Use of PVM for MAP image restoration: A parallel implementation of the ARTUR algorithm

Markov Random Field (MRF) based algorithms used for restoring images affected by blur and noise are generally very effective, but most of them are computationally very heavy. The recently proposed ARTUR algorithm is a deterministic method that belongs to this family. In this work, a parallel impleme...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Fernández, G.J., Jacobo-Berlles, J., Borensztejn, P., Bauzá, M., Mejail, M., Kacsuk P., Podhorszki N., Dongarra J, Foundation for the Technological Progress of the Industry; Microsoft; Myricom; Silicon Computers
Formato: SER
Materias:
Algorithms
Image reconstruction
Markov processes
Deterministic methods
Markov Random Fields
Parallel implementations
Message passing
Acceso en línea:http://hdl.handle.net/20.500.12110/paper_03029743_v1908_n_p113_Fernandez
Aporte de:
Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA) de Universidad de Buenos Aires
Descripción
Sumario:Markov Random Field (MRF) based algorithms used for restoring images affected by blur and noise are generally very effective, but most of them are computationally very heavy. The recently proposed ARTUR algorithm is a deterministic method that belongs to this family. In this work, a parallel implementation of this algorithm is proposed ona PVM environment. Various results are shown and the performance is analysed comparing the proposed parallel implementation against the sequential one. © Springer-Verlag Berlin Heidelberg 2000.

Ejemplares similares