基于网格系统的人脸识别算法的实现

Human face recognition algorithm's realization based on gird system

网格计算是近年来逐渐兴起并不断发展的技术,目前广泛应用于分布式超级计算、分布式仪器系统、数据密集型计算和远程沉浸等多个研究领域。常规人脸识别系统中的更新算法都是基于单机的串行算法,在处理大量数据的时候速度慢,效率低,介绍了网格计算在人脸识别系统中的应用,把原来的人脸数据更新串行算法改为并行算法并通过编写M P I并行程序移植到该网格计算平台中运行,实现了原有人脸识别系统中更新算法的分布式处理,增强了系统处理大量数据的能力,以达到提高系统性能的目的。

Grid computing is the technology that arises gradually and develops continuously in these years, and it is used widely in numerous research domain including Distributed Supercomputing, Distributed Instrumentation System, Data Intensive Computing and Tele - immersion. The algorithm of upgrading of general face recognition system is based on all serial algorithm in a single computer, it is very slow to deal with a large amount of data, and the efficiency is low. so,this article introduces the application of grid computing in face recognition system, upgrades the original serial algorithm of face data into parallel algorithm in the grid platform by using MPI parallel program, realizes the already existing updated algorithm of the recognition of face to be processed in the distributed computers, which has strengthened the systematic ability to deal with a large amount of data, in order to improve systematic performance.