Five common problems and solutions for high-definition network monitoring

In terms of transmission and storage, system integrators can solve the problem of high cost investment; High definition manufacturers need to invest some effort in improving the underlying software and hardware for display and management; The investment in improving night vision is relatively small, and it is also the easiest solution to increase user satisfaction.

Compared with standard definition video surveillance, high-definition network surveillance has obvious advantages, but after more than two years of market promotion, it is inevitable to encounter the following problems:

1. Transmission aspect

High definition network cameras generally use 720P or 1080P solutions. To achieve ideal image effects, a network bandwidth of 2-8M is required. If a system has hundreds of points, the data flow of the entire network will be quite large, and the video stream is a real-time data stream, which will continue to put continuous pressure on the network related switching equipment. This requires sufficient homework in network planning.

Solution:

Due to the continuous reduction of fiber optic cable costs, local area networks can be easily composed of fiber optic cables and high-performance switches to form high-speed networks, with gigabit to desktop connections being a breeze. With the popularization of broadband networks, especially the "three networks in one" just launched this year, the speed of entering the home has reached tens of megabytes to hundreds of megabytes. The light into copper and back out finally brings us cheap high-speed Internet. For remote transmission of high-definition video images, it no longer takes too long.

2. Storage aspect

The previous question mentioned the issue of massive data flow. If data is aggregated to the center or stored at various nodes, it poses a significant challenge to the bandwidth pressure and storage capacity at the center entrance.

Solution:

At present, heavy investment can only be made in IPSAN and storage media. It is best to use an embedded architecture for IPSAN, with fewer plug-in boards and high stability. However, the disadvantage is poor compatibility and less flexibility in secondary development. Most NVR products on the market are based on X86 architecture, which means adding an array card to the server, installing application software, and combining the server and disk array into one for better compatibility. If using IPSAN, the storage medium needs to use enterprise grade hard drives in order to fully leverage the advantages of IPSAN in data storage speed and security. For some small and medium-sized systems, if the storage time is not too long and the data security requirements are not too high, mainstream servers in the market can be used to complete the storage function.

3. Display aspect

Many high-definition manufacturers only focus on the pixels of the front-end camera, without developing corresponding high-definition video decoders. They can only watch high-definition images around the screen of the PC, and cannot display high-definition images independently on the large screen.

Solution:

For high-definition images on the TV wall, we have developed a high-definition decoder that effectively restores the high-definition graphics transmitted from the front-end, connects to the high-definition input interface of the large screen, and perfectly displays the high-definition images on the screen.

4. Management aspect

Many high-definition manufacturers still follow the traditional approach of analog cameras to make high-definition cameras, buying ready-made hardware solutions and using third-party standard algorithms provided by MPEG4. They only provide a network interface for simple digitization. As for how the platform manages the hardware and data streams of the camera later, due to the lack of underlying core algorithms and technologies, it is simply impossible to achieve.

Solution:

Choose front-end devices with rich hardware interfaces as much as possible, such as alarms, SD cards, intercoms, audio, 485, etc. In the project plan, flexible combinations can be made according to the requirements of Party A, without the need for additional customized functions. In terms of software, there should be multiple access methods, such as WEB, client, large platforms, etc. SDK development packages should be provided to flexibly integrate with third-party software. The functional modules should be as complete as possible, such as alarms, streaming media, storage services, etc., which can be directly used in the project.

5. Nighttime aspect

It is very easy to achieve high definition during the day when there is sufficient light, but once it is applied to situations where the light is weak or night monitoring is needed, high definition becomes a useless place for heroes. Due to the fact that most high-definition cameras use CMOS sensors, their sensitivity to light is low. Without infrared fill light, they will be pitch black at night, and the time period we need to guard against most of our surveillance is precisely at night. This will make the highly anticipated high-definition video system a mere decoration.

Solution:

We have accumulated rich experience in night vision monitoring and provide different solutions according to different on-site environments. Such as infrared fill light scheme, IRCUT technology, low light chips, etc.

Summary:

Behind our obsession with high-definition products, we should analyze the five practical problems mentioned above more calmly. How to solve each problem is something that many manufacturers rarely consider. Their focus is often limited to the product itself, thinking that as long as the pixel is high, it is high-definition and a high-end system solution. The solutions provided are only theoretically feasible and cannot be truly implemented in practical projects. Some even upgrade the front-end with only 1.3 million pixels to 3 or even 4 million pixels through software processing.