【摘要】：隨著我國(guó)人口老齡化程度不斷加劇和社會(huì)生活壓力的日漸增長(zhǎng),空巢老人的比例不斷上升。老人獨(dú)自居住造成很多社會(huì)問(wèn)題,尤其是老人的健康問(wèn)題,其中老人意外跌倒損傷是影響老人健康的主要原因之一。老人跌倒后沒(méi)有被及時(shí)處理會(huì)增加二次傷害,甚至導(dǎo)致老人意外死亡。據(jù)相關(guān)統(tǒng)計(jì)50%以上的老人跌倒發(fā)生在家中,如果能夠?qū)先说牡剐袨檫M(jìn)行實(shí)時(shí)的監(jiān)控,當(dāng)老人發(fā)生跌倒行為后能準(zhǔn)確識(shí)別,并向其監(jiān)護(hù)人發(fā)送提醒信息,使老人得到及時(shí)的救助將極大減少跌倒對(duì)老人健康的傷害�；诖�,本文展開(kāi)了基于深度學(xué)習(xí)和WebRTC的智能跌倒監(jiān)控系統(tǒng)的研究。主要工作如下:在分析基于深度學(xué)習(xí)和WebRTC的智能跌倒監(jiān)控系統(tǒng)的功能需求的基礎(chǔ)上,提出了系統(tǒng)的技術(shù)方案。該方案使用深度學(xué)習(xí)技術(shù)實(shí)現(xiàn)老人跌倒行為的智能識(shí)別,基于WebRTC視頻傳輸架構(gòu)實(shí)現(xiàn)遠(yuǎn)程視頻傳輸。展開(kāi)了基于深度學(xué)習(xí)的跌倒行為識(shí)別的研究。首先,提出并仿真實(shí)現(xiàn)了基于視頻幀和VGGNet-16卷積神經(jīng)網(wǎng)絡(luò)模型的跌倒識(shí)別方法;對(duì)Le2i、SDU、UCF-101開(kāi)源視頻數(shù)據(jù)集,進(jìn)行水平翻轉(zhuǎn)、對(duì)比度和亮度調(diào)節(jié)、加噪等數(shù)據(jù)增強(qiáng)處理后構(gòu)建卷積神經(jīng)網(wǎng)絡(luò)的訓(xùn)練集和測(cè)試集,對(duì)上述方法進(jìn)行訓(xùn)練和測(cè)試。實(shí)驗(yàn)結(jié)果表明:該方法識(shí)別結(jié)果強(qiáng)烈依賴(lài)于訓(xùn)練場(chǎng)景。其次,針對(duì)上述的問(wèn)題,提出了一種基于雙流卷積神經(jīng)網(wǎng)絡(luò)的跌倒識(shí)別方法。該方法:一路采用場(chǎng)景相減法檢測(cè)視頻中的運(yùn)動(dòng)目標(biāo),將視頻幀中運(yùn)動(dòng)目標(biāo)加框標(biāo)記后輸入到3D-CNN模型中進(jìn)行跌倒識(shí)別;另一路采用光流法提取視頻的光流圖,將光流圖輸入到VGGNet-16模型中進(jìn)行跌倒識(shí)別;最后將兩路模型的跌倒識(shí)別結(jié)果進(jìn)行線(xiàn)性加權(quán)融合。實(shí)驗(yàn)結(jié)果表明:基于雙流卷積神經(jīng)網(wǎng)絡(luò)的跌倒識(shí)別方法跌倒識(shí)別率為96%,比基于視頻幀和VGGNet-16卷積神經(jīng)網(wǎng)絡(luò)模型的跌倒識(shí)別方法識(shí)別率提高了51%,比基于運(yùn)動(dòng)目標(biāo)檢測(cè)和3D-CNN的跌倒識(shí)別方法提高了4%,比基于光流圖和VGGNet-16的跌倒識(shí)別方法提高了3%,且有良好的泛化能力。展開(kāi)了基于WebRTC的遠(yuǎn)程視頻監(jiān)控的研究。提出了基于WebRTC的遠(yuǎn)程視頻監(jiān)控方案,搭建了視頻傳輸?shù)男帕罘⻊?wù)器和穿網(wǎng)服務(wù)器,并基于WebRTC實(shí)現(xiàn)了視頻采集端、遠(yuǎn)程視頻監(jiān)控端。通過(guò)搭建含NAT的網(wǎng)絡(luò)實(shí)驗(yàn)環(huán)境對(duì)系統(tǒng)進(jìn)行測(cè)試,實(shí)驗(yàn)結(jié)果表明:本文實(shí)現(xiàn)的信令服務(wù)器、穿網(wǎng)服務(wù)器工作正常,視頻傳輸客戶(hù)端功能正常,且能夠穿越防火墻和NAT的限制實(shí)現(xiàn)P2P的視頻傳輸。
[Abstract]:With the aggravation of the aging population and the increasing pressure of social life, the proportion of empty nest elderly is increasing. Living alone causes many social problems, especially the health problems of the elderly, among which the accidental fall and injury of the elderly is one of the main reasons affecting the health of the elderly. Failure to be dealt with in time after a fall can increase secondary injuries and even lead to accidental death. According to related statistics, more than 50% of the elderly fall at home. If the fall behavior of the elderly can be monitored in real time, it can be accurately identified when the fall behavior occurs, and send a warning message to their guardian. Timely assistance to the elderly will greatly reduce the harm of falls to the health of the elderly. Based on this, the research of intelligent fall monitoring system based on deep learning and WebRTC is carried out in this paper. The main work is as follows: based on the analysis of the functional requirements of the intelligent fall monitoring system based on deep learning and WebRTC, the technical scheme of the system is proposed. This scheme uses depth learning technology to realize the intelligent recognition of fall behavior of the elderly, and realizes remote video transmission based on WebRTC video transmission architecture. The research on recognition of fall behavior based on deep learning is carried out. Firstly, a fall recognition method based on video frame and VGGNet-16 convolution neural network model is proposed and simulated. The training set and test set of convolutional neural network are constructed after the Le2i,SDU,UCF-101 open source video data set is flipped horizontally, the contrast and brightness are adjusted, and the noise is increased. The above methods are trained and tested. The experimental results show that the recognition results are strongly dependent on the training scene. Secondly, a fall recognition method based on two-flow convolution neural network is proposed. This method uses scene subtraction method to detect moving target in video frame, then marks moving object in video frame, then inputs it into 3D-CNN model for fall recognition. On the other hand, the optical flow image of video is extracted by optical flow method, and the optical flow graph is input into the VGGNet-16 model for fall recognition. Finally, the fall recognition results of the two models are fused linearly. The experimental results show that the fall recognition rate based on two-stream convolution neural network is 96, which is 51% higher than that based on video frame and VGGNet-16 convolution neural network model. Compared with the moving target detection and the fall recognition method based on 3D-CNN, the proposed method has a better generalization ability than the optical flow graph and the fall recognition method based on VGGNet-16. The research of remote video surveillance based on WebRTC is carried out. A remote video surveillance scheme based on WebRTC is proposed. The signalling server and the netting server of video transmission are built. The video collection terminal and remote video monitoring terminal are implemented based on WebRTC. The experimental results show that the signaling server, the network piercing server, and the video transmission client function are working well, and the video transmission client function is normal. And can pass through the firewall and NAT restrictions to achieve P2P video transmission.
【學(xué)位授予單位】：華東交通大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類(lèi)號(hào)】：TP391.41;TP18

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 羅治情;陳娉婷;官波;沈祥成;;基于HTML5+WebRTC的農(nóng)業(yè)專(zhuān)家在線(xiàn)教學(xué)系統(tǒng)的設(shè)計(jì)與實(shí)現(xiàn)[J];農(nóng)業(yè)網(wǎng)絡(luò)信息;2016年03期

2 劉洪彬;常發(fā)亮;;權(quán)重系數(shù)自適應(yīng)光流法運(yùn)動(dòng)目標(biāo)檢測(cè)[J];光學(xué)精密工程;2016年02期

3 孫勁光;孟凡宇;;基于深度神經(jīng)網(wǎng)絡(luò)的特征加權(quán)融合人臉識(shí)別方法[J];計(jì)算機(jī)應(yīng)用;2016年02期

4 陳冠宇;安凱;李向;;基于卷積神經(jīng)網(wǎng)絡(luò)的不良地質(zhì)體識(shí)別與分類(lèi)[J];地質(zhì)科技情報(bào);2016年01期

5 趙曉健;曾曉勤;;基于稠密光流軌跡和稀疏編碼算法的行為識(shí)別方法[J];計(jì)算機(jī)應(yīng)用;2016年01期

6 彭博;臧笛;;基于深度學(xué)習(xí)的車(chē)標(biāo)識(shí)別方法研究[J];計(jì)算機(jī)科學(xué);2015年04期

7 於正強(qiáng);潘峗;宦若虹;;一種結(jié)合幀差法和混合高斯的運(yùn)動(dòng)檢測(cè)算法[J];計(jì)算機(jī)應(yīng)用與軟件;2015年04期

8 沈志豪;洪建光;沈?yàn)t軍;戚偉強(qiáng);;WebRTC技術(shù)在信息機(jī)房巡檢中的應(yīng)用[J];電力信息與通信技術(shù);2015年04期

9 付斌;楊鑫;王松;林鴻;;WebRTC技術(shù)研究及其應(yīng)用[J];電信科學(xué);2013年09期

10 屈振華;李慧云;張凌;楊新章;龍顯軍;;WebRTC關(guān)鍵技術(shù)分析及運(yùn)營(yíng)商引入策略研究[J];電信科學(xué);2013年S1期

相關(guān)碩士學(xué)位論文 前7條

1 薛冰霞;基于多模特征融合的人體跌倒檢測(cè)算法研究[D];山東大學(xué);2015年

2 譚貞剛;基于多分辨率卷積網(wǎng)絡(luò)的交通標(biāo)志檢測(cè)與識(shí)別[D];大連理工大學(xué);2015年

3 聶曉飛;基于WebRTC的跨平臺(tái)視頻會(huì)議系統(tǒng)的設(shè)計(jì)與實(shí)現(xiàn)[D];北京交通大學(xué);2014年

4 龔丁禧;稀疏自組合時(shí)空卷積神經(jīng)網(wǎng)絡(luò)動(dòng)作識(shí)別方法及其并行化[D];廈門(mén)大學(xué);2014年

5 田旭;基于WebRTC技術(shù)的遠(yuǎn)程面試系統(tǒng)的設(shè)計(jì)與實(shí)現(xiàn)[D];華中師范大學(xué);2014年

6 章焱;基于深度神經(jīng)網(wǎng)絡(luò)的橋牌識(shí)別系統(tǒng)研究[D];廈門(mén)大學(xué);2014年

7 謝靖;基于頭部運(yùn)動(dòng)軌跡和3D視覺(jué)的跌倒檢測(cè)系統(tǒng)[D];南京大學(xué);2011年

，

本文編號(hào)：2362637