The Application Of OpenCV In The Embedded Intelligent Surveillance System

Video surveillance system is widely used in traffic, frontier defense, public securityand other fields, its inevitable trend of development is intelligent, network-based andhigh-definition. In order to deal with emergencies timely,intelligent video analysisalgorithms is integrated in the video surveillance system, so that passive monitoring can bechanged to active monitoring and the after-analysis can be turned into real-time analysisand pre-alarming. In this paper, an embedded intelligent surveillance system is achieved onOpenCV open-source computer vision library.The system hardware is based on ARM CortexTM-A8S5PV210,along with remotevideo monitoring and face detection intelligence capabilities.Man-machine interface isdeveloped by using Qt graphical library;The video acquisition module is designed onV4L2driver framework;The system also realizes the local display of surveillancevideo;The h.264compression of surveillance video uses S5PV210built-in MFC; Thecompressed h.264stream is sent to the Internet in accordance with the RTP/RTCPprotocol;The effectiveness of surveillance video network-based transmission isdemonstrated by VLC software. Face detection is developed by using Adaboost algorithm,based on OpenCV open-source computer vision library.After testing, the expected design goal is achieved. The significance of this paper is tobuild a development platform for embedded intelligent video surveillance system, based onOpenCV.And more intelligent surveillance system can be developed according to differentrequirements based on the platform.