Resumo

GIRAL RAMIREZ, Diego Armando; ROMERO ROMERO, Ricardo  e  MARTINEZ SANTA, Fernando. Parallel processing on FPGA for image convolution using Matlab. Tecnura [online]. 2015, vol.19, n.43, pp.119-138. ISSN 0123-921X.  https://doi.org/10.14483/udistrital.jour.tecnura.2015.1.a09.

This paper shows the design of two convolution image filter architectures, which use Hardware cosimulation through Xilinx System Generator Matlab toolbox to be implemented in a Xilinx Spartan 3AN FPGA. The purpose of the project is to evaluate the performance of parallel image processing versus the serial one. Initially the design and hardware implementation of the two architectures are performed, after, from the measurement of specific variables the best architecture is selected as an alternative for parallelization, using the tools provided by the toolbox evaluates the relationship between the degree of parallelism, execution times and hardware resources used , and finally among the most important conclusions are obtained that the processing time of the image is significantly reduced with increasing the degree of parallelization of the image.

Palavras-chave : convolution; degree of parallelization; programmable gate array; hardware cosimulation; Xilinx System Generator.

        · resumo em Espanhol     · texto em Espanhol     · Espanhol ( pdf )