Parallel Best Neighborhood Matching Algorithm on PC Cluter Platform

Yan Yan Zhang; Li Qiang He

doi:10.4028/www.scientific.net/AMR.383-390.4109

Paper Titles

Research on Photovoltaic Grid-Connected Generation System
p.4084

The Research and Development of NTC Automatic Tool Selection Based on Aucol
p.4090

Study on the Power Control Method of a Stand-Alone Wind Energy System Based on DSEG
p.4096

The Design of the Supervisory System Based on the SOPC
p.4103

Parallel Best Neighborhood Matching Algorithm on PC Cluter Platform
p.4109

Research of Roll-Isolated Gimbal Platform and a Two-Position Initial Alignment Scheme of SINS in Spinning Projectile
p.4115

Power-Fail Detection and Data Storage Design for Control System
p.4121

A PSO Weighted Estimating the Mean Downstream Demand to Mitigate the Bullwhip Effect in a Supply Chain
p.4125

Rough Sets Based Simplified Analysis of Energy Loss Indicators in Power Plant
p.4130

HomeAdvanced Materials ResearchAdvanced Materials Research Vols. 383-390Parallel Best Neighborhood Matching Algorithm on...

Parallel Best Neighborhood Matching Algorithm on PC Cluter Platform

Abstract:

Best Neighborhood Matching (BNM) has been shown to be an effective technique to recover a broken image due to a failed network transmission. For high definition image, the time of recovering using BNM algorithm is relative too long and not suitable for real time processing. And the original BNM only targets at gray scale image, and not considers the color image case. In this paper, we extend the BNM algorithm to color image area and use PC clusters to parallelly recover the image and improve the processing speed dramatically. The experiment result from a four nodes cluster system shows that our parallel BNM implementation improves the processing speed up to 7.52 times without any obvious recover quality decrease.

You might also be interested in these eBooks

Manufacturing Science and Technology, ICMST2011

View Preview

Info:

Periodical:

Advanced Materials Research (Volumes 383-390)

Pages:

4109-4114

DOI:

https://doi.org/10.4028/www.scientific.net/AMR.383-390.4109

Citation:

Cite this paper

Online since:

November 2011

Authors:

Yan Yan Zhang, Li Qiang He

Keywords:

BNM, Image Recover, Parallel Implementation, Pc-Cluster

Export:

RIS, BibTeX

Price:

Permissions:

Request Permissions

References

[1] Z. Wang, Y. L. Yu, and D. Zhang. Best neighborhood matching: An information loss restoration technique for block-based image coding systems[J]. IEEE Transactions on Image Process, July, 1998, Vol. 7, No. 7: 1056–1061.

DOI: 10.1109/83.701166

Google Scholar

[2] Z. Wang, Z. Wang, and Y. Yu. Volutionary strategy with its application to best-neighbourhood-matching error concealment algorithm [J]. Journal of China Institute of Communication , Dec, 2002, Vol. 21 No. 12: 52-58.

Google Scholar

[3] H. Liang, Y. Mao, H. Wu, and L. Xiao. Two step best neighborhood matching for fast image resotration. [J]. Journal of Computer-aided Design and Computer Graphics, Sep, 2004, Vol·16, No·9: 1279-1283.

Google Scholar

[4] W. Li, D. Zhang An Efficient Parallel Algorithm for Image Restoration and Its Key Techniques[J]. Journal of Computer Research and Development, 2002, Vol. 39, No. 7: 848-854.

Google Scholar

[5] D. Zhang and Z. Wang. Image information restoration based on long-range correlation[J]. IEEE Transactions on Circuits and Systems for Video Technology, May, 2002: 331-341.

DOI: 10.1109/tcsvt.2002.1003472

Google Scholar

[6] W. Li, D. Zhang, Z. Liu and X. Qiao. Fast block-based image restoration employing the Improved Best Neighborhood Matching approach, IEEE Transactions on Systems, Man and Cybernetics[J]. 2005, Vol. 35, No. 4: 546-555.

DOI: 10.1109/tsmca.2005.850605

Google Scholar