首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于嵌入式系统的室内人体跌倒行为检测方法的实现
引用本文:姚成玉,赵佳伟,陈毅强.基于嵌入式系统的室内人体跌倒行为检测方法的实现[J].燕山大学学报,2008,32(6).
作者姓名:姚成玉  赵佳伟  陈毅强
作者单位:1. 燕山大学,工业计算机控制工程河北省重点实验室,河北,秦皇岛,066004
2. 秦皇岛视听机械研究所,河北,秦皇岛,066000
基金项目:河北省教育厅科学研究计划资助项目  
摘    要:提出了基于视觉和嵌入式系统来检测室内人体跌倒行为的方法,采用多线程算法进行图像采集和处理.运用基于简化高斯模型的自适应背景减除算法得到二值前景图像,并引入了形态学滤波,通过简化的目标宽高比算法判断人体跌倒行为是否发生,并进行虚报警判断.试验证明该系统的整体设计达到了实时性、准确性兼顾的特点.

关 键 词:嵌入式系统  多线程  跌倒行为检测  图像处理  形态学滤波

Realization of the detection method for human falling action in room based on embedded system
YAO Cheng-yu,ZHAO Jia-wei,CHEN Yi-qiang.Realization of the detection method for human falling action in room based on embedded system[J].Journal of Yanshan University,2008,32(6).
Authors:YAO Cheng-yu  ZHAO Jia-wei  CHEN Yi-qiang
Institution:YAO Cheng-yu1,ZHAO Jia-wei1,CHEN Yi-qiang2 (1. Key Lab of Industrial Computer Control Engineering of Hebei Province,Yanshan University,Qinhuangdao,Hebei 066004,China,2. Qinhuangdao Audio-Visual Machinery Research Institute,Hebei 066000,China)
Abstract:Resolution of falling-action detection for human based on vision and embedded system is introduced. Multi-process algorithm is designed to cover image capture and image processing. Capture of two valued foreground image depends on adaptive background subtraction algorithm based on predigested Single Gaussian Model and falling action is estimated by proportion between width and height after morphology filter, moreover, false alarms is rejected. The experiment result shows that it has reached the feature of r...
Keywords:embedded system  multi-process  falling action detection  image processing  morphology filter  
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号