[Graphic] MPEG-2 compression coding technology principle application (8)

Guo Bin, Professor of the Department of television engineering, Beijing Broadcasting Institute

3. Application of MPEG-2

The MPEG-2 can basically meet most of the needs of the broadcast and television system, such as: Suitable for interlaced and line-by-line scanning images; and image sampling; theoretically up to 16000 pixels
× 16000 lines of Multi-image decomposition force and common field and frame frequencies in broadcast; the hierarchical feature of encoding allows sdtv or ldtv decoder to extract the required information from the data stream of higher-level HDTV.
In order to adapt to the different needs of coding methods, operating modes, performance and price ratio in different occasions, at the March 1993 Sydney conference and the July New York Conference, basically identified the MPEG-2 type shown in table 5
And level specifications. "Type" is a subset of all encoding methods. It is encoded by compression.AlgorithmComplexity. In table 5, six subsets are defined from the right to the left. The "level" parameter is subject to different limitations of Type encoding parameters.
According to the image resolution level, in table 5, the pixel decomposition force of four levels is defined from top to bottom. Obviously, MPEG-2 standards have a wide range of universal, in order to meet the needs of a variety of different applications
Different video encoding algorithms are integrated into a single syntax. However, if the receiver decoder must all meet the video encoding algorithm specified in syntaxes, the decoder design will become complex and costly.
Encoder is neither possible nor necessary to implement all functions of the MPEG-2. Therefore, it is proposed that different applications should meet the requirements of different parts in syntaxes.

1. Type level concept of MPEG-2

to meet the needs of different scenarios, the concept of profile and level provides a way to define a subset of syntaxes. Based on a single syntax, according to the complexity of the compression encoding algorithm
and different applications, six syntax subsets are defined, and each syntax subset is one type. MPEG-2 type: simple type (SP), main type (MP), MPEG-2
type (added later), signal-to-noise ratio (snrp) there are 6 types of Space-level (SSP) and high-level (HP. In the same type, the image parameters to be processed, such as image size, frame rate, and
rate, are also different. For example, table 5 contains four different image sizes and four different bit rates. Only the frame rate is the same. To this end, the MPEG-2 also defines a low level (LL), Master level
(ML), high level 1440 (h14l), advanced (HL) a total of four levels, to show the differences between different parameters in the same type. Clearly, the model defines the classification of data streams and the color space decomposition force; The level defines the
image resolution and the maximum bit rate of each model. That is, each model defines a new set of algorithms, such as the nature of the model, the color format, and whether there are two-way frames. Different Combinations have different algorithms. Definition of each level in the same type
defines the value range of the parameters involved in the operation, such as the number of lines of images per frame, number of workers per line, frame rate, and bit rate. Visible from table 7, compared to the type and level of the MPEG-2 when the MPEG-2 standard passed in 1994,
added 1 MPEG-2 422 @ m l. Because the MPEG-2 m p @ M
L using color subsampling, multi-edition replication can only be 2 generations, does not comply with the ITU-R 601 standard. The use of can be multi-generation replication, also meets the requirements of the ITU-R 601 standard.

space in Table 5 is a combination that cannot appear or does not exist. From high to simple, the compression ratio increases from high to low. Table 5 lists the types and levels that have been defined, and lists available and levels. For example, the most common
descriptor is the primary level, which can be expressed as main profile @ Main level, abbreviated as MP @
ML. For NTSC videos, the resolution is equivalent to 720x480, the frame rate is 30fps, And the bit rate is less than 15 Mbps. Similarly, for a pal video,
is equivalent to a resolution of 720 × 576, with a frame rate of 25 FPS and a bit rate of less than 15 Mbps. Images for ITU-R 601, such as the United States using MP @
ML satellite live broadcast, digital video disk is also used. Likewise, MP @ HL can be used for advanced TVs (ATV), such as the use of MP @
HL indicators by the U. S. HDTV (GA); for MP @ h14l, it can be used for HDTV; the European Experiment HDTV adopts SSP @ h14l. For MP @
LL, it can be used for video calls or video conferences. (To be continued)