论文部分内容阅读
约在10年前,安防行业提出了视频智能化的发展方向和目标,当时主要是为了针对大量增加的监控图像,解决事后查看、案情研判工作中警力不足以及由于视觉疲劳造成信息遗漏等问题。其技术原理为将场景中的目标和背景进行分离,得到包括目标位置、尺寸、形状、速度、停留时间等基本形态和动态信息并完成目标的跟踪和行为理解之后,与预先设定的规则进行判定,直到触发报警。此后,随着智能分析应用需求的不断升级,又引入了即时报警、主动预警的理念,各种智能分析技术如雨后春笋般地快速发展了起来,逐步应用在了银行、机场、车站、博物馆、
About 10 years ago, the security industry put forward the development direction and target of video intelligence. At that time, it mainly aimed at a large number of surveillance images to solve problems such as checking afterwards, insufficient police force in case judgment and information missing due to visual fatigue. The technical principle is to separate the target and the background in the scene, obtain the basic form and dynamic information including the target position, size, shape, speed and residence time, and complete the tracking and behavior comprehension of the target, and then carry out the pre-set rules Judgment until the alarm is triggered. Since then, with the continuous upgrading of the demand for intelligent analysis applications, the concept of instant alarm and active warning has also been introduced. Various intelligent analysis technologies have mushroomed rapidly and gradually applied to banks, airports, railway stations, museums,