There are two sets of standards for frequency codec technology, ITU-T standards H.261, H.263, H.263+, etc.; and ISO MPEG standards Mpeg1, Mpeg2, Mpeg4, etc. H.264/AVC is the latest standard jointly introduced by the two organizations to combine the advantages of H.263+ and Mpeg4. The most valuable part is undoubtedly a higher data compression ratio. Under the same image quality conditions, H.264 data compression ratio is 2 times higher than H.263 and 1.5 times higher than MPEG-4.
This article refers to the address: http://
Below we briefly introduce the concept and development of H.264 and explore the possibility of practical application of H.264 technology.
What is H.264/AVC?
The H.264/AVC standard was jointly developed by ITU-T and ISO/IEC and is targeted at covering the entire video application, including: low-rate wireless applications, standard definition and high-definition television broadcasting applications, and the Internet. Video streaming applications, delivering high-definition DVD video and high-quality video applications for digital cameras.
ITU-T named this standard H.264 (formerly known as H.26L), and ISO/IEC calls it MPEG-4 Advanced Video Coding (AVC), and it will become the MPEG-4 standard. 10 parts. Since AVC is an extension of the current MPEG-4 standard, it will certainly benefit from the well-developed infrastructure of MPEG-4 (such as system layering and audio). Obviously, MPEG-4 AVC, which is MPEG-4 Advanced Simple Profile (ASP), will outperform the current MPEG-4 video compression standard, and it will be mainly used in applications with high compression and hierarchical quality. direction.
As seen in the "Video Coding History" table below, ITU-T and ISO/IEC are responsible for the customization of all previous international video compression standards. By far the most successful video standard is MPEG-2, which has been widely accepted in various market segments such as DVD, digital TV broadcasting (covering cable and communication satellites) and digital set-top boxes. Since the birth of MPEG-2 technology, the new H.264/MPEG-4 AVC standard has greatly improved the coding efficiency and quality. Over time, H.264/MPEG-4 AVC will replace MPEG-2 and MPEG-4 in many existing applications, including some emerging markets (such as ADSL video).
The evolution of digital video codec technology
International standards are usually developed by the International Organization for Standardization ISO based on the technical recommendations of the International Telecommunication Union ITU. The digital video codec standard has undergone several changes. The H264 standard has raised the moving image compression technology to a higher stage. Providing high-quality image transmission at a lower bandwidth is a bright spot for H.264 applications. The promotion and application of H.264 has high requirements for video terminals, gatekeepers, gateways, MCUs, etc., which will effectively promote the continuous improvement of video conferencing software devices in all aspects.
The core competitiveness of H.264
The most valuable part of H.264 is undoubtedly a higher data compression ratio. The basic principle of compression technology is to filter non-critical information in video files so that data can be transmitted over the network faster. Under the same image quality conditions, the data compression ratio of H.264 is 2-3 times higher than that of MPEG-2 used in current DVD systems, and 1.5-2 times higher than MPEG-4. Because of this, H.264 compressed video data requires less bandwidth and is more economical during network transmission.
When MPEG-4 requires a transmission rate match of 6 Mbps, H.264 only requires a transmission rate of 3 Mbps to 4 Mbps. We use transportation to make a more vivid metaphor: the same is to transport a large box with a truck. If MPEG-4 can reduce the weight of the box by half, then H.264 can reduce the weight of the box to 1/4 of the original. With the same weight on the car, H.264 has doubled the load on trucks than MPEG-2.
The cost of obtaining superior performance at H.264 is a significant increase in computational complexity, such as hierarchical design, multi-frame parametrics, multi-mode motion estimation, improved intra prediction, etc., which significantly improve prediction accuracy and thus obtain better than others. The standard is much better than the compression performance.
Increasing hardware processing power and continuously optimized software algorithms are the basis for H.264's popularity. As early as ten years ago, the CPU with a frequency of several tens of megabytes reached the top level. Today, the average desktop computer has a CPU clock speed of several gigabytes. According to Moore's Law, the capacity per unit area of ​​a chip doubles every 18 months, so the computational complexity added by H.264 is negligible compared to the performance improvement effect. What's more, the new calculation methods are endless, and the H.264 demand for processing speed is relatively relieved.
Comparison of H.264 and MPEG-4
At very low bit rates (32-128Kbps), H.264 has a performance multiplication effect compared to MPEG-4, ie: H.26L media streams of the same bit rate are compared to MPEG-4 media streams, H. The 26L has a gain of about 3 decibels (the image quality level is doubled). The 32Kbps H.26L media stream has a similar signal-to-noise ratio as the 128K MPEG-4 media stream. That is, under the same picture quality, the code rate of H.264 is only a quarter of that of MPEG-4.
H.264/AVC core technology overview
This new standard consists of the following processing steps:
Inter and intra prediction
Transform (and inverse transform)
Quantization (and inverse quantization)
Loop filtering
Entropy coding
A single stream of pictures composes a video that can be divided into 16x16 pixel "macroblocks" that simplify the processing of each step in the video compression algorithm. For example, a picture taken from a standard definition standard video stream solution (720X480) is divided into 1350 (45 x 30) macroblocks and then further processed at the macroblock level.
Inter prediction
Improved motion estimation. Motion estimation is used to determine and eliminate temporal redundancy between different pictures present in the video stream. When the motion estimation search is a picture according to the past direction, the encoded picture is referred to as a "P frame picture", and when the search is based on pictures in both the past and the future, the encoded picture is referred to as a "B frame picture". .
In order to improve coding efficiency, in order to include and separate motion macroblocks in the "H.264 Motion Estimation - Improved Motion Estimation" diagram, macroblocks are split into smaller blocks. Then, the motion vector of the previous or future picture is used to predict a given block. H.264/MPEG-4 AVC invented a smaller block with better flexibility and higher prediction accuracy in terms of motion vectors.
Intra prediction
Where motion estimation cannot be used, intra-frame estimation is used to eliminate spatial redundancy. The inner estimate predicts the current block by inferring neighboring pixels in neighboring blocks in different directions in a predefined set. The difference between the predicted block and the real block is then encoded. This method is unique to H.264/MPEG-4 AVC and is especially useful for flat backgrounds where space redundancy is often present.
Transformation
The motion estimation and the internal estimated result are transformed from the spatial domain to the frequency domain by transformation. H.264/MPEG-4 AVC uses an integer DCT4X4 transform. MPEG-2 and MPEG-4 use floating point DCT8X8 transform.
Smaller blocks of H.264/MPEG-4 AVC reduce blockiness and significant artifacts. The integer coefficient eliminates the loss of precision caused by floating point coefficient operations in MPEG-2 and MPEG-4.
Quantification
The transformed coefficients are quantized, reducing the amount of prediction of the integer coefficients and eliminating the high frequency coefficients that are not easily perceived. This step is also used to control the bit rate of the output to be maintained at a substantially constant constant.
Loop filtering
The H.264/MPEG-4 AVC standard defines a deblocking filtering process for 16X16 macroblocks and 4X4 block boundaries. In the case of macroblocks, the purpose of filtering is to eliminate artifacts caused by different motion estimation types (such as motion estimation and internal estimation) or different quantization parameters of adjacent macroblocks. In the case of block boundaries, the purpose of filtering is to eliminate artifacts that may be caused by differences in transform/quantization and motion vectors from neighboring blocks. Loop filtering modifies two pixels on the same side of a macroblock/block boundary by a content adaptive nonlinear algorithm.
Entropy coding
Prior to entropy coding, the 4X4 quantized coefficients must be reordered. The reordering serialized stream is created by selecting different scan types for the motion estimation or the internal estimation based on the prediction algorithm originally adopted by these coefficients. The scan type sorts these coefficients in order from low frequency to high frequency. Since the high-frequency coefficients tend to be mostly zero, the number of zeros can be reduced by using run-length coding, thereby efficiently achieving the purpose of entropy coding.
Entropy coding-serialization of coefficients
The entropy encoding step represents the motion vector, the quantized coefficients, and the macroblock header by the byte stream of the mapped symbols. Entropy coding is designed to represent frequently used symbols with a small number of bits, and more bits are used to represent symbols that are used infrequently.
3.00mm Board To Board Connector
Shenzhen CGE Electronics Co.,Ltd , https://www.cgeconnector.com