A Comparative Study of Color Space Conversion on Homogeneous and Heterogeneous Multicore

Chen Chu; Jian Wang; Sen Ke Hou; Qi Lv; Guo Qiang Ma; Xiao Yong Ji

doi:10.4028/www.scientific.net/AMM.519-520.724

Paper Titles

Individual Tree Detection from High Spatial Resolution Imagery Using Color and Texture Features
p.703

An Optimized Video Capture and Preprocessing Strategy Based on FPGA
p.708

Fabric Image Edge Detection Based on Octonion and Echo State Networks
p.714

A Data Parallel Implementation Scheme of Geometric Operations
p.719

A Comparative Study of Color Space Conversion on Homogeneous and Heterogeneous Multicore
p.724

A Clustering Algorithm Based on Gravitation
p.731

Trend Analysis of Product Function Using Sequential Pattern Mining
p.736

The Satellite Attitude Control Law Design Based on Machine Learning
p.741

A Novel Distributed Jointly Sparse Optimization Algorithm
p.747

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vols. 519-520A Comparative Study of Color Space Conversion on...

A Comparative Study of Color Space Conversion on Homogeneous and Heterogeneous Multicore

Abstract:

Color space conversion (CSC) is an important kernel in the area of image and video processing applications including video compression. As a matrix math, this operation consumes up to 40% of processing time of a highly optimized decoder. Therefore, techniques which efficiently implement this conversion are desired. Multicore processors provide an opportunity to increase the performance of CSC by exploiting data parallelism. In this paper, we present three novel approaches for efficient implementation of color space conversion suitable for homogeneous and heterogeneous multicore. We compare the performance of color space conversion on a variety of platforms including OpenMP running on homogeneous multicore CPUs, CUDA with NVIDIA GPUs and OpenCL running on both NVIDIA and ATI GPUs. Our experimental results show that the speedup of 3×, 17× and 15× can been obtained, respectively.

You might also be interested in these eBooks

View Preview

Info:

Periodical:

Applied Mechanics and Materials (Volumes 519-520)

Pages:

724-728

DOI:

https://doi.org/10.4028/www.scientific.net/AMM.519-520.724

Citation:

Cite this paper

Online since:

February 2014

Authors:

Chen Chu, Jian Wang*, Sen Ke Hou, Qi Lv, Guo Qiang Ma, Xiao Yong Ji

Keywords:

CUDA, GPGPU, Homogeneous and Heterogeneous Multicore, OpenCL, OpenMP

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

* - Corresponding Author

References

[1] H. Kim and R. Bond: Signal Processing Magazine, IEEE, Vol. 26 (2009), p.80.

Google Scholar

[2] E. Ayguad_l , e, N. Copty, A. Duran, J. Hoeinger, Y. Lin, F. Massaioli, X. Teruel, P. Unnikrishnan, and G. Zhang: Parallel and Distributed Systems, IEEE Trans. Vol. 20 (2009), p.404.

DOI: 10.1109/tpds.2008.105

Google Scholar

[3] G. Taboada, C. Teijeiro, J. Tourino, B. Fraguela, R. Doallo, J. Mourino, D. Mallon, and A. Gomez: High Performance Computing and Communications, IEEE International Conference, (2009) p.69.

DOI: 10.1109/hpcc.2009.88

Google Scholar

[4] H. Yang, J. Wang, and X. Ji: Advanced Materials Research Vol. 716 (2013), p.505.

Google Scholar

[5] A. Munshi. The OpenCL Specification version 1. 1. Khronos OpenCL Working Group, (2011).

Google Scholar

[6] E. Dupuis: Information on http: /lestourtereaux. free. fr/papers/data/yuvrgb. pdf.

Google Scholar

[7] Y. Yang, P. Yuhua, and L. Zhao: Consumer Electronics, IEEE Trans. Vol. 53 (2007), p.1490.

Google Scholar

[8] A. Shahbahrami, B. H. H. Juurlink, and S. Vassiliadis: Architecture and Code Optimization, ACM Trans. (2008), p.1.

Google Scholar

[9] F. Bensaali, A. Amira, and S. Chandrasekaran. Electronics Circuits and Systems, IEEE International Conference, (2006) p.164.

Google Scholar