Selection of video encoding resolution

At present, the following resolutions are mainly used in the monitoring industry: SQCIF, QCIF, CIF, 4CIF.

The advantages of SQCIF and QCIF are that they have low storage capacity and can be used in narrow bands. Products using this resolution are inexpensive; the disadvantage is that the image quality is often poor and not accepted by users.

CIF is currently the mainstream resolution in the surveillance industry. Its advantage is that it has a low storage capacity, can be transmitted in ordinary broadband networks, and the price is relatively low. Its image quality is good and accepted by most users. The disadvantage is that the image quality cannot meet the requirements of high definition.

4CIF is standard definition resolution, and its advantage is that the image is clear. Disadvantages are high storage capacity, high network transmission bandwidth requirements, and high prices.

New choice of resolution-528x384

2CIF (704x288) has been adopted by some products to solve the shortcomings of insufficient clarity of CIF, high storage capacity and high price of 4CIF. However, since 704x288 is only an improvement in horizontal resolution, the improvement in image quality is not particularly obvious.

After testing, we found another 2CIF resolution of 528x384, better than 704x288 can solve the problem of CIF, 4CIF. Especially in the 512Kbps-1Mbps code rate, it can obtain stable high-quality images to meet the user's higher image quality requirements. At present, this resolution has been adopted by many network multimedia broadcasts and accepted by the majority of users. For example, Hangzhou Netcom's online cinema adopts 512x384 resolution, which can stably obtain image quality similar to DVD under 768k.

The best way to realize video encoding in surveillance

At present, video coding is in a period of rapid technological change, and the compression performance of video coding is constantly being improved.

ASCI and DSP are mainly used in monitoring. Because the design and production cycle of ASIC chip is too long, it cannot keep up with the development speed of video coding. The DSP chip, due to its universal design, enables it to implement various video encoding algorithms, and can update the video encoder in time to keep up with the development speed of video encoding. In addition, using the DSP chip can configure the encoder more flexibly than the ASIC, so that the encoder achieves the best performance.

Hikvision's current technical standards

Hikvision products use the most advanced H.264 video compression algorithm and high-performance DSP processor.

The powerful H.264 video compression engine enables the product to obtain extremely high compression ratio, high-quality image quality and good network transmission performance. The high-performance DSP processor can flexibly configure the video encoder / decoder: dynamically set the resolution, frame rate, code rate, image quality, etc .; it can output dual streams to achieve the functions of local storage and network transmission.

With TM130X DSP products, a single chip can compress one channel of the following resolution video in real time: SQCIF, QCIF, CIF, 2CIF (PAL: 704x288 or 528x384).

With DM642 DSP products, a single chip can compress 4 channels of video with the following resolutions in real time: SQCIF, QCIF, CIF, 2CIF (PAL: 704x288 or 528x384). A single chip can compress 2 4CIF videos in real time.

1. Basic concepts: embedded, real-time and multitasking

a. Embedded: software (including operating system and functional software) is integrated into the hardware system, which is simply a system with software and hardware integrated.
b. Real-time: respond to events within the prescribed time limit. The response that times out is a failed response.
c. Multi-tasking: respond to multiple requests simultaneously.
d. Real-time system and time-sharing system: real-time and non-real-time, time-sharing and non-time-sharing (exclusive).

2. Features of embedded hard disk video recorder

a. Embedded, real-time, multi-task equipment;
b. The software and hardware are highly professional and have no redundant functions;
c. The structure is simple and compact, and the volume is small.