Tipo di prodotto  Contributo in atti di convegno 

Titolo  A minimax entropy method for blind separation of dependent components in astrophysical images 
Anno di pubblicazione  2006 
Formato  Elettronico 
Autore/i  Caiafa C. F., Kuruoglu E. E., Proto A. N. 
Affiliazioni autori  Lab. de Sistemas Complejos, Fac. de Ingeniería  UBA, Buenos Aires, Argentina; CNRISTI, Pisa, Italy; Comisión de Inv. Científicas de la Prov. de Buenos Aires, Argentina, Italy 
Autori CNR e affiliazioni 

Lingua/e 

Abstract  We develop a new technique for blind separation of potentially non independent components in astrophysical images. Given a set of linearly mixed images, corresponding to different measurement channels, we estimate the original electromagnetic radiation sources in a blind fashion. Specifically, we investigate the separation of cosmic microwave background (CMB), thermal dust and galactic synchrotron emissions without imposing any assumption on the mixing matrix. In our approach, we use the Gaussian and nonGaussian features of astrophysical sources and we assume that CMBdust and CMBsynchrotron are uncorrelated pairs while dust and synchrotron are correlated which is in agreement with theory. These assumptions allow us to develop an algorithm which associates the Minimum Entropy solutions with the nonGaussian sources (thermal dust and galactic synchrotron emissions) and the Maximum Entropy solution as the only Gaussian source which is the CMB. This new method is more appropriate than ICA algorithms because independence between sources is not imposed which is a more realistic situation. We investigate two specific measures associated with entropy: Gaussianity Measure (GM) and Shannon Entropy (SE) and we compare them. Finally, we present a complete set of examples of separation using these two measures validating our approach and showing that it performs better than FastICA algorithm. The experimental results presented here were performed on an image database that simulates the measurements expected from the instruments that will operate onboard ESA's Planck Surveyor Satellite to measure the CMB anisotropies all over the celestial sphere. 
Lingua abstract  inglese 
Altro abstract   
Lingua altro abstract   
Pagine da   
Pagine a  81 
Pagine totali  88 
Rivista  AIP conference proceedings Attiva dal 1970 Editore: American Institute of Physics,  New York Paese di pubblicazione: Stati Uniti d'America Lingua: inglese ISSN: 0094243X Titolo chiave: AIP conference proceedings Titolo proprio: AIP conference proceedings. Titolo abbreviato: AIP conf. proc. Titoli alternativi:

Numero volume della rivista  872 
Serie/Collana   
Titolo del volume   
Numero volume della serie/collana   
Curatore/i del volume  A. M. Djafari 
ISBN   
DOI   
Editore   
Verificato da referee  Sì: Internazionale 
Stato della pubblicazione  Published version 
Indicizzazione (in banche dati controllate)   
Parole chiave  Source separation, Dependent component analysis, Minimax entropy, Astrophysical microwave radiation images 
Link (URL, URI)  http://aip.scitation.org/doi/abs/10.1063/1.2423263 
Titolo convegno/congresso  Bayesian Inference and Maximum Entropy Methods In Science and Engineering 
Luogo convegno/congresso  Paris, France 
Data/e convegno/congresso  0813/07/2006 
Rilevanza  Internazionale 
Relazione  Contributo 
Titolo parallelo   
Note/Altre informazioni  Codice puma: cnr.isti/2006A275 
Strutture CNR 

Moduli/Attività/Sottoprogetti CNR 

Progetti Europei   
Allegati  A minimax entropy method for blind separation of dependent components in astrophysical images (documento privato ) Tipo documento: application/pdf 
Area disciplinare  Computer Science & Engineering 

Note  In: Bayesian Inference and Maximum Entropy Methods In Science and Engineering (Paris, France, July 813, 2006). Proceedings, pp. 8188. A. M. Djafari (ed.). American Institute of Physics, 2006. 
Descrizione sintetica del prodotto  ABSTRACT: We develop a new technique for blind separation of potentially non independent components in astrophysical images. Given a set of linearly mixed images, corresponding to different measurement channels, we estimate the original electromagnetic radiation sources in a blind fashion. Specifically, we investigate the separation of cosmic microwave background (CMB), thermal dust and galactic synchrotron emissions without imposing any assumption on the mixing matrix. In our approach, we use the Gaussian and nonGaussian features of astrophysical sources and we assume that CMBdust and CMBsynchrotron are uncorrelated pairs while dust and synchrotron are correlated which is in agreement with theory. These assumptions allow us to develop an algorithm which associates the Minimum Entropy solutions with the nonGaussian sources (thermal dust and galactic synchrotron emissions) and the Maximum Entropy solution as the only Gaussian source which is the CMB. This new method is more appropriate than ICA algorithms because independence between sources is not imposed which is a more realistic situation. We investigate two specific measures associated with entropy: Gaussianity Measure (GM) and Shannon Entropy (SE) and we compare them. Finally, we present a complete set of examples of separation using these two measures validating our approach and showing that it performs better than FastICA algorithm. The experimental results presented here were performed on an image database that simulates the measurements expected from the instruments that will operate onboard ESA's Planck Surveyor Satellite to measure the CMB anisotropies all over the celestial sphere. 