[ICME 2014, paperId 293]AN APPROACH FOR FAST AND PARALLEL VIDEO PROCESSING ON APACHE HADOOP CLUSTERS

ABSTRACT

This paper proposes an approach for fast and parallel video processing on MapReduce-based clusters such as Apache Hadoop. By utilizing clusters, the approach is able to handle large-scale of video data and the processing time can be significantly reduced. Technique details of performing video analysis on clusters are revealed, including method of porting typical video processing algorithms designed for a single computer to the proposed system. As case studies, face detection and motion detection and tracking algorithms have been implemented on clusters. Performance experiments on an Apache Hadoop cluster of six computers show that the system is able to reduce the running time of the two implemented algorithms to below 25% of that of a single computer. The applications of the system include smart city video surveillance, services provided by video sites and satellite image processing.

NOTES

The FULL TEXT is available inSEE ALSO(Conference members can find the full text in your ICME USB stick). And software packages required to setup the proposed system are listed here (Of course, you can google yourself for how to install them):

0. Redhat Linux or CentOS is recommended as the operating system (tested).

1. Hadoop Releases Download (The recommended release is 1.0.4 as it is tested, version >= 2.0 is not recommended as it might be incompatible with Fuse-DFS).

2 .Fuse Download (Depended by Fuse-DFS)

3. How to install Fuse-DFS (Necessary to mount HDFS to local filesystem.)

4. JDK download (Java Environment)

5. JavaCV Download (This package is optional now as OpenCV starts to support Java itself. If you still use APIs of OpenCV 1.0, you can use this.)

6. OpenCV Download (The version is supposed to be the same as JavaCV if you use JavaCV.)

It will take some efforts before you master Hadoop system well. However, once you master the system, you will find it amazingly efficient and can help you accelerate you image or video processing.

See also:

https://sites.google.com/site/tomheavenresearch/published-works/icme2014