Modeling of Multi-Image/Video Flow on a Multiprocessor Surveillance System

Modeling of Multi-Image/Video Flow on a Multiprocessor Surveillance System

Athanasios Tsitsoulis (Assistive Technologies Research Center (ATRC), College of Engineering and Computer Science, Wright State University, Dayton, OH, USA) and Nikolaos Bourbakis (Assistive Technologies Research Center (ATRC), College of Engineering and Computer Science, Wright State University, Dayton, OH, USA)
DOI: 10.4018/ijmstr.2013010101
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

DIAS is a distributed, multiprocessor picture information system architecture. It consists of four main cooperating parts (Input-Preprocessing; Planning-Control; Image Processing/Analysis/Understanding; Backend). The performance of the DIAS system is mainly based on the performance of the Image Processing/Analysis/Understanding (IA) section, which is a multiprocessor array structure. Thus, the functional modeling of the IA section can provide a satisfactory picture of the overall DIAS system function. The analytical model is used to evaluate what parameters are involved in the Image Analysis array structure. Experimental results demonstrate its performance in strenuous conditions and its ability to guarantee real-time constraints.
Article Preview

2. General Configuration

The overall organization of the DIAS system architecture is illustrated in Figure 1. The DIAS system receives image either through a set of 2-D photoarrays (PAs) from the direct environment, or from a storage area. The values of the image pixels are then fed to a set of preprocessors (Cs), which process these values extracting critical information (parameters), such as average intensities of various picture regions, the number of pixels per region, the locations of informative areas, etc. (Davis & Burns, 2011). Since the preprocessors must function very quickly, they are implemented in hardware.

Figure 1.

General organization of DIAS

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 5: 4 Issues (2017): 2 Released, 2 Forthcoming
Volume 4: 4 Issues (2016)
Volume 3: 4 Issues (2015)
Volume 2: 4 Issues (2014)
Volume 1: 4 Issues (2013)
View Complete Journal Contents Listing