vol. 133, no 5 (054307).

Mazet V. (2005). Développement de méthodes de traitement de signaux spectroscopiques : estimation de la ligne de base et du spectre de raies. Thèse de doctorat non publiée, Université Henri Poincaré, Nancy 1.

Mazet V. (2011). Joint Bayesian decomposition of a spectroscopic signal sequence. IEEE Signal Processing Letters, vol. 18, no 3, p. 191-184.

Nguyen Thi H.-L., Jutten C. (1995). Blind source separation for convolutive mixtures. Signal Processing, vol. 45, no 2, p. 209-229.

Perret B., Mazet V., Collet C., Slezak E. (2011). Hierarchical multispectral galaxy decomposition using a MCMC algorithm with multiple temperature simulated annealing. Pattern Recognition, vol. 44, no 6, p. 1328-1342.

Poisson L., Maksimenska R., Soep B., Mestdagh J.-M., Parker D., et al. (2010). Unusual quantum interference in the S1 state of DABCO and observation of intramolecular vibrational redistribution. Journal of Physical Chemistry A, vol. 114, p. 3313-3319.

Poisson L., Mestdagh E. G. J.-M., Soep B., Gonzalez A., Chergui M. (2008). Direct observation of microscopic solvation at the surface of clusters by ultrafast photoelectron imaging. Journal of Physical Chemistry A, vol. 112, no 39, p. 9200-9210.

Robert C., Casella G. (2004). Monte Carlo statistical methods (2e éd.). Springer.

Roueff A., Chanussot J., Mars J., Nguyen M.-Q. (2004a). Partition d’une séquence d’images temps-échelle pour la séparation d’ondes dans un profil sismique. Traitement du signal, vol. 21, p. 17-36.

Roueff A., Chanussot J., Mars J., Nguyen M.-Q. (2004b). Unsupervised separation of seismic waves using the watershed algorithm on time-scale images. Geophysical Prospecting, vol. 52, p. 287-300.

Rue H., Held L. (2005). Gaussian Markov random fields: Theory and applications. Chapman & Hall/CRC.

Sahnoun S., Djermoune E.-H., Soussen C., Brie D. (2011). Sparse multiresolution modal estimation. In Ieee statistical signal processing workshop. Nice, France.

Stolow A., Bragg A., Neumark D. (2004). Femtosecond time-resolved photoelectron spectroscopy. Chemical Reviews, vol. 104, no 4, p. 1719-1757.

Tomasi G., van den Berg F., Andersson C. (2004). Correlation optimized warping and dynamic time warping as preprocessing methods for chromatographic data. Journal of Chemometrics, vol. 18, p. 231-241.

Wang L., Liu J., Li S. (2000). MRF parameter estimation by MCMC method. Pattern Recognition, vol. 33, no 11, p. 1919-1925.