Publication: Design of a parallel application of image segmentation based on Corba : feasibility study
Loading...
Identifiers
Publication date
2011-05-05
Defense date
2011-09-01
Authors
Advisors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Impact
Export
Abstract
In computer vision, segmentation refers to the process of partitioning a digital image into multiple segments. This process involves a high consumption of resources, both memory and CPU. Therefore, these applications are specially suited for HPC (High-performance computing) environments. Nowadays, a wide variety of methods and algorithms have been available recently to deal with the trouble of the segmentation of images, one of them is used in this project that was called PSRG. This algorithm was designed to be executed parallel. Parallel executions use MPI for performing the data distribution. We have considered interesting to enable these simulations on machines with different architectures and distributed over wide area networks, allowing them to run on heterogeneous architectures and/or operative systems. To achieve it, we started this project with an implementation of a CORBA−based communications library. Once the development finished, we proceed to evaluate the performance of this new version of PSRG.
Description
Keywords
Visión artificial, Proceso de imágenes, Proceso en paralelo, Segmentación de imágenes