Human Detection by Surveillance Camera

Human Detection by Surveillance Camera

Larbi Guezouli (LaSTIC Laboratory - University of Batna2, Batna, Algeria), Hanane Boukhetache (LaSTIC laboratory, Batna, Algeria) and Imene Kebi (LaSTIC laboratory, Batna, Algeria)
Copyright: © 2018 |Pages: 13
DOI: 10.4018/IJRAT.2018010102

Abstract

Security problems and decreasing costs, leads to the rapid development of video surveillance systems. It is necessary to implement analytical tools capable of identifying objects that may appear in the video sequence. The work presented in this article consists of designing a video surveillance system for the automatic detection of humans in a video sequence acquired by a fixed camera. The principle of this work is based on the modeling and subtraction of the background. In order to determine the nature of the objects, the authors make the detection of the contours of the foreground image, then by matching this contour with the images of a base, silhouette images of people in different positions. The acquisition of the frames is carried out in real time, the matching of the images takes a considerable time and this time becomes increasingly longer based on the size of the base. To solve this problem, the authors have used the parallelism.
Article Preview

Video Surveillance And Motion Detection

Surveillance cameras can be an effective technique for protecting public safety and detecting or discouraging criminal activity. The analysis of human behaviour since the video attracts the interest of a large number of researchers despite the fact that the problems are diverse and varied. Most approaches in the literature estimate the movement as low-level information (Sarah & Mahfoud, 2016).

Motion detection is usually a control algorithm that helps the surveillance camera to choose when to start capturing the detected event.

Video surveillance becomes a fundamental component of our society as was the case for the mobile phone. Insecurity is also the new element in this need to monitor, statistics prove it.

Video surveillance is a system of cameras arranged in a space to be monitored. These cameras are connected to a computer system which allows the processing and analysis of the received data. Data analysis and integration are increasingly automated and require less human intervention.

Video surveillance is a system that automatically detects the movement of an object. It also allows following this object, it is based on the notion of the dissimilarity between neighbor frames of a video stream. It also exploits the properties of the object like: size, position and speed (Le, 2009; Teresa, 2016).

The Design of a Video Surveillance System

The motion detection step plays a very important role in the video surveillance system because the result of this step will influence all the following steps. Figure 1 shows the general architecture of a video surveillance system.

Figure 1.

General architecture of a video surveillance system

Detection of Human Presence

We all have a common morphology, although we are all different. We all have two arms, two legs, one head, etc. which implies that a human silhouette can be differentiated from a silhouette of an animal for example.

Human detection involves segmenting moving regions.

The movement may in particular be detected relative to a background image, to the time difference or to the optical flow (Nguyen, Li, & Ogunbona, 2016; Paul, Haque, & Chakraborty, 2013; Sardeshmukh, Kolte, & Joshi, 2016).

Complete Article List

Search this Journal:
Reset
Open Access Articles
Volume 7: 2 Issues (2019): Forthcoming, Available for Pre-Order
Volume 6: 2 Issues (2018): 1 Released, 1 Forthcoming
Volume 5: 2 Issues (2017)
Volume 4: 2 Issues (2016)
Volume 3: 2 Issues (2015)
Volume 2: 2 Issues (2014)
Volume 1: 2 Issues (2013)
View Complete Journal Contents Listing