Title :
Bayesian prior for reconstruction of compressively sampled astronomical images
Author :
Colonnese, S. ; Cusani, Roberto ; Rinauro, Stefano ; Scarano, G.
Author_Institution :
Dipt. di Ing. dell´Inf., Elettron. e delle Telecomun., Univ. di Roma La Sapienza, Rome, Italy
Abstract :
This paper presents an original reconstruction procedure for compressively sampled astronomical images. The procedure stems from adoption of a Bayesian prior on the original astronomical image. Simulation results show that the Bayesian prior enhances the iterative greedy support estimation procedure usually adopted in the literature leading to improved convergence performance and reconstruction accuracy.
Keywords :
Bayes methods; astronomical image processing; convergence of numerical methods; data compression; greedy algorithms; image coding; image reconstruction; image sampling; iterative methods; Bayesian prior; compressively sampled astronomical image reconstruction; convergence performance improvement; iterative greedy support estimation procedure enhancement; reconstruction accuracy improvement; Bayes methods; Convergence; Estimation; Image coding; Image reconstruction; Indexes; Vectors; Bayesian prior; Compressive sampling; astronomical imaging;
Conference_Titel :
Visual Information Processing (EUVIP), 2013 4th European Workshop on
Conference_Location :
Paris
