Parallel codebook design for vector quantization on a message passing MIMD architecture

Authors:
Hazem M. Abbas;Mohamed M. Bayoumi
Affiliations:
Mentor Graphics Corp., Egypt;Department of Electrical and Computer Engineering, Queen's University, Kingston, Ont., Canada, K7L 3N6
Venue:
Parallel Computing - Parallel computing in image and video processing
Year:
2002

Citing 7
Cited 4

Trillium operating system

C3P Proceedings of the third conference on Hypercube concurrent computers and applications: Architecture, software, computer systems, and general issues - Volume 1
Competitive learning algorithms for vector quantization

Neural Networks
Parallel processing in industrial real-time applications

Parallel processing in industrial real-time applications
Comparison of parallel and serial implementation of feedforward neural networks

Journal of Microcomputer Applications
Performance prediction of large MIMD systems for parallel neural network simulations

Future Generation Computer Systems - Special issue: massive parallel computing
A Massively Parallel Implementation of the Full Search Vector Quantization Algorithm

HPCN Europe 1994 Proceedings of the nternational Conference and Exhibition on High-Performance Computing and Networking Volume I: Applications
Progressive vector quantization on a massively parallel SIMD machine with application to multispectral image data

IEEE Transactions on Image Processing

A Low-Cost Parallel K-Means VQ Algorithm Using Cluster Computing

ICDAR '03 Proceedings of the Seventh International Conference on Document Analysis and Recognition - Volume 2
Efficient parallel processing of competitive learning algorithms

Parallel Computing
Hierarchical parallel processing of large scale data clustering on a PC cluster with GPU co-processing

The Journal of Supercomputing
Multi-grain parallel processing of data-clustering on programmable graphics hardware

ISPA'04 Proceedings of the Second international conference on Parallel and Distributed Processing and Applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

Vector quantization (VQ) is a widely used algorithm in speech and image data compression. One of the problems of the VQ methodology is that it requires large computation time especially for large codebook size. This paper addresses two issues. The first deals with the parallel construction of the VQ codebook which can drastically reduce the training time. A master/worker parallel implementation of a VQ algorithm is proposed. The algorithm is executed on the DM-MIMD Alex AVX-2 machine using a pipeline architecture. The second issue deals with the ability of accurately predicting the machine performance. Using communication and computation models, a comparison between expected and real performance is carried out. Results show that the two models can accurately predict the performance of the machine for image data compression. Analysis of metrics normally used in parallel realization is conducted.