本文關(guān)鍵詞：基于嵌入式視頻監(jiān)控的摔倒檢測系統(tǒng)的設計　出處：《電視技術(shù)》2014年15期 　論文類型：期刊論文

  更多相關(guān)文章： 嵌入式 遠程監(jiān)控 摔倒檢測

【摘要】：為了迅速獲得老人跌倒信息,使其得到及時救助,設計和實現(xiàn)了基于嵌入式視頻監(jiān)控的摔倒檢測系統(tǒng)。該系統(tǒng)無需人體佩戴任何設備,即可自動檢測老人跌倒信息并報警。對監(jiān)控到的視頻流采用H.264標準進行編碼壓縮,運用B/S網(wǎng)絡結(jié)構(gòu)模式實現(xiàn)遠程視頻監(jiān)控。同時,對監(jiān)控到的視頻流進行人體摔倒檢測以實現(xiàn)報警。摔倒檢測使用改進的高斯混合模型算法對背景進行更新,采用背景減除法來分割運動目標。經(jīng)實驗表明該嵌入式系統(tǒng)可實現(xiàn)遠程監(jiān)控并準確分割出人體目標,對其姿態(tài)進行自動判別,出現(xiàn)跌倒時進行報警。該系統(tǒng)滿足實時性要求,誤報率低,魯棒性好。
[Abstract]:In order to get the fall information of the elderly quickly and get the timely rescue, a fall detection system based on embedded video surveillance is designed and implemented, which does not need any equipment to be worn by the human body. Can automatically detect the elderly fall information and alarm. The video stream monitored by H.264 standard coding compression, using the B / S network structure mode to achieve remote video monitoring. At the same time. Human fall detection is performed on monitored video streams to achieve alarm. Fall detection uses an improved Gao Si hybrid model algorithm to update the background. The background subtraction method is used to segment the moving target. The experiment shows that the embedded system can realize remote monitoring and accurately segment the human body target, and automatically distinguish the posture of the target. The system meets the real-time requirement, the false alarm rate is low and the robustness is good.
【作者單位】： 中國計量學院機電工程學院;
【分類號】：TN919.81;TP274
【正文快照】： 當前老齡化社會已經(jīng)到來,研究表明跌倒是老年人常見的傷害事件,是老年人群傷殘、失能和死亡的主要原因。如果能及時發(fā)現(xiàn)老年人跌倒并進行有效的處理,就會降低跌倒對老人的傷害。據(jù)全國調(diào)查顯示,老年人的跌倒有一半以上是在家中發(fā)生的[1],而現(xiàn)在許多家庭的子女由于多種原因,沒，

本文編號：1386966