Fast Multidimensional Image Processing with OpenCL
Universidade Federal de Sergipe, Departamento de Computacao
IEEE International Conference on Image Processing (ICIP), 2016
@inproceedings{dantas2016fast,
title={Fast multidimensional image processing with OpenCL},
author={Dantas, Daniel Oliveira and Leal, Helton Danilo Passos and Sousa, Davy Oliveira Barros},
booktitle={2016 IEEE International Conference on Image Processing (ICIP)},
pages={1779–1783},
year={2016},
organization={IEEE}
}
Multidimensional image data, i.e., images with three or more dimensions, are used in many areas of science. Multidimensional image processing is supported in Python and MATLAB. VisionGL is an open source library that provides a set of image processing functions and can help the programmer by automatically generating code. The objective of this work is to augment VisionGL by adding multidimensional image processing support with OpenCL for high performance through use of GPUs. Benchmarking experiments were run with window and point operations to compare Python, MATLAB and VisionGL when processing 1D to 5D images. As a result, speedups of up to two orders of magnitude were obtained.
August 23, 2016 by hgpu
