【摘要】：隨著網(wǎng)絡(luò)的迅速發(fā)展,要傳輸?shù)臄?shù)據(jù)種類也越來(lái)越多,其中流媒體數(shù)據(jù)是比重非常大的一部分,并且傳輸時(shí)需要較大的帶寬。選擇合理的傳輸方式非常重要。同時(shí)隨著監(jiān)控領(lǐng)域的發(fā)展,人們需要同時(shí)觀看多個(gè)采集設(shè)備發(fā)送過(guò)來(lái)的數(shù)據(jù),及播放端能夠同時(shí)解碼播放多路視頻,并且人們希望在播放視頻時(shí)提供一些智能化的功能。本文研究實(shí)現(xiàn)視頻多對(duì)多傳輸、播放端同時(shí)解碼播放多路視頻、視頻在播放時(shí)提供人臉檢測(cè)和人臉識(shí)別的功能,通過(guò)人臉識(shí)別用戶可以方便的檢測(cè)出視頻畫面中的人臉是否在某個(gè)人臉庫(kù)中或者是否是要尋找的人。本文的主要研究?jī)?nèi)容如下:1)視頻傳輸:一個(gè)發(fā)送端可以將視頻碼流同時(shí)發(fā)往多個(gè)接收端,一個(gè)接收端可以同時(shí)接收多個(gè)發(fā)送端發(fā)送來(lái)的視頻碼流。發(fā)送端在應(yīng)用層通過(guò)RTP協(xié)議將視頻碼流封裝成RTP報(bào)文,然后通過(guò)傳輸層使用UDP協(xié)議通過(guò)組播的方式將流媒體數(shù)據(jù)發(fā)送給組播組中的成員。接收端通過(guò)加入不同的組播組同時(shí)接收多個(gè)發(fā)送端發(fā)送來(lái)的視頻碼流。2)視頻的解碼播放:通過(guò)編寫Windows應(yīng)用程序,實(shí)現(xiàn)接收端同時(shí)接收多路視頻,并通過(guò)不同的線程同時(shí)進(jìn)行解碼播放,接收端能夠同時(shí)顯示多個(gè)發(fā)送端發(fā)送來(lái)的視頻碼流,并可以將通過(guò)RTP協(xié)議傳輸過(guò)來(lái)的H.264碼流保存成MP4格式的文件。播放本地視頻文件時(shí)能夠快進(jìn)快退并且音視頻能夠同步播放。3)人臉檢測(cè)和識(shí)別:Windows應(yīng)用程序在播放視頻時(shí),以一定的頻率對(duì)視頻畫面中的人臉進(jìn)行檢測(cè)并判斷畫面中的人臉是否在某個(gè)人臉庫(kù)中。通過(guò)判斷畫面中的人臉是否在某個(gè)人臉庫(kù)中來(lái)判斷某個(gè)人是否為要尋找的人或者是否有權(quán)限進(jìn)入某個(gè)區(qū)域。
[Abstract]:With the rapid development of the network, there are more and more kinds of data to be transmitted, in which streaming media data is a very large part of the transmission, and the transmission requires a large bandwidth. It is very important to choose a reasonable transmission mode. At the same time, with the development of surveillance field, people need to watch the data sent by multiple acquisition devices at the same time, and the player can decode and play the multi-channel video at the same time, and people hope to provide some intelligent functions when playing the video. In this paper, many-to-many video transmission is studied, and the player decodes and plays multi-channel video at the same time. The video provides the functions of face detection and face recognition when playing. Face recognition users can easily detect whether the face in the video image is in a certain face database or whether it is the person to be looked for. The main contents of this paper are as follows: 1) Video transmission: one transmitter can send the video stream to multiple receivers at the same time, and one receiver can simultaneously receive the video stream sent from multiple transmitting terminals. The sender encapsulates the video stream into RTP message through RTP protocol in the application layer, and then sends the streaming media data to the members of the multicast group by using the UDP protocol in the transport layer. By joining different multicast groups, the receiver simultaneously receives multiple video streams sent by the transmitting end. 2) Decoding and playing the video: by writing a Windows application program, the receiver can receive multiple video at the same time. And through different threads to decode and play at the same time, the receiver can display multiple video streams sent by the sending end at the same time, and the H.264 stream transmitted by RTP protocol can be saved as a MP4 format file. Local video files can be played fast forward and back, and audio and video can be played synchronously. 3) face detection and recognition: when the Windows application plays the video, The face in the video screen is detected at a certain frequency and whether the face in the screen is in a person's face database or not. Determine whether the face in the screen is in a person's database to determine whether a person is the person you are looking for or whether you have permission to enter a certain area.
【學(xué)位授予單位】：電子科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TN919.8;TP316.7

【參考文獻(xiàn)】

相關(guān)期刊論文 前8條

1 蘇獲;劉曉東;;基于OpenCV的智能視頻監(jiān)控告警系統(tǒng)的研究[J];電子設(shè)計(jì)工程;2016年03期

2 紀(jì)凌;李怡謙;;基于RTP協(xié)議流媒體識(shí)別算法的設(shè)計(jì)及實(shí)現(xiàn)分析[J];計(jì)算機(jī)安全;2014年02期

3 施培蓓;胡玉娟;;基于快速級(jí)聯(lián)分類的行人檢測(cè)系統(tǒng)[J];計(jì)算機(jī)工程;2013年08期

4 李校林;劉利權(quán);張杰;;基于RTP的H.264視頻流實(shí)時(shí)打包傳輸?shù)难芯縖J];計(jì)算機(jī)工程與科學(xué);2012年05期

5 陳玨奇;劉峰;;流媒體傳輸技術(shù)[J];計(jì)算機(jī)技術(shù)與發(fā)展;2012年03期

6 張瑩;李勇平;敖新宇;;基于OpenCV的通用人臉檢測(cè)模塊設(shè)計(jì)[J];計(jì)算機(jī)工程與科學(xué);2011年01期

7 李志泉;;對(duì)流媒體服務(wù)的分析[J];大眾科技;2009年08期

8 肖吉軍;陳利學(xué);;H.264在網(wǎng)絡(luò)視頻監(jiān)控系統(tǒng)中的應(yīng)用研究[J];軟件導(dǎo)刊;2009年06期

，

本文編號(hào)：2468184