Security systems have always had an important role in society for a long time. The purpose of these systems has always remained primarily focused on the security aspect. In most cases the efficiency of current security systems depend on human factors. The frames have to be processed by a person to reach a definitive conclusion. In recent years, security systems have become increasingly intelligent as technology continues to develop. Methods that implement automatic analysis of any given frame in a video have been introduced and some of these methods are already standardized. The aim of this study is to develop a prototype and stress test it to identify possible flaws. This study is primarily a feasibility study to show the possibilities of what can be achieved with limited resources. With a fully functional prototype available a series of stress tests will be specified to measure the system’s limits. The results have been positive overall but a number of critical flaws have been identified.
[1]
Mislav Grgic,et al.
SCface – surveillance cameras face database
,
2011,
Multimedia Tools and Applications.
[2]
Kari Pulli,et al.
Realtime Computer Vision with OpenCV
,
2012,
ACM Queue.
[3]
Mubarak Shah,et al.
Automated Visual Surveillance in Realistic Scenarios
,
2007,
IEEE MultiMedia.
[4]
R. Weiss.
Learning from strangers : the art and method of qualitative interview studies
,
1995
.
[5]
Brian Thorne.
Introduction to Computer Vision in Python
,
2009
.
[6]
G. Thiel.
Automatic CCTV surveillance-towards the VIRTUAL GUARD
,
2000
.
[7]
Mubarak Shah,et al.
Tracking and Object Classification for Automated Surveillance
,
2002,
ECCV.