In the rapid development of the video surveillance application industry chain, video applications have always followed the trend towards higher definition, higher frame rate, and higher compression rate. The video coding technology H.26L was first proposed in 1997 as a highly-compressed digital video encoder standard. This standard is now commonly referred to as H.264/MPEG-4 AVC, and is simply referred to as H.264. With this standard, the compression efficiency at the same image quality is more than double that of the previous standard. With its excellent encoding performance, it continuously provides smooth high-quality images and superior fault tolerance in unstable environments. In the past decade, it has swept the world and penetrated into every audio and video field. At the same time, it is indispensable for Fuze to go to the security industry. At the same time, with the application of video streaming network transmission technology, high-definition network monitoring began to develop rapidly, and the quality and sharpness of the network were rapidly updated. The 720P picture displayed for the surveillance room was relished a few years ago. Today, 1080P already monopolizes network monitoring. The era has spread to every corner of the streets.
With the development of technology, the pursuit of ultra-clear picture quality has also made the security industry focus on 4K. 4K LCD monitors have been launched in major brands, security manufacturers have also launched a front-end 4K cameras and back-end 4K network hard disk recorders, HDMI transmission protocol also supports 4K, so it seems that 4K monitoring system has been supporting the formation, ready to go. However, the problem has arisen, 4K picture is still large-scale stream of 16Mbps under H.264 coding technology, in today's various industry applications also face the stringent requirements of bandwidth, storage, energy consumption and real-time, so more efficient compression technology Standards should be born in the long-time - H.265/HEVC standard.
Two, H.265 coding techniques described
In 2013, the International Telecommunication Union ITU formally approved the H.265/HEVC standard. The standard is called High Efficiency Video Coding. Under the same picture quality, compared to H.264, H.265 encoded video. The stream size will be reduced by 30 to 50%.
1. Technical highlights of H.265 (HEVC)
As a next-generation video coding standard, HEVC (H.265) still belongs to a predictive-plus-transformation hybrid coding framework. However, compared to H.264, H.265 has revolutionized many aspects. The final requirement is to adopt a flexible coding structure.
In H.265, the size of the macroblock is expanded from 16x16 of H.264 to 64x64 to facilitate the compression of high-resolution video. At the same time, a more flexible coding structure is adopted to improve coding efficiency, including Coding Units, Predict Units, and Transform Units. as the picture shows:
Figure 1 CU, PU, â€‹â€‹CU
The coding unit is similar to the concept of macroblock in H.264/AVC and is used in the coding process. The prediction unit is the basic unit for prediction, and the transformation unit is the basic unit for transformation and quantization. The separation of these three units makes the transformation, prediction, and coding process more flexible. It also helps the division of each stage more in line with the texture features of the video image, which is beneficial for each unit to perform its functions more optimally.
The following figure shows an example. The left figure is the traditional H.264 standard. Each macro block size is fixed. The right picture is the H.265 standard. The coding unit size is determined based on the amount of area information.
Figure 2 Difference between H.264 and H.265 macroblock cutting
In the past, H.264 will be divided into several large blocks of the same size in 16 x 16 pixels (or 16 x 8, 8 x 8, 8 x 4, 4 x 4, etc.) The block is the smallest element when coding. H.265 is to cut the screen work from the user manually set, transferred to the encoder to determine, so that the encoder can be visually sized to 16X16, 32 x 32, 64 x 64, the screen is cut into several coding trees In general, the larger the block size, the better the compression efficiency.
It is through a series of optimization that H.265's video encoding compression technology can reduce the size of the same picture and quality video file by half, and the occupied bandwidth can also be reduced by half, providing a more perfect network bandwidth. Video compression solution. Under the same video picture, the video size encoded by H.265 saves 30% to 50% of the video stream size compared to H.264 encoding. As a result, the 4K video stream originally required 16 Mbps, and after encoding based on the H.265 technology, the 4K video stream only required 8 Mbps. The original 1080P stream, which was widely used in network HD systems, was also compressed from H.265 encoding from 4 to 6 Mbps, and was reduced to 2 to 3 Mbps. This means that with the construction of the MMIS vehicle wireless system in Shanghai rail transit, the 100 Mbps bandwidth can satisfy the simultaneous transmission of the maximum 30-40-channel on-vehicle video to meet the real-time demand for on-board video surveillance on the ground.
In the IEEE test, the H.265 encoding performance was tested on 9 videos. The best sample saved flow was 66.6%. The worst performing sample, the saving flow was also close to 30%, and the average reached 49.3%. At the same time, H.265 supports 4K (4096Ã—2160) and 8K (8192Ã—4320) pixels of Ultra HD video. The data volume of 4K video is 4 to 8 times that of HD, and it requires a more efficient compression encoding algorithm, while the H.265 encoding efficiency is twice that of H.264, which can well meet the encoding requirements of 4K video and save storage. Resources.
Figure 3 H.265 and H.264 stream bandwidth comparison
In the application of rail transit video surveillance, with the adoption of the Anti-Terrorism Law at the 18th Session of the Standing Committee of the Twelfth National People's Congress on December 27, 2015, Article 32 of the Regulations clearly states that: The management unit of the target shall establish a management system for on-duty monitoring of public safety video image information systems, information preservation and use, operation and maintenance, and ensure the normal operation of related systems. The retention period of the collected video and image information shall not be less than 90 days.â€ The venue contains key targets such as government agencies, banks, large city event venues, airports, train stations, and urban rail transit stations. The video recording must be adjusted to more than 90 days for 30 days or more. This is for the construction of a rail transit video surveillance system. The investment put forward higher demand. Taking Shanghai as an example, currently using H.264-encoded 1080P/25-frame video, with an average of 100 points per station, for 90 days of storage, requires 372T of hard-disk storage capacity. If H.265 encoding is used, then the storage device is put into use. It can be reduced to half.
Currently all major video surveillance vendors have launched mature H.265 products. Video compression in the field of video surveillance, H.265 replaced by H.264, mainly affects the compression module products and decoding module products, namely IPC, NVR, decoder, and platform software with decoding display. The key lies in the use of code chip ASIC or hardware acceleration unit for hard coding and hard decoding. Currently, mainstream chip manufacturers such as Hass, Anba, Qualcomm, SONY, NVIDIA, and Intel have proposed H.265 with 4K ultra-high-definition codec SOC or CPU+ASIC codec solutions. H.265-based IPC, NVR, platform and other product forms have become the industry standard.
III. Application of H.265 Technology in Shanghai Rail Transit
At present, the technical framework of the Shanghai rail transit high-definition video surveillance system is transmitted over an IP network. The GB-28181 SIP protocol is an inter-platform control protocol, ONVIF is a front-end device access protocol, and H.264 is an encoding protocol. Since H.265 is an iterative upgrade of H.264, upgrades and transitions only require the upgrade of the corresponding codec equipment and control protocol, which is a smooth, low-cost upgrade. GB-28181 â€œPublic Safety Food Surveillance Networking Systemâ€”Technical Requirements for Information Transmission, Exchange, and Controlâ€ is in the process of revision. It adds the H.265 unified coding provision and also extends the control protocol accordingly. The ONVIF (Open Network Video Interface Forum) also announced the profile draft of Profile T in July 2017. The draft supports advanced streaming media features, including support for H.265 video compression standards and extended feature sets, extending system integration. The ONVIF video configuration function of merchants and end users. This enables the Shanghai Rail Transit video surveillance system to introduce the H.265 video compression standard. It is only necessary to upgrade the existing control protocol to iterate over the control protocol.
Fourth, the conclusion
Application of H.265 coding technology Under the current background of increasing bandwidth requirements and storage requirements for security monitoring, it is possible to push security monitoring to a clearer and smarter chapter in a more efficient and cost-effective manner. Its inheritance of the previous H.264 system architecture also allows users to upgrade technology with minimal cost. Shanghai Rail Transit currently has the ability to upgrade to H.265. At the same time, with the increase in demand for 90 days of storage, on-the-spot navigation of on-board video surveillance systems, and even mobile video surveillance, there is also a need for technical upgrades.
Research topics: Shanghai Shentong Metro Group Co., Ltd. Zhou Ming Technology Center; Special Note: reprint will be required to indicate the author, and units Source
Nanpi Jiantong Hardware Manufacturing Co.,Ltd. offers a wide range of Stainless Steel Stamping Parts.
Material grade: stainless steel 301, stainless steel 304, stainless steel 316
brass H59, H62
galvanized steel sheet DX51D
Stock thickness ranges from 0.30mm-6.0mm for stainless steel
0.30mm-3.0mm for copper
0.30mm-6.0mm for aluminum
0.20mm-6.0mm for carbon steel
Surface finishing: polishing, satin blasted,powder coating, electro plating,anodize
Application: springs, kitchen ware, home appliance, tanks, motor housing, metal cover, caps, frame, assembly, steel base etc.
Metal Stamping Parts
Metal Stamping Parts,Metal Stamping,Custom Metal Stamp,Metal Stamping Tools
Nanpi Jian Tong Hardware Manufacturing Co., Ltd. , https://www.sheetmetals.nl