High-speed rail communication command system video surveillance requirements analysis solution

High-speed railway is different from general railway system. High-speed railway itself is a systematic and integrated large-scale project. The communication department alone involves more than 10 subsystems, including wired, data, transmission, dispatching, emergency communication, video surveillance, etc. High-speed railway is very different from ordinary railway or subway. For example, subway usually runs at a speed of about 60 kilometers per hour, and the train interval is about 3 minutes, while the speed of high-speed railway may reach 300 kilometers per hour, but the time interval may be similar to that of subway. This puts high demands on the communication and command system of high-speed railway. At the same time, as an important auxiliary facility, the requirements for video surveillance system are also very high.

1. Features of high-speed rail video surveillance system

The video surveillance system of high-speed rail requires the use of advanced video surveillance technology, based on the IP network of the railway system, to build a digital, intelligent, and distributed network video surveillance system to meet the needs of business departments such as public security, safety supervision, passenger transport, dispatching, vehicle management, locomotive maintenance, engineering, electrical, vehicle, power supply, and disaster prevention monitoring, rescue and emergency management, and to achieve the sharing of video network resources and information resources. The video surveillance system of high-speed rail is generally based on a network architecture to realize the functions of video acquisition, encoding, compression, storage, forwarding, and virtual matrix. The video signal collected by the camera is connected to the DVR or encoder through a coaxial cable to realize video acquisition, encoding, compression, and transmission. The control signal of the PTZ camera is transmitted through RS485; the encoder sends the video stream to the NVR through the network for centralized storage and backup; the storage server can focus on backing up the video data of the DVR or NVR; the streaming media server can perform centralized video forwarding when multiple users visit to reduce the pressure on the network and front-end equipment; the decoder is connected to the TV wall to realize the centralized large-screen display restoration of the video.

2. Video Analysis Technology

The characteristics of high-speed rail are large system span and wide geographical distribution. The environment of video analysis is complex. Phenomena such as wind, frost, rain, snow, fog, camera shaking, train lights, city lights, insects, and cloud shadows are all problems that may be encountered in video analysis. A good VCA system should be able to balance the problem between missed reports and false reports. Railways are different from laboratories. Any adjustment of the camera angle, focal length, etc. requires certain manpower and material resources. Video analysis has high requirements for the scene (FOV), and it needs to be adjusted and confirmed in the future configuration. Therefore, it is not difficult to understand that the video analysis cameras of most video surveillance systems also use PTZ cameras instead of fixed cameras. After the analysis mode is fixed and the camera FOV is adjusted, it is necessary to enter the analysis settings. Usually, only one mode can be used for a camera video. In railway applications, there are two main VCA modes. One is to set up intrusion detection in important sections and throat areas to identify people or animals invading the high-speed rail tracks (most of the high-speed rail lines are closed or protected by fences and other physical methods, but there is still a possibility of intrusion); high-altitude falling object analysis is set up in the road-to-rail area to prevent high-altitude falling objects from affecting train operation. At present, these two video analysis application modes have been used in railway video surveillance and have performed well.

At present, there are two main architectures for video analysis technology, one is based on the back-end server, and the other is the front-end DSP method (DVS or IPC). In the DSP method, that is, the distributed intelligent analysis architecture, the video analysis unit is generally located near the video acquisition device, so that the system can be set up selectively so that the system only transmits the video to the control center or storage center when an alarm occurs. Compared with the server method, it can save network burden and storage space. Video analysis is a complex process that requires a large amount of system computing resources, so the number of ways that the server method can perform video analysis at the same time is very limited. For the above reasons, the mainstream video analysis technology on the market currently adopts the DSP method, based on cameras or encoders. It should be noted that for video analysis equipment based on the front-end DSP method, once the video analysis point needs to be adjusted, such as adding or canceling the video analysis function, it is usually necessary to replace the DVS or IPC, while the mode based on the back-end analysis can be adjusted directly in the machine room or control center without replacing the front-end hardware. Usually, the cost of replacing the DVS or IPC method is very high in railway projects.

3. Redundancy technology (reliability)

High-speed rail video surveillance systems need to be highly reliable in order to support the various needs of different departments, such as operation, security, and maintenance. For network video surveillance systems, different redundancy methods can be used to enhance stability. Redundancy technology can be implemented from front-end encoders, transmission networks to forwarding servers, management servers, and storage systems.

Encoder

The encoder can use dual power supplies and dual network cards to enhance stability; or use "N+1" redundancy to enhance reliability to ensure that the system can continue to operate when a single machine fails or the device is replaced.

NVR

NVR can use "N+1" redundancy to enhance reliability, to ensure continuous operation of the system when a single machine fails or equipment is replaced.

CMS

CMS can use "dual-machine redundant hot standby" to enhance stability and ensure that the system runs without downtime.

storage

The storage system can use RAID technology of disk array to achieve high-reliability data storage.

As shown in the fully redundant network video surveillance system architecture, redundancy is implemented in the front-end equipment IPC, DVS, transmission network, NVR, storage device NVR, and core management platform CMS, ensuring 24-hour high-reliability operation of the system and reducing system downtime or data loss caused by network, power supply, hardware, storage, software, etc. failures.

IPC is a single device. The way to improve stability is to perform local storage when the network is temporarily interrupted. For a large number of DVS, the "N+1" method can be used for redundancy to prevent the failure of the hardware and software of a single device. For NVR, the "N+1" method can be used for backup to prevent the failure of the hardware and software of a single device. For storage devices, mature RAID technology can be used to achieve redundancy protection. For the network, "dual networks" can be used to achieve high-reliability data transmission. For CMS, a dual-machine hot standby method is used, and the two machines share the RAID disk array to achieve redundancy. [page]

Key technologies of high-speed rail video surveillance system

1. Coding and compression technology

Video coding compression is the premise and foundation of network video surveillance. Uncompressed video data is massive. At present, the typical video coding compression standards are MPEG-4 and H.264. The MPEG-4 standard still uses the basic coding framework similar to the previous standards (H.261/3 and MPEG-1/2), that is, the typical three steps: predictive coding, transform quantization and entropy coding. New compression coding standards are designed based on the idea of ​​optimization, and some technologies in previous standards are improved. For example, motion compensation technology with 1/4 and 1/8 pixel accuracy is proposed on the basis of the original, which greatly improves the performance of predictive coding. The MPEG-4 standard not only provides specific compression algorithms, but is formulated for the needs of integration and compression technology such as digital television, interactive multimedia applications, and video surveillance. MPEG-4 integrates multiple multimedia applications in a complete framework and provides corresponding categories (Profile) and grades (Level) for different applications. H.264, which is also the tenth part of the MPEG-4 standard, is a high-compression video coding standard proposed by the Joint Video Team (JVT) jointly formed by the ITU-T Video Coding Experts Group (VCEG) and the ISO/IEC Moving Picture Experts Group (MPEG). Like previous standards, H.264 also adopts a hybrid coding mode of predictive coding plus transform coding. It concentrates the advantages of various previous coding standards and absorbs the experience accumulated in the process of standard formulation, and obtains much better compression performance than other previous coding methods. The biggest advantage of the H.264 standard is its high data compression ratio. Under the premise of the same image quality, the compression ratio of H.264 coding is 1.5 to 2 times that of MPEG-4. H.264 adopts a "network-friendly" structure and syntax, which is conducive to the processing of bit errors and packet loss to meet the needs of different rates, different resolutions, and different network transmission and storage occasions.

In network video surveillance systems, the ability of mutual encoding and decoding between products is very important, which can reduce the difficulty of system integration, facilitate expansion and protect costs. At present, the mutual encoding and decoding of equipment from different manufacturers is not ideal, that is, the compatibility is poor. The main reason is that different manufacturers use different encoding profiles and levels, and the data encapsulation format is incompatible, followed by the addition of private information (in the basic code stream/packet).

4. Video storage application technology

Storage deployment should be flexible, and you can choose alarm-triggered storage, preset schedule storage, manual start-stop storage, etc. The storage architecture should be mainstream architecture such as DAS, NAS, SAN, etc., and generally use RAID5 redundancy. Storage requirements generally require that normal recording and alarm recording be separated and set to different periods, such as 7 days for normal recording and 30 days for alarm recording. The planning and design of the storage system should be deployed according to project requirements and network conditions.

It should support manual backup storage of video image information; video information recording storage, event-triggered storage and planned storage functions; the system should support users to retrieve and play back according to various conditions such as time, location, event, etc.; the system should support multiple users to call and retrieve video images at the same time; DVR and NVR can work in multiple storage and archiving modes; the system should support downloading to local playback and remote direct playback of recordings.

Video archiving storage services (secondary storage, alarm storage) have the following features:

The storage architecture is a completely independent secondary architecture;

Archive Server is usually divided into "important video archive" and "alarm video archive";

Archive Server can be deployed anywhere on the network;

One archive server can archive multiple DVRs or NVRs;

·Users do not need to specify the playback source from DVR, NVR or Archive Server, the system will automatically index;

You can choose to archive certain channels of a DVR or NVR.

5. Comprehensive video platform

The system software platform includes the core data part and the client workstation. The core data server contains the system database and core software, and the workstation system should include functions such as user management, authority management, configuration management, fault management, and log management. The platform should support monitoring and maintenance of various video equipment resources and the operating status of the equipment, configuration of system equipment parameters through software, and management of system user registration, authentication, deletion, and authority allocation. The system should have a log management function, which can record the time when operators enter and exit the system and the main operation conditions through the operation log, and support functions such as log information query and report generation.

Data management: the types, versions, quantities and numbers of all equipment on the entire line are centrally "registered" in the database; monitoring function: real-time image calls can be made for all monitoring points on the entire line through the network, providing a reference for leaders to display during emergency command and decision-making and dispatch; network management function: monitor and manage the operation of network element equipment in all monitoring systems on the entire line, and display the operation data and alarm information of various equipment in real time; alarm management: record and display the alarms of equipment, system and communication parts, record the alarm processing status, and can export it; control function: PTZ control and relay output node control of all monitoring points on the entire line; voice intercom: for monitoring points with voice equipment installed at the front end, real-time voice intercom and one-way "shouting" function are realized; data storage: supports local storage and central backup of videos, with a variety of automatic and manual storage methods.