Research And Implementation Of Opencv-Region-Based Intrusion Detection

Intelligent Video Surveillance is all application field of computer vision that attracted much attention in recent years.It can process,analyze and understand video signal and control the whole system using computer vision and image processing techniques. For example, a surveillance system can recognize different objects automatically, or detect unusual events in the surveillance scene.Then it provides the helpful information or alarms to monitoring staff.This can help the monitoring staff solving urgent problems quickly and effectively.Additionally, intelligent surveillance system can filtrate plenty of redundant information, only preserve key information.Intelligent video surveillance techniques call solve the problems such as huge processing data problem, long response times and human inherent weaknesses,which often occur in traditional video surveillance systems, lead to the inefficient monitoring and cumbersome workload.Intelligent video surveillance covers a wide range of systematic project involving image processing, and many technical areas of computer vision, pattern recognition, artificial intelligence technology, communications technology and network technology application, the paper focuses on the application one direction of intelligent video surveillance technology: a specific area of intrusion detection algorithm, and in-depth research involved in a number of specific issues related to the main study which includes:1. Research and achieve some of the basic technology in the intelligent video processing, including the use of three-layer adaptive mixture Gaussian model to model the background and the extraction of the moving target, under the condition of well suppressing noise and improving the integrity of the target.2. Study and realize the time- domain difference with the combination of the Canny operator to analyze the regional invasion, and the differential coefficient β is defined as a regional invasion threshold.3. Analyze the different background modeling method to extract the foreground image.4. Extract and analyze the contours of the moving target, and the target contour information could calculate the number of invasion in a specific area.