Implementation of a Bayesian Statistics Segmentation Algorithm in GPUs

F. FIORETTI; D. LEMMA; E. MILLAN JUJTIUK; R. ISOARDI; E.M. BRINGA

Buenos Aires

Congreso; HPC 2012; 2012

Universidad Nacional de La Plata

The increasing interest in the implementation of scientific algorithms in GPUs (Graphic Processing Units) stems from the high levels of computing acceleration which can be obtained through their use, for instance in image processing. Tomographic image segmentation using a Bayesian approach relies on the statistical inference of global parameters unsuitable for local data treatment, which is the strength of the highly parallel processing architecture of GPUs. We solve this problem with a parallel algorithm, and obtain a performance which is up to 57x of the serial CPU version of the code.