MPEG2视频编码标准ISO/IEC13818-2详解

5星 · 超过95%的资源需积分: 9 88 浏览量更新于2024-07-31 收藏 2.03MB PDF 举报

"MPEG2编码标准 ISO IEC13818-2" MPEG-2（Moving Picture Experts Group - 2）是国际标准化组织（ISO）和国际电工委员会（IEC）联合制定的一种数字视频编码标准，其具体规范被记录在ISO/IEC 13818-2文档中。这个标准主要应用于电视广播、DVD存储、有线电视以及数字视频录像等领域。第二版发布于2000年12月15日，旨在提供高效的视频压缩技术，以便在有限的带宽下传输或存储高质量的视频内容。 MPEG-2编码标准的主要目标是提高视频压缩效率，同时保持良好的图像质量。它采用了一系列先进的编码技术，包括： 1. 分帧编码：MPEG-2将视频流分割成I帧（帧内编码）、P帧（前向预测编码）和B帧（双向预测编码）。这种分帧方式允许对不同类型的帧进行不同程度的压缩，以优化带宽使用。 2. DCT（离散余弦变换）：通过将像素数据转换到频率域，可以消除空间冗余，从而实现数据压缩。 3. 运动补偿：通过对相邻帧之间的运动矢量进行估计，减少帧间相似性造成的冗余，进一步压缩数据。 4. 量化与熵编码：对DCT系数进行量化处理，并使用熵编码（如哈夫曼编码）来减少信息的统计冗余。 5. 键帧和预测：I帧用于初始化解码过程，因为它包含完整的图像信息；P帧和B帧则依赖于前一帧或后一帧的信息进行预测编码，从而节省存储空间。 6. 动态范围扩展：MPEG-2支持不同比特率和分辨率的视频流，以适应不同的传输环境和显示设备。 7. 复用与打包：MPEG-2标准还包括了音频和视频数据的复用，以及节目和系统信息的打包，这些信息可以一起传输，形成一个完整的传输流。 MPEG-2标准的广泛应用证明了其在数字视频编码领域的成功。尽管后来出现了更高级的编码标准，如H.264/AVC和H.265/HEVC，但MPEG-2因其兼容性和稳定性，至今仍在很多领域中被广泛使用。然而，需要注意的是，该标准的版权由Adobe Systems Incorporated持有，因此在使用此PDF文件时，需要遵守Adobe的软件许可政策，特别是在编辑内容时。ISO中央秘书处对这方面的责任不承担责任，下载文件的各方应确保不侵犯Adobe的版权政策。

ISO/IEC 13818-2 : 2000 (E)

6 ITU-T Rec. H.262 (2000 E)

3.103 progressive: The property of film frames where all the samples of the frame represent the same instances in

time.

3.104 quantisation matrix: A set of sixty-four 8-bit values used by the dequantiser.

3.105 quantised DCT coefficients: DCT coefficients before dequantisation. A variable length coded representation

of quantised DCT coefficients is transmitted as part of the coded video bitstream.

3.106 quantiser scale: A scale factor coded in the bitstream and used by the decoding process to scale the

dequantisation.

3.107 random access: The process of beginning to read and decode the coded bitstream at an arbitrary point.

3.108 reconstructed frame: A reconstructed frame consists of three rectangular matrices of 8-bit numbers

representing the luminance and two chrominance signals. A reconstructed frame is obtained by decoding a coded frame.

3.109 reconstructed picture: A reconstructed picture is obtained by decoding a coded picture. A reconstructed

picture is either a reconstructed frame (when decoding a frame picture), or one field of a reconstructed frame (when

decoding a field picture). If the coded picture is a field picture, then the reconstructed picture is the top field or the

bottom field of the reconstructed frame.

3.110 reference field: A reference field is one field of a reconstructed frame. Reference fields are used for forward

and backward prediction when P-pictures and B-pictures are decoded. Note that when field P-pictures are decoded,

prediction of the second field P-picture of a coded frame uses the first reconstructed field of the same coded frame as a

reference field.

3.111 reference frame: A reference frame is a reconstructed frame that was coded in the form of a coded I-frame or

a coded P-frame. Reference frames are used for forward and backward prediction when P-pictures and B-pictures are

decoded.

3.112 re-ordering delay: A delay in the decoding process that is caused by frame re-ordering.

3.113 reserved: The term "reserved" when used in the clauses defining the coded bitstream, indicates that the value

may be used in the future for ITU-T | ISO/IEC defined extensions.

3.114 sample aspect ratio (SAR): This specifies the relative distance between samples. It is defined (for the

purposes of ITU-T Rec. H.262 | ISO/IEC 13818-2), as the vertical displacement of the lines of luminance samples in a

frame divided by the horizontal displacement of the luminance samples. Thus, its units are (metres per line) ÷ (metres per

sample).

3.115 scalable hierarchy: Coded video data consisting of an ordered set of more than one video bitstream.

3.116 scalability: Scalability is the ability of a decoder to decode an ordered set of bitstreams to produce a

reconstructed sequence. Moreover, useful video is output when subsets are decoded. The minimum subset that can thus

be decoded is the first bitstream in the set which is called the base layer. Each of the other bitstreams in the set is called

an enhancement layer. When addressing a specific enhancement layer, "lower layer" refer to the bitstream which

precedes the enhancement layer.

3.117 side information: Information in the bitstream necessary for controlling the decoder.

3.118 16 ×

××

× 8 prediction: A prediction mode similar to field-based prediction but where the predicted block size is

16 × 8 luminance samples.

3.119 run: The number of zero coefficients preceding a non-zero coefficient, in the scan order. The absolute value of

the non-zero coefficient is called "level".

3.120 saturation: Limiting a value that exceeds a defined range by setting its value to the maximum or minimum of

the range as appropriate.

3.121 skipped macroblock: A macroblock for which no data is encoded.

3.122 slice: A consecutive series of macroblocks which are all located in the same horizontal row of macroblocks.

3.123 SNR scalability: A type of scalability where the enhancement layer(s) contain only coded refinement data for

the DCT coefficients of the lower layer.

3.124 source; input: Term used to describe the video material or some of its attributes before encoding.

ISO/IEC 13818-2 : 2000 (E)

ITU-T Rec. H.262 (2000 E) 7

3.125 spatial prediction: Prediction derived from a decoded frame of the lower layer decoder used in spatial

scalability.

3.126 spatial scalability: A type of scalability where an enhancement layer also uses predictions from sample data

derived from a lower layer without using motion vectors. The layers can have different frame sizes, frame rates or

chrominance formats.

3.127 start codes (system and video): 32-bit codes embedded in that coded bitstream that are unique. They are used

for several purposes including identifying some of the structures in the coding syntax.

3.128 stuffing (bits); stuffing (bytes): Code-words that may be inserted into the coded bitstream that are discarded in

the decoding process. Their purpose is to increase the bitrate of the stream which would otherwise be lower than the

desired bitrate.

3.129 temporal prediction: Prediction derived from reference frames or fields other than those defined as spatial

prediction.

3.130 temporal scalability: A type of scalability where an enhancement layer also uses predictions from sample data

derived from a lower layer using motion vectors. The layers have identical frame size, and chrominance formats, but can

have different frame rates.

3.131 top field: One of two fields that comprise a frame. Each line of a top field is spatially located immediately

above the corresponding line of the bottom field.

3.132 top layer: The topmost layer (with the highest layer_id) of a scalable hierarchy.

3.133 variable bitrate: Operation where the bitrate varies with time during the decoding of a coded bitstream.

3.134 variable length coding (VLC): A reversible procedure for coding that assigns shorter code-words to frequent

events and longer code-words to less frequent events.

3.135 video buffering verifier (VBV): A hypothetical decoder that is conceptually connected to the output of the

encoder. Its purpose is to provide a constraint on the variability of the data rate that an encoder or editing process may

produce.

3.136 video sequence: The highest syntactic structure of coded video bitstreams. It contains a series of one or more

coded frames.

3.137 xxx profile decoder: Decoder able to decode one or a scalable hierarchy of bitstreams of which the top layer

conforms to the specifications of the xxx profile (with xxx being any of the defined Profile names).

3.138 xxx profile scalable hierarchy: Set of bitstreams of which the top layer conforms to the specifications of the

xxx profile.

3.139 xxx profile bitstream: A bitstream of a scalable hierarchy with a profile indication corresponding to xxx. Note

that this bitstream is only decodable together with all its lower layer bitstreams (unless it is a base layer bitstream).

3.140 zigzag scanning order: A specific sequential ordering of the DCT coefficients from (approximately) the

lowest spatial frequency to the highest.

4 Abbreviations and symbols

The mathematical operators used to describe this Specification are similar to those used in the C programming language.

However, integer divisions with truncation and rounding are specifically defined. Numbering and counting loops

generally begin from zero.

4.1 Arithmetic operators

+ Addition

– Subtraction (as a binary operator) or negation (as a unary operator)

+ Increment, i.e. x+ + is equivalent to x = x + 1

–

– Decrement, i.e. x– – is equivalent to x = x – 1

Multiplication

ISO/IEC 13818-2 : 2000 (E)

8 ITU-T Rec. H.262 (2000 E)

^ Power

/ Integer division with truncation of the result toward zero. For example, 7/4 and –7/–4 are truncated to

1 and –7/4 and 7/–4 are truncated to –1.

// Integer division with rounding to the nearest integer. Half-integer values are rounded away from zero

unless otherwise specified. For example 3//2 is rounded to 2, and –3//2 is rounded to –2.

DIV Integer division with truncation of the result toward minus infinity. For example 3 DIV 2 is rounded

to 1, and –3 DIV 2 is rounded to –2.

÷ Used to denote division in mathematical equations where no truncation or rounding is intended.

% Modulus operator. Defined only for positive numbers.

Sign( ) Sign(x) =

1 x > 0

0 x =

= 0

–1 x < 0

Abs( ) Abs(x) =

x x >= 0

–x

x < 0

i = a

i < b

f (i) The summation of the f (i) with i taking integral values from a up to, but not including b.

4.2 Logical operators

|| Logical OR

&& Logical AND

! Logical NOT

4.3 Relational operators

> Greater than

>= Greater than or equal to

< Less than

<= Less than or equal to

= Equal to

!= Not equal to

max [, … ,] The maximum value in the argument list

min [, … ,] The minimum value in the argument list

4.4 Bitwise operators

& AND

| OR

>> Shift right with sign extension

<< Shift left with zero fill

4.5 Assignment

= Assignment operator

4.6 Mnemonics

The following mnemonics are defined to describe the different data types used in the coded bitstream.

bslbf Bit string, left bit first, where "left" is the order in which bit strings are written in this Specification.

Bit strings are generally written as a string of 1s and 0s within single quote marks, e.g. '1000 0001'.

Blanks within a bit string are for ease of reading and have no significance. For convenience, large

strings are occasionally written in hexadecimal; in this case, conversion to a binary in the

conventional manner will yield the value of the bit string. Thus, the left most hexadecimal digit is

first and in each hexadecimal digit the most significant of the four bits is first.

ISO/IEC 13818-2 : 2000 (E)

ITU-T Rec. H.262 (2000 E) 9

uimsbf Unsigned integer, most significant bit first.

simsbf Signed integer, in twos complement format, most significant (sign) bit first.

vlclbf Variable length code, left bit first, where "left" refers to the order in which the VLC codes are

written. The byte order of multibyte words is the most significant byte first.

In the clauses describing the syntax, any syntactic element that can only take positive or unsigned values (such as a flag

that can be equal to 0 or 1) is described with the mnemonic 'uimsbf'. If the syntactic element can have a negative value, it

is described with the mnemonic 'simsbf'. If the syntactic element has a constant value (e.g. marker_bit) then it is

described with the mnemonic 'bslbf'. If the syntactic element represents a variable-length code, it is described with the

mnemonic 'vlclbf'.

4.7 Constants

π 3,141 592 653 58...

e 2,718 281 828 45...

5 Conventions

5.1 Method of describing bitstream syntax

The bitstream retrieved by the decoder is described in 6.2. Each data item in the bitstream is in bold type. It is described

by its name, its length in bits, and a mnemonic for its type and order of transmission.

The action caused by a decoded data element in a bitstream depends on the value of that data element and on data

elements previously decoded. The decoding of the data elements and definition of the state variables used in their

decoding are described in 6.3. The following constructs are used to express the conditions when data elements are

present, and are in normal type:

This syntax uses the 'C-code' convention that a variable or expression evaluating to a non-zero value is equivalent to a

condition that is true and a variable or expression evaluating to a zero value is equivalent to a condition that is false. In

many cases a literal string is used in a condition. For example:

if ( scalable_mode =

= "spatial scalability" ) …

In such cases the literal string is that used to describe the value of the bitstream element in 6.3. In this example, we see

that "spatial scalability" is defined in Table 6-10 to be represented by the two bit binary number '01'.

while ( condition ) {

data_element

. . .

}

If the condition is true, then the group of data elements occurs

next in the data stream. This repeats until the condition is not true.

do {

data_element

. . .

} while ( condition )

The data element always occurs at least once.

The data element is repeated until the condition is not true.

if ( condition ) {

data_element

. . .

} else {

data_element

. . .

}

If the condition is true, then the first group of data elements occurs

next in the data stream.

If the condition is not true, then the second group of data elements

occurs next in the data stream.

for ( i = m; i < n; i+ +) {

data_element

. . .

}

The group of data elements occurs (m – n) times. Conditional

constructs within the group of data elements may depend on the

value of the loop control variable i, which is set to zero for the

first occurrence, incremented by one for the second occurrence, and so forth.

/* comment ... */ Explanatory comment that may be deleted entirely without in any

way altering the syntax.

ISO/IEC 13818-2 : 2000 (E)

10 ITU-T Rec. H.262 (2000 E)

As noted, the group of data elements may contain nested conditional constructs. For compactness, the {} are omitted

when only one data element follows:

data_element [n] data_element [n] is the n + 1th element of an array of data.

data_element [m][n] data_element [m][n] is the m + 1, n + 1th element of a two-dimensional array of data.

data_element [l][m][n] data_element [l][m][n] is the l + 1, m + 1, n + 1th element of a three-dimensional array

of data.

While the syntax is expressed in procedural terms, it should not be assumed that 6.2 implements a satisfactory decoding

procedure. In particular, it defines a correct and error-free input bitstream. Actual decoders must include means to look

for start codes in order to begin decoding correctly, and to identify errors, erasures or insertions while decoding. The

methods to identify these situations, and the actions to be taken, are not standardised.

5.2 Definition of functions

Several utility functions for picture coding algorithm are defined as follows.

5.2.1 Definition of bytealigned() function

The function bytealigned () returns 1 if the current position is on a byte boundary, that is the next bit in the bitstream is

the first bit in a byte. Otherwise it returns 0.

5.2.2 Definition of nextbits() function

The function nextbits () permits comparison of a bit string with the next bits to be decoded in the bitstream.

5.2.3 Definition of next_start_code() function

The next_start_code() function removes any zero bit and zero byte stuffing and locates the next start code.

This function checks whether the current position is byte aligned. If it is not, zero stuffing bits are present. After that any

number of zero stuffing bytes may be present before the start code. Therefore, start codes are always byte aligned and

may be preceded by any number of zero stuffing bits.

5.3 Reserved, forbidden and marker_bit

The terms "reserved" and "forbidden" are used in the description of some values of several fields in the coded bitstream.

The term "reserved" indicates that the value may be used in the future for ITU-T | ISO/IEC defined extensions.

The term "forbidden" indicates a value that shall never be used (usually in order to avoid emulation of start codes).

The term "marker_bit" indicates a one bit integer in which the value zero is forbidden (and it therefore shall have the

value '1'). These marker bits are introduced at several points in the syntax to avoid start code emulation.

next_start_code() { No. of bits Mnemonic

while ( !bytealigned() )

zero_bit 1 '0'

while ( nextbits() != '0000 0000 0000 0000 0000 0001' )

zero_byte 8 '0000 0000'

}

剩余219页未读，继续阅读

hurrypoint

粉丝: 0
资源: 8

MPEG2视频编码标准ISO/IEC13818-2详解

mpeg-2标准，iso 13818全

MPEG2-2(13818中文版).doc

《MPEG2-2(13818中文版)》

MPEG2编码标准ISO IEC13818-3.pdf

MPEG2编码标准ISO IEC13818-1.pdf

ISO/IEC_13818-3.pdf

音频AAC编码标准(ISO-IEC-13818-7-2006) pdf文档

ISO-IEC-13818-1_2000_MPEG2_Systems.pdf

MPEG2 AAC编码标准详解：ISO/IEC 13818-7 (2014)与MPEG4-AAC的关系

MPEG2（包含TS封装） ISO/IEC 13818-1 技术中文/英文 对照完整文档

最新资源

MPEG2（包含TS封装） ISO/IEC 13818-1 技术中文/英文对照完整文档