Image Processing Reference
In-Depth Information
CHAPTER 8
Security surveillance
applications utilizing parallel
video-processing techniques
in the spatial domain
Leonidas Deligiannidis
1
;
Hamid R. Arabnia
2
1
Wentworth Institute of Technology, Department of Computer Science, Boston, MA, USA
2
University of Georgia, Computer Science, Athens, GA, USA
Abstract
We present several solutions to security surveillance and their applications while utilizing parallel pro-
cessing of video streams for improved performance. The algorithms presented in this article are ex-
plained in detail and their implementations are provided for educational purposes. These algorithms
with their implementations are being taught in our Parallel Processing course utilizing NVidia's CUDA
language and framework. We chose the topic of security surveillance in our Parallel Processing course
because the results are visual and applicable in many situations such as surveillance of parking lots (e.g.,
monitoring our car), dormitory rooms, offices, etc. Image-processing techniques and algorithms are dis-
cussed in this article, and extended to real-time high-definition video-processing focusing on security
surveillance.
Keywords
Parallel processing
Surveillance applications
Video processing
Acknowledgments
We would like to thank NVIDIA Corporation for providing equipment and financial support
through their CUDA Teaching Center initiative. We would also like to thank Axis Communica-
tions for donating more than 20 network-enabled surveillance cameras. Their support is greatly
appreciated.
Search WWH ::
Custom Search