基于嵌入式视频监控的摔倒检测系统的设计  

Rate-control Algorithm of H. 264 /AVC Based Human Visual System

作　　者：白云飞[1] 李瞡 何金龙[1] 

机构地区：[1]中国计量学院机电工程学院,浙江杭州310018

出　　处：《电视技术》2014年第15期231-236,共6页Video Engineering

摘　　要：为了迅速获得老人跌倒信息,使其得到及时救助,设计和实现了基于嵌入式视频监控的摔倒检测系统。该系统无需人体佩戴任何设备,即可自动检测老人跌倒信息并报警。对监控到的视频流采用H.264标准进行编码压缩,运用B/S网络结构模式实现远程视频监控。同时,对监控到的视频流进行人体摔倒检测以实现报警。摔倒检测使用改进的高斯混合模型算法对背景进行更新,采用背景减除法来分割运动目标。经实验表明该嵌入式系统可实现远程监控并准确分割出人体目标,对其姿态进行自动判别,出现跌倒时进行报警。该系统满足实时性要求,误报率低,鲁棒性好。In order to quickly get information when old man fall down and need timely assistance, a kind of intelligent video surveillance system is designed and implemented which people need no device to wear. To monitor the video stream for human detection, technology like the H. 264 video standard, remote monitoring is realized by using B/S network structure model. An improved method of Gauss mixture model is used to update the background. The background subtraction method is used to segment the moving object. The experiment indicates that the client can be remotely monitored and accurate segmentation of the human targets, automatically identify the person＇s gesture. It can make an alert when people fall down or other anomalies occurred. The system satisfies the requirement of real-time, low rate of false positives and good robustness.

关 键 词：嵌入式 远程监控 摔倒检测 

分 类 号：TP316[自动化与计算机技术—计算机软件与理论]