Large Scale Image Processing Using Distributed and Parallel Architecture
Large Scale Image Processing Using Distributed and Parallel Architecture
复制标题
使用分布式并行架构的大规模图像处理
DOI：
--
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
M. Potdar
中科院分区：
文献类型：
--
作者：
Helly M. Patel;Krunal J. Panchal;Prashant Chauhan;M. Potdar
--通过互联网上传和使用的图像数量巨大，而且这一比例仍在急剧增加。因此，立即需要处理这些数据，并根据用户和应用程序的要求对其进行定制和过滤。本文介绍了Hadoop、HIPI、Map Reduce、CUDA、MPI等并行分布式处理技术在海量数据库处理中的应用。基于MapReduce的大规模图像处理，在分布式并行环境下具有很高的可靠性和可扩展性。这些方法都是高效的并行数据处理方法，可以用来分析和处理海量图像数据。在当今高度数字化的世界中，大量的多媒体数据需要有效地存储、处理和分析。关键词-图像处理、并行和分布式处理、Mapreduce、Hadoop、HIPI、CUDA、MPI
--Enormous amount of images are uploaded and used via internet and this ratio is still drastically increasing .So there is instant need to handle this data and customize and filter it as per user and application requirement. This paper describe some parallel and distributed processing techniques like Hadoop, HIPI, Map reduce, CUDA, MPI to process massive database. Mapreduce based large-scale images processing, which exhibit high reliability and scalability in distributed and parallel environment. As highly efficient parallel data processing all these methods are used to analyze and process massive image database. In today’s highly digitalized world large volume of multimedia data need to efficiently store, process and analyze This paper shows usefulness of parallel and distributed techniques. Keywords---Image Processing, Parallel and distributed processing, Mapreduce, Hadoop, HIPI, CUDA, MPI