MRKDSBC: A Distributed Background Modeling Algorithm Based on MapReduce

扫码查看

原文链接

NETL

外文摘要：Video surveillance is a widely used technology。 Moving object detection is the most important content of video surveillance。 Background modeling is an important method in moving object detection。 However， background modeling algorithm is usually computationally intensive when the size of video is large。 Kernel density estimation method based on Chebyshev inequality (KDSBC) is a new background modeling algorithm。 This paper present MRKDSBC based on MapReduce which is a distributed programming model。 Further more， we prove the correctness of the algorithm theoretically and implement it on Hadoop platform。 Finally， we compare it with traditional algorithm。

外文关键词：

Chebyshev inequalitybackground modelingKernel density estimationdistributedMapReduceHadoop

作者：

Cong Wan、Cuirong Wang、Kun Zhang

展开 >

作者单位：

College of Information Science and Engineering,Northeastern University, Shenyang 110044, China

会议名称：

International symposium on neural networks

会议地点：

Shengyang(CN)

会议母体文献：

Advances in neural networks - ISNN 2012

页码：

668-677

出版时间：

2012