Design and Implementation of Digital Audio and Video Surveillance System Based on IP Network

Monitoring systems are widely used in security, production management and other places where unattended operations are required. Although analog surveillance systems based on closed-circuit television have matured, the current emerging digital video surveillance systems have many advantages. The digital video surveillance system has high intelligence and reliability, can provide remote access capability, and is easy to store and manage video data, and can be developed and upgraded. This paper introduces a design and implementation scheme of digital video surveillance system based on IP network. It realizes the whole process digitization from acquisition and transmission to terminal control.

l Problems that the system needs to solve

1.1 Digital audio and video compression standards and selection of compression methods

The existing compression algorithm has H. 263 series, M-JPEG, MPEG-1 VCD compression standard, MPEG-2 DVD compression standard, WAVELET wavelet transform, MPEG-4 standard. These algorithms have their own advantages and disadvantages, and also determine their applicability in different industries, H. 263 is suitable for applications such as video telephony and video conferencing that do not require high image size and quality; MJPEG, MPEG-1, and MPEG-2 are not suitable for network transmission due to poor real-time performance and large data volume; MPEG-4 The emergence of video compression technology has led to a revolution in the field of compression. He basically overcomes the shortcomings of other compression algorithms, uses a very narrow bandwidth, compresses and transmits data through frame reconstruction technology, in order to obtain the best data with the least amount of data. Image Quality. MPEG-4 attempts to achieve two goals:

(1) Multimedia communication at a low bit rate;

(2) It is a synthesis of multimedia communication in multiple industries.

According to this goal, MPEG-4 introduces AV objects (Audio/Visaul Objects), making more interactive operations possible. Although MPEG-4 is not designed for video surveillance compression, it is also suitable for video compression in CIF (352×288) or higher resolution (768×576).

There are two ways to implement the compression algorithm, software compression and hardware compression. The hardware compression has good real-time performance and reliable performance. There are also dedicated MPEG-4 compression chips or boards on the market.

In order to meet the requirements of real-time, the system uses the hardware compression method of MPEG-4 compression algorithm.

1.2 Channel environment and real-time considerations

Currently available channels are PSTN, N-ISDN, Ethernet, etc. Most of the monitoring system applications use video surveillance in a relatively small area, so that the widely used Ethernet can be used as the channel for video transmission of the digital hard disk recording system. At present, the bandwidth of 100BASE-T Ethernet has reached 100Mb/s, which can meet the requirements of digital hard disk recording system to provide high-quality clear images and simultaneous transmission of multiple channels of video. Therefore, this paper chooses 100BASE-T Ethernet as the main transmission channel.

The task of this paper is mainly to solve the real-time transmission and multicast problems of digital video around Ethernet. Considering that in some applications, the video stream needs to be transmitted over a long distance. Therefore, when designing the network transmission system, the limitation of the channel bandwidth is fully considered, and the dynamic adjustment mechanism of the code stream and the frame rate is well adapted to meet the remote monitoring. The requirements for image quality and image continuity.

In order to achieve real-time performance, the optical and video capture part must be real-time, and the transmission part must meet real-time requirements. According to the test, MPEG-4 should reach 25 frames/s, and 256kb/s bandwidth is needed. 100Mb/s Ethernet can be seen. The network can meet the multiplexing requirements.

1.3 Control of network protocols and transport mechanisms

In the OSI network reference model developed by the ISO organization, the transport layer is built on top of the IP layer and contains two transport protocols: one is the transport control protocol TCP, which is a connection-oriented network protocol; the other is the user datagram protocol UDP. He is connectionless. Among them, TCP is not suitable for real-time transmission of audio and video data, and the commonly used UDP-based RTP protocol.

Since UDP has no error control and is unreliable packet delivery, in order to achieve reliable delivery and flow control, the IETF (Internet Engineering Department) proposes two protocols, RTP and RTCP. All real-time media data is transmitted using RTP, and RTCP provides the function of receiving feedback from the sender. They are all based on UDP.

2 system design

2.1 Functional design of digital transmission system network transmission

The system block diagram is shown in Figure 1.

He consists of 9 modules. The audio and video capture and compression processing is completed by the video capture card hardware. The capture card communicates with the network transmission module through the attached SDK function interface. When the video capture card completes the video capture and compression processing, the RTP protocol The encapsulation module encapsulates and sorts the data blocks, and then transfers them to the UDP network transmission module for transmission on the IP network. The work done on the receiving end is basically similar to that of the transmitting end, but is responsible for reorganizing and translating the audio and video data packets transmitted from the network. The code is played back.

2.2 System hardware components

Figure 2 shows the hardware components of the whole system, including the camera, the front-end acquisition computer and the central server. The front-end acquisition computer is equipped with a video capture card, and the corresponding number of cameras can be equipped according to the number of channels of the capture card.

2.3 Software Design

The system works in C/S mode and consists of three parts: acquisition, transmission, server display and control.

The software development of audio and video capture is based on the SDK provided by the capture card manufacturer. Since the size of the video data packet and the code stream affects the real-time nature of the video transmission in the network and the degree of jitter of the video during playback on the receiving end, the audio and video data packet size and code stream settings should be real-time during transmission and A compromise with the jitter situation during playback.

The streaming, encapsulation and transmission process of the transmitting end adopts the 32-bit operating system preemptive multi-threading task mechanism to solve the problem of low parallel efficiency of the CPU. The whole is divided into three buffers and multi-threaded structure, that is, adopting the stream buffer and the package buffer. Three buffers, such as the area and the send buffer, are allocated four threads, such as a stream encapsulation thread, a memory switching thread, a video image sending thread, and a program main thread, which utilize the stream buffer buffer, the stream buffer full, and the package. 7 events, such as buffer empty, package buffer full, send buffer empty, send buffer full and allow transmission, improve the efficiency of video image transmission.

When using the RTP protocol to encapsulate the video and audio composite stream, the common practice is to load the dynamic link library (DLL) of the RTP protocol in the Windows operating system, and then perform corresponding framing on the data stream output by the video encoder of the transmitting end. The algorithm forms a video stream package suitable for the RTP protocol format, submits it to the RTP protocol packet processing module, adds the packet header of the protocol, and adds a time stamp according to the current sampling clock, marks the sequence number, and gives a frame rate. Parameters such as resolution, corresponding compression format, etc. are completed by multi-address transmission. At the receiving end, when the real-time video data arrives, the header of the layer protocol is removed, and the tier is applied to the upper layer according to the nickname applied by the socket. The RTP packet module processes the submitted data packet, performs authentication according to its session identifier and serial number, and delivers the valid packet to the corresponding decoding buffer to implement synchronization within the video stream.

In order to avoid the broadcast storm, IP multicast is implemented on the PC platform. For this purpose, an IP multicast class CMulticastSocket based on the Microsoft base class library MFC is tailored. The IP multicast class CMulticastSocket is derived from the asynchronous Socket class CAnsycsocket. Each member of the group can dynamically join and quit. The information sent by a member of the group, all other authorized members in the group. Can receive, he is a branch of UDP Sockets.

Since the digital hard disk recording system (DVR) also needs to provide network control functions and transmission system information to the client, in a specific network programming application, a programming mechanism of UDP Socket and TCP Socket coexistence is adopted.

3 Performance indicators

The performance indicators of this system are as follows:

  • Supports simultaneous transmission of 8-channel CIF format video in 100Base-T Ethernet environment;

  • Support 50 remote clients to access at the same time;

  • On the client side, the delay of the network video and audio transmission is less than 1 000 ms, and there is no obvious jitter, and the frame rate of each video reconstructed by the client is greater than 25 frames/s;
  • Supports video transmission in a QCIF format under PSTN line conditions, and the video frame rate reconstructed by the client is greater than 5 frames/s;

  • Support alarm return, telephone line automatically alarms, automatic recording of alarm events.

    4 Conclusion

    The audio-visual monitoring system based on IP network and using universal computer combined with video capture card introduced in this paper has been successfully applied to unattended large-scale warehouses. The usage shows that its performance is good. In the future, with the development of related technologies, embedded digital monitoring systems based on IP networks with more powerful functions and smaller size will be more and more widely used.



  • :
    0 times
    Window._bd_share_config = { "common": { "bdSnsKey": {}, "bdText": "", "bdMini": "2", "bdMiniList": false, "bdPic": "", "bdStyle": " 0", "bdSize": "24" }, "share": {}, "image": { "viewList": ["qzone", "tsina", "tqq", "renren", "weixin"], "viewText": "Share to:", "viewSize": "16" }, "selectShare": { "bdContainerClass": null, "bdSelectMiniList": ["qzone", "tsina", "tqq", "renren" , "weixin"] } }; with (document) 0[(getElementsByTagName('head')[0] || body).appendChild(createElement('script')).src = 'http://bdimg.share. Baidu.com/static/api/js/share.js?v=89860593.js?cdnversion=' + ~(-new Date() / 36e5)];

    Bearing Rollers

    Needle Rollers For Axle,Needle Rollers For Drive Shaft,Bearing Needle Rollers,Bearing Steel Cylinderical Rollers

    Qingdao Yukun Bearing Needle Roller Co.,Ltd. , https://www.ykneedleroller.com