The concept and principle of intelligent video surveillance analysis technology

The essential difference between intelligent video analysis and motion detection is that the former can accurately identify the real moving target in the video, while the latter can only judge the content of the picture change, and can not distinguish the target and background interference. Therefore, intelligent video analysis has a qualitative improvement in anti-interference ability compared to motion detection. Using intelligent analysis technology, users can preset different alarm rules in different camera scenes according to the actual application. Once the target has violated the predefined rules in the scene, the system will automatically issue an alarm. The alarm information can be in various forms, including local driving alarm equipment and sending alarm data to the back-end monitoring center. The monitoring station controls the pop-up video, automatically pops up alarm information, and drives alarm devices.

The technical principle of intelligent video analysis is to access various video devices such as various cameras and DVRs, DVSs, and streaming media servers, and to automatically alert various security events through intelligent image recognition processing technology, and to analyze alarms through real-time analysis. Conductive integrated monitoring platform and client. Specifically, the intelligent video analysis system "discovers the enemy's situation" in real time through the camera and "sees" the surveillance target in the field of vision, and at the same time judges whether the behavior of these monitored targets has a security threat through its own intelligent recognition algorithm. Or threats that will occur, promptly send an alarm to the integrated monitoring platform or background management personnel through voice, video and other types.

Intelligent video analysis technology has been developed and applied in foreign countries for nearly 10 years. The most famous professional intelligent video analysis vendors in the world are VCATechnology, IOImage, ObjectVideo, Bosch, Axis, and IBM, Sony, Panasonic, PELCO, Honeywell. Companies such as Seoul and Siemens also have quite influential overall solution products in this field. The domestic intelligent video analytics solution vendors mainly include the Institute of Software Application Technology of the Chinese Academy of Sciences in Guangzhou, Haikang, Dahua, Bokang, and some manufacturers specializing in intelligent video analysis, such as UDI Digital, Bellson, Gao Dewei, Zhongsheng Yihua, etc. the company.

Intelligent video analysis technology is widely used in public safety related systems, building intelligence, intelligent transportation and other related fields.

There are two types of intelligent video analytics technology for video surveillance solutions. The first is a front-end solution based on intelligent video processors. In this mode, all target tracking, behavior judgment, and alarm triggering are performed by the front-end intelligent analysis device, and only the alarm information is transmitted to the monitoring center through the network. The second is a back-end intelligent video analytics solution based on industrial computers. In this mode, all front-end cameras only have basic video capture capabilities, and all video analysis must be aggregated to the back end or key nodes for unified processing by the computer. In the market, the first method is mostly applied, and the video analysis device is placed behind the IP camera, which can effectively save the bandwidth occupied by the video stream. Industrial computer-based solutions can only control a number of critical monitoring points and require high levels of computer performance and network bandwidth.