Sciweavers

Share
PDCN
2004

K-Means VQ algorithm using a low-cost parallel cluster computing

10 years 29 days ago
K-Means VQ algorithm using a low-cost parallel cluster computing
It is well-known that the time and memory necessary to create a codebook from large training databases have hindered the vector quantization based systems for real applications. To overcome this problem, we present a parallel approach for the K-means Vector Quantization (VQ) algorithm based on master/slave paradigm and lowcost parallel cluster computing. Distributing the training samples over the slaves' local disks reduces the overhead associated with the communication process. In addition, models predicting computation and communication time have been developed. These models are useful to predict the optimal number of slaves taking into account the number of training samples and codebook size. The experiments have shown the efficiency of the proposed models and also a linear speed up of the vector quantization process used in a two-stage Hidden Markov Model (HMM)-based system for recognizing handwritten numeral strings.
Paulo Sergio Lopes de Souza, Alceu de Souza Britto
Added 31 Oct 2010
Updated 31 Oct 2010
Type Conference
Year 2004
Where PDCN
Authors Paulo Sergio Lopes de Souza, Alceu de Souza Britto Jr., Robert Sabourin, Simone do Rocio Senger de Souza, Díbio Leandro Borges
Comments (0)
books