摘要
为了解决住宅小区内智能监控系统依赖性强,检测效率低等问题,提出了一种基于DM6437的斜率阈值匹配算法的异常行为检测方法,使用自适应的背景相减法来检测运动物体,根据运动斜率判断人体运动是否异常。通过整个系统的硬件和软件的设计,保证了系统的可实施性。采用高性能的数字信号处理器(DM6437)作为系统的核心处理器,提高了运算速度。本文利用斜率阈值匹配法对所检测的运动目标进行判断,结果表明,该方法实现简单,具有较好的实时性,可以作为实时监控系统中异常行为检测的有效方法。
In order to solve strong dependence, low detection efficiency of the residential district intelligent monitoring system, we put forward art unusual behavior detection method based on the DM6437 slope threshold matching algorithm, use adaptive background subtrac- tion method to detect objects in motion, according to the movement slope to judge whether the human movement is abnormal. Through the whole system hardware and software design to ensure the system implementation. Using the high performance digital signal processor （DM6437） as the eore of the system processor, improve the computational speed. The results show that this method is simple, and has good real--time performance , can be used as a real--time monitoring system abnormal behavior detection method.
出处
《计算机测量与控制》
北大核心
2013年第9期2342-2345,共4页
Computer Measurement &Control
关键词
DM6437
阈值匹配
自适应背景相减法
异常行为检测
DM6437
threshold matching
adaptive phase background subtraction
abnormal behavior detection
