CN1980395A - Entropy coding method and decoding method based on text - Google Patents
Entropy coding method and decoding method based on text Download PDFInfo
- Publication number
- CN1980395A CN1980395A CN 200510126222 CN200510126222A CN1980395A CN 1980395 A CN1980395 A CN 1980395A CN 200510126222 CN200510126222 CN 200510126222 CN 200510126222 A CN200510126222 A CN 200510126222A CN 1980395 A CN1980395 A CN 1980395A
- Authority
- CN
- China
- Prior art keywords
- coefficient
- nonzero coefficient
- numerical value
- nonzero
- context model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Abstract
The method includes following steps: scanning DCT coefficient blocks according to sequence so as to obtain last one nonzero coefficient (NC) and the coefficient in front of the last one NC in coefficient block, and number value of NC; based on the number value of NC to calculate difference value between number value of NC in this block and number value of NC in adjacent block, and constructing context model for number value of NC; calculating flag bit of NC for last one NC and coefficients before the last one NC; selecting context model to encode flag bit of NC; subtracting 1 from absolute value of NC according to scanning sequence or scanning inverted sequence; and selecting context model to encode number value of NC, and its position in the scan etc; encoding sign bit. The invention also discloses corresponding decoding method.
Description
Technical field
The present invention relates to a kind of context modeling method that is used for Signal Compression DCT coefficient self adaptation arithmetic entropy coding, and provided a kind of novel coding/decoding system, be applicable to the entropy coding of various image/videos in the signal processing field/audio frequency dct transform data.
Background technology
Video coding technique is key technologies for application such as digital media storage and transmission.Core technology in the video coding system comprises predictive coding, transition coding, entropy coding.Wherein, entropy coding is to the syntactic element that produces in the cataloged procedure, carries out the comentropy coding as motion vector (mv), macroblock coding masterplate (cbp), piece prediction residual coefficient etc., and its effect is a statistical redundancy of removing data.
H.261, the 2D-VLC entropy coder of Moving Picture Experts Group-2 current video encoding standard has the coding method of piece prediction residual coefficient:; H.263, the 3D-VLC entropy coder of MPEG-4 standard and H.264/AVC the CAVLC entropy coder and the CABAC entropy coder of video encoding standard.
The 2D-VLC entropy coder adopts (level to the coding of residual error coefficient, run)+coded system of EOB, level wherein represents the value of nonzero coefficient, and run represents the quantity of the zero coefficient between current nonzero coefficient and the previous nonzero coefficient, and EOB represents end of block flag.EOB monopolizes a position in code table, promptly EOB distributes a code word separately.The 3D-VLC entropy coder is combined in EOB that (level run) counts centering, forms three-dimensional amount (level, run, last), vacateed like this that EOB occupies in the code table than the short code word, and utilized the joint probability of level, run and EOB, so code efficiency there is raising.The CAVLC entropy coder of JVT video encoding standard has utilized contextual information fully owing to it, and has carried out good optimization at 4 * 4 block sizes, opens coding though therefore level, run are divided, and has still obtained good coding result.Above entropy coder belongs to variable length encoder, its basic skills is the Huffman coding, though performance increases, do not make full use of the statistical probability (no adaptivity) of source symbol, simultaneously can't efficient coding greater than 0.5 source symbol for probability.
H.264/AVC the CABAC entropy coder of video encoding standard is a kind of adaptive arithmetic encoder, the performance of arithmetic encoder is higher than variable length encoder, because CABAC can utilize context to carry out modeling better, the probability of occurrence of source symbol had more accurately estimate that code efficiency is very high.Although than variable length encoder, the computation complexity height of arithmetic encoder, hardware is realized complicated.Arithmetic encoder its efficiently coding efficiency be subjected to the attention of more and more industrial quarters and academia: current nearly all audio/video coding standard is (as JPEG, JPEG-LS, MPEG Audio/Video, H264 etc.) all comprised the expansion technique of arithmetic entropy coding, JPEG 2000 has then directly adopted the self adaptation two-value arithmetic coding.
Utilizing context to carry out modeling is that arithmetic encoder has the very key of high coding efficiency, and therefore, the quality of context modeling method has very big influence to the code efficiency of arithmetic encoder.In video encoding standard H.264, the process to context modeling is described, and is in the present invention to propose a kind of new hereinafter modeling method different with last H.264 video encoding standard.
Summary of the invention
The objective of the invention is to overcome the statistical probability that does not make full use of source symbol in variable-length encoding/coding/decoding method, the shortcoming that code efficiency is lower provides a kind of based on contextual adaptive entropy coding method and coding/decoding method.
To achieve these goals, the invention provides a kind ofly, comprising based on contextual adaptive entropy coding method:
1), according to scanning sequency scanning DCT coefficient block, obtain all coefficients before last nonzero coefficient and this coefficient in the current DCT coefficient block, and the individual numerical value of nonzero coefficient;
2), obtain the individual numerical value of the nonzero coefficient of current DCT coefficient block by step 1), ask the difference of a current block nonzero coefficient numerical value and an adjacent block nonzero coefficient numerical value, and the amount of information of the nonzero coefficient numerical value of adjacent block changed into some intervals, for each interval structure and the relevant context model of a nonzero coefficient numerical value, so that the difference of a nonzero coefficient numerical value is encoded;
3), last nonzero coefficient that step 1) is obtained and each coefficient before this coefficient ask for the nonzero coefficient flag bit, if above-mentioned coefficient nonzero coefficient, then establishing the flag bit corresponding with coefficient is 1, if above-mentioned coefficient zero coefficient, then establishing flag bit is 0;
4), according to the scanning sequency of each coefficient in scanning process, select context model for the pairing nonzero coefficient flag bit of each coefficient and encode, the position in the scanning sequency at described context model and this coefficient place is relevant;
5), the absolute value of nonzero coefficient is subtracted 1 by the backward of scanning sequency or scanning, and, absolute value utilize in macro block (mb) type, the piece nonzero coefficient number and the choice of location context model in scanning to encode its sign bit of encoding again after the nonzero coefficient absolute value of having encoded subtracts 1 for subtracting 1 result.
In the technique scheme, before the absolute value to the difference of a described nonzero coefficient numerical value, nonzero coefficient subtracts 1 value coding, they are made binary conversion treatment, be converted into binary number.
The scanning of the described DCT coefficient in the described step 1) is a process that two dimension or multidimensional DCT coefficient is arranged as one-dimensional sequence.
Described scanning sequency is zigzag scanning sequency or vertical only scanning sequency or horizontal only scanning sequency.
Described step 2) in, the difference of a described current block nonzero coefficient numerical value and an adjacent block nonzero coefficient numerical value is: the individual numerical value of current block nonzero coefficient deducts the average of a nonzero coefficient numerical value of two adjacent blocks, resulting result; The information of the nonzero coefficient of a described adjacent block numerical value is meant the average of a nonzero coefficient numerical value of two adjacent blocks of current block.
Described step 2) in, if the adjacent block of described current block does not exist, then the nonzero coefficient of an adjacent block numerical value equals 0; If only have an adjacent block, then the average of the nonzero coefficient of the adjacent block of a current block numerical value is exactly a nonzero coefficient numerical value of existing adjacent block.
Described step 2) in, described context model is relevant with the number of nonzero coefficient, comprise elementary context model and secondary context model, described elementary context model is relevant with DCT coefficient place macro block (mb) type, and described secondary context model is relevant with the average of the nonzero coefficient number of adjacent block.
In the described step 5), it is relevant that described context model and coding nonzero coefficient absolute value subtract 1 result, comprise one-level context model, secondary context model and three grades of context models, wherein, the one-level context model is relevant with the type of DCT coefficient place macro block, the secondary context model is relevant with the number of nonzero coefficient in the macro block, and the three grades of nonzero coefficient that context model is to be encoded positions in scanning are relevant.
A kind of based on contextual adaptive entropy coding/decoding method, its step is as follows:
A), according to the context model of an adjacent block nonzero coefficient numerical information, decoding obtains the difference of a current block nonzero coefficient numerical value and an adjacent block nonzero coefficient numerical value, according to the information of the numerical value of nonzero coefficient in the adjacent block, obtain the individual numerical value of nonzero coefficient in the current block then;
B), by scanning sequency, select the context model decoding to obtain the value of nonzero coefficient flag bit on each scanning position according to current coefficient place scanning position successively, if decode value is 0, then this locational coefficient value is 0; As if decode value is 1, then this locational coefficient value non-0; Equal the individual numerical value of nonzero coefficient in this piece that step a) obtains up to the nonzero coefficient numerical value of decoding;
C), successively according to the number of nonzero coefficient in the type of macro block, the macro block, the choice of location context model decoding of nonzero coefficient in the scanning that obtains by step b) to be decoded, the absolute value that obtains each nonzero coefficient subtracts 1, and decoding obtains the sign bit of nonzero coefficient then; Equal the nonzero coefficient number that step a) obtains up to the individual numerical value that decodes;
D), the zero coefficient values that obtains according to decoding and the value and the positional information recovery coefficient piece of zero coefficient.
In the described step c), decoding nonzero coefficient absolute value subtracts 1 order and coding nonzero coefficient absolute value, and to subtract 1 order identical.
The invention has the advantages that:
Of the present invention based on contextual entropy coding method and entropy decoding method, make full use of contextual information, reduce redundancy, further improved the compression performance of system.
Description of drawings
Fig. 1 is the schematic diagram of 8 * 8 DCT coefficient block and zigzag scanning sequency;
Fig. 2 is the schematic diagram of current block and adjacent block;
Fig. 3 is the flow chart based on contextual adaptive entropy coding method of the present invention;
Fig. 4 is the flow chart based on contextual adaptive entropy coding/decoding method of the present invention.
Embodiment
Below in conjunction with the drawings and specific embodiments, method of the present invention is described further.
By predictive coding in the video coding and transition coding stage, original image is converted to the DCT coefficient block, and method of the present invention is to do further operation to the DCT coefficient block.
As shown in Figure 1, supposing has one 8 * 8 DCT coefficient block, and this piece is the luminance block of two-way inter coded images, represents with B_Luma, and 64 coefficients are arranged in this coefficient block, comprises zero coefficient and nonzero coefficient.As shown in Figure 3, describe based on the cataloged procedure of contextual adaptive entropy coding method below in conjunction with of the present invention this DCT coefficient block.
9 ,-2,3,0,1,0,0 ,-2,0,0,1 to the DCT coefficient block of 8 * 8 among Fig. 1, according to the zigzag scanning sequency, the one-dimensional sequence that the DCT coefficient of two dimension is arranged in is as follows:.Because the 11st position parameter value is 1 to be last nonzero coefficient, therefore, the zero coefficient behind this coefficient omits.In these 11 coefficients, nonzero coefficient has 6, and therefore, the TC value of this DCT coefficient block is 6.
Suppose that current block represents that with C as shown in Figure 2, current block has two adjacent blocks, these two adjacent blocks are represented with A and B respectively.If the TC of current block is designated as TC (C), then the TC of adjacent block remembers respectively and makes TC (A) and TC (B).When the difference that will calculate between current block TC and adjacent block TC, be the average that the TC value of current block will be deducted the TC value of two adjacent blocks.Therefore, current block C is adjacent the difference of the TC of piece and represents with the variable Δ, Δ=TC (C)-(TC (A)+TC (B))/2.When (TC (A)+TC (B))/2 are not integer, should round downwards.Difference DELTA can be got negative, positive number and 0, but can nonzero integer be made coding and consider compression performance owing to arithmetic encoder, therefore before being input to arithmetic encoder, earlier difference is done a mapping, all be mapped to nonnegative integer 6, the mapping method between difference DELTA and nonnegative integer δ is as shown in table 1:
| Difference DELTA | 0 | Δ＞0 | Δ＜0 |
| Value δ after the mapping | 0 | 2×Δ-1 | 2×(-Δ)+1 |
Table 1
With the average of the nonzero coefficient number of nonnegative integer variables A ver record adjacent block, that is: Aver=(TC (A)+TC (B))/2.Before structure context (Context) model relevant, at first to obtain elementary (Primary) context sequence number according to the type of DCT coefficient place macro block with the TC value.Corresponding relation between the type of described macro block and elementary context sequence number is as shown in table 2.
| Macro block (mb) type | In-frame encoding picture brightness (I_Luma) | Forward direction inter coded images brightness (P_Luma) | Two-way inter coded images brightness (B_Luma) | Colourity (Chroma) |
| Elementary context sequence number | 0 | 1 | 2 | 3 |
Table 2
Under the situation of same type,, select different context models according to the difference of the value of Aver.The value of Aver under each elementary context model is quantized into some levels (except that tint volume changes into 3 grades, all the other all are 4 grades), obtains secondary context sequence number:
| Elementary context sequence number 0 | Aver | [0，4] | [5，12] | [13，18] | [19，64] |
| Secondary context model number | 0 | 1 | 2 | 3 | |
| Elementary context sequence number 1 | Aver | [0，1] | [2，5] | [6，10] | [11，64] |
| Secondary context model number | 0 | 1 | 2 | 3 | |
| Elementary context sequence number 2 | Aver | 0 | [1，4] | [5，9] | [10，64] |
| Secondary context model number | 0 | 1 | 2 | 3 | |
| Elementary context sequence number 3 | Aver | 0 | [1，3] | [4，64] | |
| Secondary adopted pattern number up and down | 0 | 1 | 2 |
Table 3
In the present embodiment, cited DCT coefficient block is not indicated its adjacent block, the TC value of supposing its adjacent block is respectively TC (A)=3 and TC (B)=7, then the difference of current block TC and adjacent block TC is Δ=TC (C)-(TC (A)+TC (B))/2=6-(3+7/2)=1, Aver=(TC (A)+TC (B))/2=5, value after the mapping is δ=2 * 1-1=1, and macro block (mb) type is B_Luma.Therefore, elementary context sequence number is 2, and Secondary pattern number=2 utilize this contextual probability to drive arithmetic encoder coding δ.
To these 11 coefficients of 9 ,-2,3,0,1,0,0 ,-2,0,0,1 in the DCT coefficient block shown in Figure 1, record nonzero coefficient flag bit significant_coeff_flag, the result who obtains is 1,1,1,0,1,0,0,1,0,0,1.
Order according to 8 * 8 block scans is quantized into several contexts, an a kind of context model relevant of context index correspondence with the nonzero coefficient flag bit, and a kind of corresponding relation between the order of scanning is as shown in table 4, in the specific implementation, according to actual conditions, different corresponding relations can be arranged:
| 0 | 1 | 5 | 5 | 5 | 5 | 5 | 13 |
| 2 | 4 | 5 | 5 | 5 | 5 | 9 | 13 |
| 3 | 4 | 4 | 4 | 7 | 8 | 9 | 13 |
| 3 | 4 | 4 | 7 | 7 | 8 | 9 | 13 |
| 3 | 4 | 7 | 7 | 7 | 8 | 9 | 13 |
| 3 | 6 | 6 | 6 | 6 | 6 | 6 | 13 |
| 3 | 11 | 11 | 11 | 11 | 11 | 11 | 11 |
| 12 | 12 | 12 | 12 | 12 | 12 | 12 | 12 |
Table 4
To 11 coefficients in the example shown in Figure 1, the context index of each position is number as follows on its scanning position:
| The scanning sequence number | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
| Coefficient value | 9 | -2 | 3 | 0 | 1 | 0 | 0 | -2 | 0 | 0 | 1 |
| significant_coeff_flag | 1 | 1 | 1 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 1 |
| Context index number | 0 | 1 | 2 | 3 | 4 | 5 | 5 | 5 | 4 | 3 | 3 |
Table 5
The coding of piece prediction residual coefficient is actually its absolute value is subtracted 1 result encodes: be defined as follows context model respectively according to nonzero coefficient number and the position of nonzero coefficient in scanning to be encoded in the type of macro block, the macro block:
Type quantification according to macro block becomes 3 kinds of one-level context models, and is as shown in table 6, for a kind of one-level division methods of model up and down, at different macro block (mb) types, gives different one-level context model number.
| Macro block (mb) type | In-frame encoding picture brightness (I_Luma) | Forward direction, two-way inter coded images brightness (P_Luma, B_Luma) | Colourity (Chroma) |
| One-level context model number | 0 | 1 | 2 |
Table 6
According to nonzero coefficient number definition secondary context model in the macro block.When definition, according between the represented nonzero coefficient number dividing regions of variable TC,, each one-level context model is refined into several secondary context models according to the difference in interval;
As shown in table 7, be a kind of division methods of secondary context model.
| One-level context model number 0 | TC | [0，10] | [11，22] | [23，64] |
| Secondary context model number | 0 | 1 | 2 | |
| One-level context model number 1 | TC | [0，10] | [11，18] | [19，64] |
| Secondary context model number | 0 | 1 | 2 | |
| One-level context model numbers 2 | TC | [0，3] | [4，10] | [11，64] |
| Secondary context model number | 0 | 1 | 2 |
Table 7
For selected type model and coefficient number model, also to further determine three grades of context models of each nonzero coefficient number, three grades of context models are definite according to nonzero coefficient to be encoded position in scanning.Concrete, it is that interval according to the sequence number of this coefficient zigzag sequential scanning (pos represents with variable, and value is 0 ~ 63) place determines.
| One-level context model number 0 | Secondary context model number 0 | pos | 0 | [1，14] | [15，63] |
| Three grades of context models number | 0 | 1 | 2 | ||
| Secondary context model number 1 | pos | [0，4] | [5，24] | [25，63] | |
| Three grades of context models number | 0 | 1 | 2 | ||
| Secondary context model numbers 2 | pos | [0，13] | [14，46] | [47，63] | |
| Three grades of context models number | 0 | 1 | 2 | ||
| One-level context model number 1 | Secondary context model number 0 | pos | 0 | [1，9] | [10，63] |
| Three grades of context models number | 0 | 1 | 2 | ||
| Secondary context model number 1 | pos | [0，4] | [5，20] | [21，63] | |
| Three grades of context models number | 0 | 1 | 2 | ||
| Secondary context model numbers 2 | pos | [0，13] | [14，25] | [26，63] | |
| Three grades of context models number | 0 | 1 | 2 | ||
| One-level context model numbers 2 | Secondary context model number 0 | pos | 0 | [1，6] | [7，63] |
| Three grades of context models number | 0 | 1 | 2 | ||
| Secondary context model number 1 | pos | [0，2] | [3，12] | [13，63] | |
| Three grades of context models number | 0 | 1 | 2 |
| Secondary context model numbers 2 | pos | [0，10] | [11，20] | [21，63] | |
| Three grades of context models number | 0 | 1 | 2 |
Table 8
To 6 nonzero coefficients in the example shown in Figure 1, its corresponding call number is pressed shown in the table 9:
| Zero coefficient values | 9 | -2 | 3 | 1 | -2 | 1 |
| Position in the sequential scanning (pos) | 0 | 1 | 2 | 3 | 6 | 9 |
| One-level context index number | 1 | 1 | 1 | 1 | 1 | 1 |
| Secondary context index number | 0 | 0 | 0 | 0 | 0 | 0 |
| Three grades of context indexs number | 0 | 1 | 1 | 1 | 1 | 1 |
Table 9
Especially, used arithmetic encoder is the two-value arithmetic coding device if encode, then value is input to arithmetic encoder coding before, the nonnegative integer δ that obtain step 12, and the absolute value of the nonzero coefficient that obtains of step 15 subtracts 1 value, does binary conversion treatment.
Described binary conversion treatment is element to be encoded will be converted into the binary number of representing with 0 or 1.The corresponding relation of decimal number and binary number can be referring to table 10, as shown in Table 10, when metric nonzero integer is converted to string of binary characters, uses big or small identical 0 and one 1 with nonnegative integer to represent.For example, when the value of the nonzero coefficient among Fig. 1 was made coding, the absolute value with nonzero coefficient subtracted 1 earlier, obtains 8,1,2,0,1,0, makes binaryzation then, obtains 000000001,01,001,1,01,1.
| Nonnegative integer | String of binary characters |
| 0 1 2 3 4 5 ... 13 14 ... | 1 0 1 0 0 1 0 0 0 1 0 0 0 0 1 0 0 0 0 0 1 - - - - - - 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 - - - - - - - - - - - - - - - - |
| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15... |
Table 10
In video coding system, transmitting terminal utilize above-mentioned realize coding based on contextual adaptive entropy coding method after, also need to realize decoding at receiving terminal.
It is a kind of based on contextual adaptive entropy coding/decoding method that the present invention also provides, and as shown in Figure 4, its step is as follows:
In this step, decoding nonzero coefficient absolute value subtracts 1 order and coding nonzero coefficient absolute value to subtract 1 order identical.
Ins and outs and entropy coding method about entropy decoding method are similar, no longer repeat in the present embodiment.
Claims (10)
1, a kind of based on contextual adaptive entropy coding method, comprising:
1), according to scanning sequency scanning DCT coefficient block, obtain all coefficients before last nonzero coefficient and this coefficient in the current DCT coefficient block, and the individual numerical value of nonzero coefficient;
2), obtain the individual numerical value of the nonzero coefficient of current DCT coefficient block by step 1), ask the difference of a current block nonzero coefficient numerical value and an adjacent block nonzero coefficient numerical value, and the amount of information of the nonzero coefficient numerical value of adjacent block changed into some intervals, for each interval structure and the relevant context model of a nonzero coefficient numerical value, so that the difference of a nonzero coefficient numerical value is encoded;
3), last nonzero coefficient that step 1) is obtained and each coefficient before this coefficient are asked for the value of nonzero coefficient flag bit, if above-mentioned coefficient nonzero coefficient, then establishing the flag bit corresponding with coefficient is 1, if above-mentioned coefficient zero coefficient, then establishing flag bit is 0;
4), according to the scanning sequency of each coefficient in scanning process, select context model for the pairing nonzero coefficient flag bit of each coefficient and encode, the position in the scanning sequency at described context model and this coefficient place is relevant;
5), the absolute value of nonzero coefficient is subtracted 1 by the backward of scanning sequency or scanning, and utilizing in macro block (mb) type, the piece nonzero coefficient number and the choice of location context model in scanning to encode for absolute value subtracts 1 result, the nonzero coefficient absolute value of having encoded subtracts its sign bit of encoding again behind 1 the result.
2, according to claim 1ly it is characterized in that, before the absolute value to the difference of a described nonzero coefficient numerical value, nonzero coefficient subtracts 1 value coding, they are made binary conversion treatment, be converted into binary number based on contextual adaptive entropy coding method.
3, according to claim 1ly it is characterized in that based on contextual adaptive entropy coding method the scanning of the described DCT coefficient in the described step 1) is a process that two dimension or multidimensional DCT coefficient is arranged as one-dimensional sequence.
4, according to claim 1ly it is characterized in that based on contextual adaptive entropy coding method described scanning sequency is zigzag scanning sequency or vertical only scanning sequency or horizontal only scanning sequency.
5, according to claim 1 based on contextual adaptive entropy coding method, it is characterized in that, described step 2) in, the difference of a described current block nonzero coefficient numerical value and an adjacent block nonzero coefficient numerical value is: the individual numerical value of current block nonzero coefficient deducts the resulting result of a nonzero coefficient numerical value of adjacent block; The nonzero coefficient of a described adjacent block numerical value is meant the average of a nonzero coefficient numerical value of two adjacent blocks of current block.
6, according to claim 5ly it is characterized in that described step 2 based on contextual adaptive entropy coding method) in, if the adjacent block of described current block does not exist, then the nonzero coefficient of an adjacent block numerical value equals 0; If only have an adjacent block, then the average of the nonzero coefficient of the adjacent block of a current block numerical value is exactly a nonzero coefficient numerical value of existing adjacent block.
7, according to claim 1 based on contextual adaptive entropy coding method, it is characterized in that, described step 2) in, described context model is relevant with the number of nonzero coefficient, comprise elementary context model and secondary context model, described elementary context model is relevant with DCT coefficient place macro block (mb) type, and described secondary context model is relevant with the average of the nonzero coefficient number of adjacent block.
8, according to claim 1 based on contextual adaptive entropy coding method, it is characterized in that, in the described step 5), it is relevant that described context model and coding nonzero coefficient absolute value subtract 1 result, comprise one-level context model, secondary context model and three grades of context models, wherein, the one-level context model is relevant with the type of DCT coefficient place macro block, the secondary context model is relevant with the number of nonzero coefficient in the macro block, and the three grades of nonzero coefficient that context model is to be encoded positions in scanning are relevant.
9, a kind of based on contextual adaptive entropy coding/decoding method, its step is as follows:
A), according to the context model of an adjacent block nonzero coefficient numerical information, decoding obtains the difference of a current block nonzero coefficient numerical value and an adjacent block nonzero coefficient numerical value, according to the information of the numerical value of nonzero coefficient in the adjacent block, obtain the individual numerical value of nonzero coefficient in the current block then;
B), by scanning sequency, select the context model decoding to obtain the value of nonzero coefficient flag bit on each scanning position according to current coefficient place scanning position successively, if decode value is 0, then this locational coefficient value is 0; As if decode value is 1, then this locational coefficient value non-0; Equal the individual numerical value of nonzero coefficient in this piece that step a) obtains up to the nonzero coefficient numerical value of decoding;
C), successively according to the number of nonzero coefficient in the type of macro block, the macro block, the choice of location context model decoding of nonzero coefficient in the scanning that obtains by step b) to be decoded, obtain each nonzero coefficient absolute value and subtract 1 result, decoding obtains the sign bit of nonzero coefficient then; Equal the nonzero coefficient number that step a) obtains up to the individual numerical value that decodes;
D), the zero coefficient values that obtains according to decoding and the value and the positional information recovery coefficient piece of zero coefficient.
10, according to claim 9ly it is characterized in that in the described step c), decoding nonzero coefficient absolute value subtracts 1 order and coding nonzero coefficient absolute value, and to subtract 1 order identical based on contextual adaptive entropy coding/decoding method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200510126222 CN100488254C (en) | 2005-11-30 | 2005-11-30 | Entropy coding method and decoding method based on text |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200510126222 CN100488254C (en) | 2005-11-30 | 2005-11-30 | Entropy coding method and decoding method based on text |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1980395A true CN1980395A (en) | 2007-06-13 |
| CN100488254C CN100488254C (en) | 2009-05-13 |
Family
ID=38131361
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200510126222 Expired - Fee Related CN100488254C (en) | 2005-11-30 | 2005-11-30 | Entropy coding method and decoding method based on text |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100488254C (en) |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102186087A (en) * | 2011-06-24 | 2011-09-14 | 哈尔滨工业大学 | Parallel non-zero coefficient context modeling method for binary arithmetic coding |
| CN102377994A (en) * | 2010-08-05 | 2012-03-14 | 富士通株式会社 | Context-adaptive variable-length coding method and system thereof |
| CN102685503A (en) * | 2011-03-10 | 2012-09-19 | 华为技术有限公司 | Encoding method of conversion coefficients, decoding method of conversion coefficients and device |
| WO2012163199A1 (en) * | 2011-05-27 | 2012-12-06 | Mediatek Inc. | Method and apparatus for line buffer reduction for video processing |
| CN102177543B (en) * | 2008-10-08 | 2013-05-15 | 弗朗霍夫应用科学研究促进协会 | Audio decoder, audio encoder, method for decoding an audio signal, method for encoding an audio signal |
| CN103250413A (en) * | 2010-11-16 | 2013-08-14 | 高通股份有限公司 | Parallel context calculation in video coding |
| CN104093020A (en) * | 2011-03-10 | 2014-10-08 | 华为技术有限公司 | Coding method and device of transformation coefficients and decoding method and device of transformation coefficients |
| CN104093018A (en) * | 2011-03-10 | 2014-10-08 | 华为技术有限公司 | Coding method and device of transformation coefficients and decoding method and device of transformation coefficients |
| CN106162201A (en) * | 2015-03-26 | 2016-11-23 | 北京君正集成电路股份有限公司 | A kind of image data compression method based on sign bit packet and device |
| CN106488238A (en) * | 2010-04-05 | 2017-03-08 | 三星电子株式会社 | Low complex degree entropy coding/decoding method and apparatus |
| CN107529060A (en) * | 2011-10-31 | 2017-12-29 | 三星电子株式会社 | Determine the method and apparatus and computer recordable media of context model |
| CN107770549A (en) * | 2012-01-19 | 2018-03-06 | 佳能株式会社 | The method for coding and decoding the validity mapping of the residual error coefficient of change of scale |
| CN103238322B (en) * | 2010-12-03 | 2018-05-04 | 维洛媒体国际有限公司 | The individually position of the last coefficient of efficiency of code video blocks in video coding |
| CN108881910A (en) * | 2010-04-13 | 2018-11-23 | Ge视频压缩有限责任公司 | Decode the device of Saliency maps |
| CN110708552A (en) * | 2019-08-27 | 2020-01-17 | 杭州海康威视数字技术股份有限公司 | Decoding method, encoding method and device |
| WO2020135371A1 (en) * | 2018-12-24 | 2020-07-02 | 华为技术有限公司 | Flag bit context modeling method and device |
| WO2020172907A1 (en) * | 2019-02-27 | 2020-09-03 | 北京大学深圳研究生院 | Method and device for selecting context model of quantization coefficient end flag bit |
| CN111988630A (en) * | 2020-09-11 | 2020-11-24 | 北京锐马视讯科技有限公司 | Video transmission method and device, equipment and storage medium |
| CN113365058A (en) * | 2019-03-09 | 2021-09-07 | 杭州海康威视数字技术股份有限公司 | Method, decoding end, encoding end and system for encoding and decoding |
| CN113810699A (en) * | 2016-04-29 | 2021-12-17 | 世宗大学校产学协力团 | Method and apparatus for encoding and decoding image signal |
| CN114071140A (en) * | 2019-12-03 | 2022-02-18 | 杭州海康威视数字技术股份有限公司 | Decoding method, encoding method, device, equipment and machine readable storage medium |
| CN114175653A (en) * | 2019-09-17 | 2022-03-11 | 北京达佳互联信息技术有限公司 | Method and apparatus for lossless codec mode in video codec |
| US11330272B2 (en) | 2010-12-22 | 2022-05-10 | Qualcomm Incorporated | Using a most probable scanning order to efficiently code scanning order information for a video block in video coding |
| CN114567786A (en) * | 2019-09-23 | 2022-05-31 | 北京达佳互联信息技术有限公司 | Method and apparatus for video encoding and decoding in 4:4:4chroma format |
| CN115379241A (en) * | 2017-09-18 | 2022-11-22 | 谷歌有限责任公司 | Method and apparatus for coding last significant coefficient flag |
| CN115604473A (en) * | 2017-10-23 | 2023-01-13 | 谷歌有限责任公司（Us） | Method and apparatus for coding blocks of video data |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2492333B (en) * | 2011-06-27 | 2018-12-12 | British Broadcasting Corp | Video encoding and decoding using transforms |
| KR20140097969A (en) * | 2011-11-07 | 2014-08-07 | 파나소닉 인텔렉츄얼 프로퍼티 코포레이션 오브 아메리카 | Image encoding method, image decoding method, image encoding device, image decoding device, and image encoding/decoding device |
-
2005
- 2005-11-30 CN CN 200510126222 patent/CN100488254C/en not_active Expired - Fee Related
Cited By (57)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102177543B (en) * | 2008-10-08 | 2013-05-15 | 弗朗霍夫应用科学研究促进协会 | Audio decoder, audio encoder, method for decoding an audio signal, method for encoding an audio signal |
| CN106488238B (en) * | 2010-04-05 | 2019-06-28 | 三星电子株式会社 | Low complex degree entropy coding/decoding method and apparatus |
| CN106488238A (en) * | 2010-04-05 | 2017-03-08 | 三星电子株式会社 | Low complex degree entropy coding/decoding method and apparatus |
| CN113556554A (en) * | 2010-04-13 | 2021-10-26 | Ge视频压缩有限责任公司 | Coding of significance maps and transform coefficient blocks |
| CN113556558A (en) * | 2010-04-13 | 2021-10-26 | Ge视频压缩有限责任公司 | Coding of significance maps and transform coefficient blocks |
| CN113556561A (en) * | 2010-04-13 | 2021-10-26 | Ge视频压缩有限责任公司 | Coding of significance maps and transform coefficient blocks |
| CN108881910B (en) * | 2010-04-13 | 2021-01-26 | Ge视频压缩有限责任公司 | Apparatus for decoding significance map |
| CN108881910A (en) * | 2010-04-13 | 2018-11-23 | Ge视频压缩有限责任公司 | Decode the device of Saliency maps |
| CN102377994B (en) * | 2010-08-05 | 2014-05-07 | 富士通株式会社 | Context-adaptive variable-length coding method and system thereof |
| CN102377994A (en) * | 2010-08-05 | 2012-03-14 | 富士通株式会社 | Context-adaptive variable-length coding method and system thereof |
| CN103250413A (en) * | 2010-11-16 | 2013-08-14 | 高通股份有限公司 | Parallel context calculation in video coding |
| CN103238322B (en) * | 2010-12-03 | 2018-05-04 | 维洛媒体国际有限公司 | The individually position of the last coefficient of efficiency of code video blocks in video coding |
| US11330272B2 (en) | 2010-12-22 | 2022-05-10 | Qualcomm Incorporated | Using a most probable scanning order to efficiently code scanning order information for a video block in video coding |
| US9571836B2 (en) | 2011-03-10 | 2017-02-14 | Huawei Technologies Co., Ltd. | Method and apparatus for encoding and decoding with multiple transform coefficients sub-blocks |
| CN104093018A (en) * | 2011-03-10 | 2014-10-08 | 华为技术有限公司 | Coding method and device of transformation coefficients and decoding method and device of transformation coefficients |
| CN102685503A (en) * | 2011-03-10 | 2012-09-19 | 华为技术有限公司 | Encoding method of conversion coefficients, decoding method of conversion coefficients and device |
| CN104093018B (en) * | 2011-03-10 | 2017-08-04 | 华为技术有限公司 | The coding method of conversion coefficient, the coding/decoding method of conversion coefficient, and device |
| CN102685503B (en) * | 2011-03-10 | 2014-06-25 | 华为技术有限公司 | Encoding method of conversion coefficients, decoding method of conversion coefficients and device |
| CN104093020A (en) * | 2011-03-10 | 2014-10-08 | 华为技术有限公司 | Coding method and device of transformation coefficients and decoding method and device of transformation coefficients |
| US10165305B2 (en) | 2011-03-10 | 2018-12-25 | Huawei Technologies Co., Ltd. | Encoding and decoding transform coefficient sub-blocks in same predetermine order |
| WO2012163199A1 (en) * | 2011-05-27 | 2012-12-06 | Mediatek Inc. | Method and apparatus for line buffer reduction for video processing |
| CN103563380B (en) * | 2011-05-27 | 2017-07-21 | 寰发股份有限公司 | Reduce the method and device of the row buffering for Video processing |
| US9986247B2 (en) | 2011-05-27 | 2018-05-29 | Hfi Innovation Inc. | Method and apparatus for line buffer reduction for video processing |
| US9866848B2 (en) | 2011-05-27 | 2018-01-09 | Hfi Innovation Inc. | Method and apparatus for line buffer reduction for video processing |
| CN107277539A (en) * | 2011-05-27 | 2017-10-20 | 寰发股份有限公司 | Reduce the method and device for the row caching in context-adaptive entropy decoding |
| US20130329784A1 (en) * | 2011-05-27 | 2013-12-12 | Mediatek Inc. | Method and Apparatus for Line Buffer Reduction for Video Processing |
| CN103563380A (en) * | 2011-05-27 | 2014-02-05 | 联发科技股份有限公司 | Method and apparatus for line buffer reduction for video processing |
| US9762918B2 (en) * | 2011-05-27 | 2017-09-12 | Hfi Innovation Inc. | Method and apparatus for line buffer reduction for video processing |
| CN102186087A (en) * | 2011-06-24 | 2011-09-14 | 哈尔滨工业大学 | Parallel non-zero coefficient context modeling method for binary arithmetic coding |
| CN107529060B (en) * | 2011-10-31 | 2020-04-21 | 三星电子株式会社 | Method and apparatus for determining context model and computer recordable medium |
| CN107529060A (en) * | 2011-10-31 | 2017-12-29 | 三星电子株式会社 | Determine the method and apparatus and computer recordable media of context model |
| US10531100B2 (en) | 2012-01-19 | 2020-01-07 | Canon Kabushiki Kaisha | Method, apparatus and system for encoding and decoding the significance map for residual coefficients of a transform unit |
| US10531101B2 (en) | 2012-01-19 | 2020-01-07 | Canon Kabushiki Kaisha | Method, apparatus and system for encoding and decoding the significance map for residual coefficients of a transform unit |
| CN107770549B (en) * | 2012-01-19 | 2020-02-14 | 佳能株式会社 | Method and apparatus for encoding and decoding image frame data, and computer readable medium |
| CN107770549A (en) * | 2012-01-19 | 2018-03-06 | 佳能株式会社 | The method for coding and decoding the validity mapping of the residual error coefficient of change of scale |
| US10516887B2 (en) | 2012-01-19 | 2019-12-24 | Canon Kabushiki Kaisha | Method, apparatus and system for encoding and decoding the significance map for residual coefficients of a transform unit |
| CN106162201A (en) * | 2015-03-26 | 2016-11-23 | 北京君正集成电路股份有限公司 | A kind of image data compression method based on sign bit packet and device |
| CN106162201B (en) * | 2015-03-26 | 2019-02-15 | 北京君正集成电路股份有限公司 | A kind of image data compression method and device based on sign bit grouping |
| CN113810699A (en) * | 2016-04-29 | 2021-12-17 | 世宗大学校产学协力团 | Method and apparatus for encoding and decoding image signal |
| CN115379241A (en) * | 2017-09-18 | 2022-11-22 | 谷歌有限责任公司 | Method and apparatus for coding last significant coefficient flag |
| CN115379241B (en) * | 2017-09-18 | 2023-06-06 | 谷歌有限责任公司 | Method and apparatus for coding a last significant coefficient flag |
| CN115604473A (en) * | 2017-10-23 | 2023-01-13 | 谷歌有限责任公司（Us） | Method and apparatus for coding blocks of video data |
| WO2020135371A1 (en) * | 2018-12-24 | 2020-07-02 | 华为技术有限公司 | Flag bit context modeling method and device |
| CN113316939A (en) * | 2018-12-24 | 2021-08-27 | 华为技术有限公司 | Context modeling method and device for zone bit |
| US11949868B2 (en) | 2019-02-27 | 2024-04-02 | Peking University Shenzhen Graduate School | Method and device for selecting context model of quantization coefficient end flag bit |
| WO2020172907A1 (en) * | 2019-02-27 | 2020-09-03 | 北京大学深圳研究生院 | Method and device for selecting context model of quantization coefficient end flag bit |
| CN113365058B (en) * | 2019-03-09 | 2023-02-28 | 杭州海康威视数字技术股份有限公司 | Method, decoding end, encoding end and system for encoding and decoding |
| CN113365058A (en) * | 2019-03-09 | 2021-09-07 | 杭州海康威视数字技术股份有限公司 | Method, decoding end, encoding end and system for encoding and decoding |
| CN110708552A (en) * | 2019-08-27 | 2020-01-17 | 杭州海康威视数字技术股份有限公司 | Decoding method, encoding method and device |
| WO2021036429A1 (en) * | 2019-08-27 | 2021-03-04 | 杭州海康威视数字技术股份有限公司 | Decoding method, encoding method, and apparatus |
| CN112449192A (en) * | 2019-08-27 | 2021-03-05 | 杭州海康威视数字技术股份有限公司 | Decoding method, encoding method and device |
| CN114175653A (en) * | 2019-09-17 | 2022-03-11 | 北京达佳互联信息技术有限公司 | Method and apparatus for lossless codec mode in video codec |
| CN114175653B (en) * | 2019-09-17 | 2023-07-25 | 北京达佳互联信息技术有限公司 | Method and apparatus for lossless codec mode in video codec |
| CN114567786A (en) * | 2019-09-23 | 2022-05-31 | 北京达佳互联信息技术有限公司 | Method and apparatus for video encoding and decoding in 4:4:4chroma format |
| CN114567786B (en) * | 2019-09-23 | 2023-08-29 | 北京达佳互联信息技术有限公司 | Method and apparatus for video encoding and decoding in 4:4:4 chroma format |
| CN114071140A (en) * | 2019-12-03 | 2022-02-18 | 杭州海康威视数字技术股份有限公司 | Decoding method, encoding method, device, equipment and machine readable storage medium |
| CN111988630A (en) * | 2020-09-11 | 2020-11-24 | 北京锐马视讯科技有限公司 | Video transmission method and device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100488254C (en) | 2009-05-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100488254C (en) | Entropy coding method and decoding method based on text | |
| CN1332563C (en) | Coding method of video frequency image jump over macro block | |
| KR100701810B1 (en) | Signal encoding method, signal decoding method, signal encoding device, signal decoding device, computer-readable recording medium | |
| US8401321B2 (en) | Method and apparatus for context adaptive binary arithmetic coding and decoding | |
| CN101800898B (en) | Coding device and method | |
| KR101622450B1 (en) | Video encoding and decoding using transforms | |
| CN103238322B (en) | The individually position of the last coefficient of efficiency of code video blocks in video coding | |
| CN101098473B (en) | Picture coding method and apparatus | |
| KR101523452B1 (en) | Coding the position of a last significant coefficient within a video block based on a scanning order for the block in video coding | |
| CN1327395C (en) | Method and system for context-based adaptive binary arithmetic coding | |
| CN100492916C (en) | Variable-length code using multiple mapping table and decoding method and apparatus | |
| US20160277765A1 (en) | Method and arrangement for coding transform coefficients in picture and/or video coders and decoders and a corresponding computer program and a corresponding computer-readable storage medium | |
| CN100499810C (en) | Image encoding method, image decoding method, image encoder, image decode, and image transmission system | |
| CN101584218B (en) | Method and apparatus for encoding and decoding based on intra prediction | |
| CN104041040A (en) | Encoding of prediction residuals for lossless video coding | |
| CN103959775A (en) | Intra-mode video coding | |
| CN104205835B (en) | The context of most end position decoding is exported for video coding | |
| CN105027561A (en) | Context derivation for context-adaptive, multi-level significance coding | |
| CN103597838A (en) | Methods and devices for coding and decoding the position of the last significant coefficient | |
| KR101038531B1 (en) | Apparatus and method for encoding image capable of parallel processing in decoding and apparatus and method for decoding image capable of parallel processing | |
| CN1589023A (en) | Coding and decoding method and device for multiple coded list lengthening based on context | |
| CN103460701A (en) | Complexity reduction of significance map coding | |
| CN100454339C (en) | Method and system for context-based adaptive binary arithmetic coding | |
| CN112995671B (en) | Video encoding and decoding method and device, computer readable medium and electronic equipment | |
| CN101686398B (en) | Method for transcoding a first codeword to a second codeword |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090513 Termination date: 20211130 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |