CN102215390B

CN102215390B - Processing method and device for image coding and decoding

Info

Publication number: CN102215390B
Application number: CN201010147587.0A
Authority: CN
Inventors: 林永兵
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2010-04-09
Filing date: 2010-04-09
Publication date: 2014-07-09
Anticipated expiration: 2030-04-09
Also published as: CN102215390A

Abstract

The embodiment of the invention provides a processing method and device for image coding and decoding. The processing method for image coding comprises the following steps that: a macroblock to be coded is subjected to a down sampling division to obtain a plurality of blocks; the plurality of blocks are subjected to spatial-domain prediction intra-frame coding to obtain prediction residuals which respectively correspond to the plurality of blocks, wherein the spatial-domain prediction intra-frame coding at least comprises an interblock prediction intra-frame coding by adopting a prediction mode; and at least one prediction residual is subjected to directional change based on the prediction mode. In the embodiment of the invention, the prediction accuracy and the coding compression performance can be effectively improved.

Description

Image coding and decoding processing method and device

Technical field

The embodiment of the present invention relates to the communications field, relates in particular to a kind of image coding and decoding processing method and device.

Background technology

Video coding compression is the correlation between a kind of spatial domain, time domain and code word of utilizing video image, removes as far as possible a kind of Video compression method of redundancy.Inner frame coding method is a kind of block-based hybrid video coding method.

In the prior art, inner frame coding method can comprise: spatial domain prediction, conversion, quantification and entropy coding.Wherein, spatial domain prediction can utilize the correlation between adjacent block, carries out spatial domain prediction coding.For instance, H.264 the spatial domain prediction in standard can utilize the adjacent block of having encoded, and the piece of current required prediction is carried out to predictive coding.Due in standard H.264, luminance component has 9 kinds of predictive modes, and chromatic component has 5 kinds of predictive modes.In these predictive modes, except DC mode, in fact all the other predictive modes have characterized the texture information of image different directions.Carrying out spatial domain prediction when coding, can carry out predictive coding to current block according to the predictive mode of choosing, obtain prediction residual, then to prediction residual convert, quantification and entropy coding, finally generate compressed bit stream.Can remove the spatial domain redundancy of video image by these steps, thereby reach the object of compressed encoding.Correspondingly, decoding end also can be decoded according to the information such as predictive mode, prediction residual, thereby obtains decoded image.

Realizing in process of the present invention, inventor finds, the data volume of high clear video image is larger, takies network bandwidth resources more when transmission, and existing network bandwidth resources are difficult to meet the transmission of high clear video image.Therefore, how further to improve Video coding compression performance and become problem demanding prompt solution.

Summary of the invention

The embodiment of the present invention provides a kind of image coding and decoding processing method and device.

The embodiment of the present invention provides a kind of Image Coding processing method, comprising:

The macro block of required coding is carried out to down-sampling division, obtain a plurality of;

Carry out spatial domain prediction intraframe coding to described a plurality of, obtain and described a plurality of corresponding prediction residual respectively, wherein said spatial domain prediction intraframe coding at least comprises that employing predictive mode carries out inter prediction intraframe coding;

At least one prediction residual in described prediction residual is carried out to the directional transform based on described predictive mode.

The embodiment of the present invention provides a kind of image decoding processing method, comprising:

To carrying out the directivity inverse transformation based on predictive mode with a plurality of prediction residual coefficients that at least one piece is corresponding in corresponding prediction residual coefficient respectively, obtain prediction residual;

Described prediction residual is carried out to the processing of spatial domain prediction intraframe decoder, obtain described a plurality of, wherein said spatial domain prediction intraframe decoder is processed and is at least comprised that the described predictive mode of employing carries out inter prediction intraframe decoder;

Carry out up-sampling to described a plurality of and synthesize, obtain decoded macro block.

The embodiment of the present invention provides a kind of image code processing device, comprising:

Coding is divided module, for the macro block of required coding is carried out to down-sampling division, obtains a plurality of;

Coded prediction module, for carrying out spatial domain prediction intraframe coding to described a plurality of, obtains and described a plurality of corresponding prediction residual respectively, and wherein said spatial domain prediction intraframe coding at least comprises that employing predictive mode carries out inter prediction intraframe coding;

Transcoding, coding transform module, for carrying out the directional transform based on described predictive mode at least one prediction residual of described prediction residual.

The embodiment of the present invention provides a kind of image decoding processing unit, comprising:

Decoding conversion module, for to carrying out the directivity inverse transformation based on predictive mode with a plurality of prediction residual coefficients that corresponding at least one piece of prediction residual coefficient is corresponding respectively, obtains prediction residual;

Decoding prediction module, for described prediction residual being carried out to the processing of spatial domain prediction intraframe decoder, obtains described a plurality of, and wherein said spatial domain prediction intraframe decoder is processed and at least comprised that the described predictive mode of employing carries out inter prediction intraframe decoder;

Decoding synthesis module, synthesizes for carrying out up-sampling to described a plurality of, obtains decoded macro block.

The embodiment of the present invention, by the macro block of required coding is carried out to down-sampling division, can make full use of the correlation between pixel, thereby improves precision of prediction, improves compression coding performance; Carry out the directional transform based on pattern by least one prediction residual in the prediction residual of obtaining after spatial domain prediction intraframe coding is processed, can further remove the Anisotropic Texture Information correlation in prediction residual, improve compression coding performance.

Brief description of the drawings

In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, to the accompanying drawing of required use in embodiment or description of the Prior Art be briefly described below, apparently, accompanying drawing in the following describes is some embodiments of the present invention, for those of ordinary skill in the art, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.

Fig. 1 is the flow chart of Image Coding processing method embodiment mono-of the present invention;

Fig. 2 is the flow chart of Image Coding processing method embodiment bis-of the present invention;

Fig. 3 is the structural representation that in Image Coding processing method embodiment bis-of the present invention, down-sampling is divided;

Fig. 4 is the structural representation that Image Coding processing method embodiment bis-of the present invention carries out infra-frame prediction;

Fig. 5 is the flow chart of Image Coding processing method embodiment tri-of the present invention;

Fig. 6 is the flow chart of Image Coding processing method embodiment tetra-of the present invention;

Fig. 7 is the structural representation that Image Coding processing method embodiment tetra-of the present invention carries out infra-frame prediction;

Fig. 8 is the flow chart of Image Coding processing method embodiment five of the present invention;

The flow chart of Fig. 9 Image Coding processing method of the present invention embodiment six;

Figure 10 is the structural representation that Image Coding processing method embodiment six of the present invention carries out infra-frame prediction;

Figure 11 is the flow chart of Image Coding processing method embodiment seven of the present invention;

Figure 12 is the flow chart of decoding processing method embodiment mono-of the present invention;

Figure 13 is the flow chart of decoding processing method embodiment bis-of the present invention;

Figure 14 is the flow chart of decoding processing method embodiment tri-of the present invention;

Figure 15 is the flow chart of decoding processing method embodiment tetra-of the present invention;

Figure 16 is the flow chart of decoding processing method embodiment five of the present invention;

Figure 17 is the flow chart of decoding processing method embodiment six of the present invention;

Figure 18 is the flow chart of decoding processing method embodiment seven of the present invention;

Figure 19 is the structural representation of image code processing device embodiment mono-of the present invention;

Figure 20 is the structural representation of image code processing device embodiment bis-of the present invention;

Figure 21 is the structural representation of image code processing device embodiment tri-of the present invention;

Figure 22 is the structural representation of image code processing device embodiment tetra-of the present invention;

Figure 23 is the structural representation of image code processing device embodiment five of the present invention;

Figure 24 is the structural representation of image decoding processing unit embodiment mono-of the present invention;

Figure 25 is the structural representation of image decoding processing unit embodiment bis-of the present invention;

Figure 26 is the structural representation of image decoding processing unit embodiment tri-of the present invention;

Figure 27 is the structural representation of image decoding processing unit embodiment tetra-of the present invention;

Figure 28 is the structural representation of image decoding processing unit embodiment five of the present invention;

Figure 29 is the structural representation of image decoding processing unit embodiment six of the present invention.

Embodiment

For making object, technical scheme and the advantage of the embodiment of the present invention clearer, below in conjunction with the accompanying drawing in the embodiment of the present invention, technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is the present invention's part embodiment, instead of whole embodiment.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtaining under creative work prerequisite, belong to the scope of protection of the invention.

Fig. 1 is the flow chart of Image Coding processing method embodiment mono-of the present invention, and as shown in Figure 1, the method for the present embodiment can comprise:

Step 101, the macro block of required coding is carried out to down-sampling division, obtain a plurality of.

Specifically, image code processing device is at the macro block (Macroblock to required coding, hereinafter to be referred as: MB) encode process time, can carry out down-sampling division to the macro block of required coding, thus obtain a plurality of (hereinafter to be referred as: Block).For instance, the MB of required coding is 16 × 16, and image code processing device can carry out 1/2 down-sampling to this MB, also the row and column of 16 × 16 MB is carried out respectively to 1/2 extraction, thereby can obtain the Block of 48 × 8.In the present embodiment, adopt down-sampling division methods to divide the MB of required coding, can make full use of the correlation between pixel, thereby improve precision of prediction, improve compression coding performance.

It should be noted that, the size of MB described in the present embodiment can be both above-mentioned 16 × 16, also can be 32 × 32,64 × 64,128 × 128 or other arbitrary dimension, this down-sampling is divided and both can be processed luminance block, also can process chrominance block, and this down-sampling divides other down-sampling mode that also can adopt except 1/2 down-sampling, the present embodiment is not restricted.

Step 102, carry out spatial domain prediction intraframe coding to described a plurality of, obtain and described a plurality of corresponding prediction residual respectively, wherein said spatial domain prediction intraframe coding at least comprises that employing predictive mode carries out inter prediction intraframe coding.

After a plurality of Block of image code processing device in the Macroblock that obtains required coding, can carry out spatial domain prediction intraframe coding to these a plurality of Block, thereby can obtain the prediction residual corresponding with these a plurality of Block difference.For instance, image code processing device can all adopt inter prediction intraframe coding processing to the Block of 48 × 8, or can carry out inter prediction intraframe coding processing to the Block of 18 × 8 in the Block of 48 × 8, can carry out predicting between pixel intraframe coding processing with reference to the Block that adopts inter prediction intraframe coding processing to other Block of 38 × 8.Wherein, inter prediction intraframe coding processing is taking adjacent coded block as with reference to predicting, predicts that intraframe coding processing is taking adjacent encoded pixels as with reference to predicting between pixel.Therefore, in the present embodiment, image code processing device can be distinguished the Block of 48 × 8 corresponding spatial domain prediction intraframe coding result and corresponding the subtracting each other of original pixels of the Block of 48 × 8, thereby can obtain respectively the prediction residual corresponding with the Block difference of 48 × 8.

It should be noted that, spatial domain prediction intraframe coding in the present embodiment can adopt spatial domain prediction inner frame coding method realization arbitrarily in prior art, for example adopt traditional inter prediction intraframe coding processing, or between employing pixel, predict other spatial domain prediction inner frame coding methods such as intraframe coding processing, adopt predictive mode to carry out inter prediction intraframe coding as long as this spatial domain prediction intraframe coding at least comprises, the present embodiment is not restricted.

Step 103, at least one prediction residual in described prediction residual is carried out to the directional transform based on described predictive mode.

Because the intra prediction mode (Intra prediction mode) of spatial domain prediction intraframe coding can reflect the grain direction information of Block itself to a certain extent, and these grain direction information still exist in the prediction residual of a plurality of Block.Therefore, the present embodiment can adopt directional transform (the Mode-Dependent Directional Transform based on predictive mode at least one prediction residual in the prediction residual corresponding with a plurality of Block difference, hereinafter to be referred as: MDDT), for example adopt the predictive mode travel direction conversion adopting in step 102, thereby can further remove the Anisotropic Texture Information correlation in prediction residual, improve compression coding performance.

Specifically, MDDT can be a conversion of various Intra prediction mode customization, with the different Anisotropic Texture Information in better adaptive prediction residual error.For instance, at key technology area (KeyTechnology Area, hereinafter to be referred as: KTA) in reference software, MDDT for except direct current (hereinafter to be referred as: DC) other Intra prediction mode pattern customizes respectively a directional transform, as other 8 kinds of Intraprediction mode except DC pattern in 8 × 8 and 4 × 4MB in H.264, respectively corresponding 8 kinds of directional transform, still adopt the conversion irrelevant with Anisotropic Texture Information for DC pattern, for example discrete cosine transform (discrete cosine transform, hereinafter to be referred as: DCT), discrete sine transform, wavelet transformation etc., these conversion are irrelevant with the predictive mode that characterizes Anisotropic Texture Information, for example irrelevant with above-mentioned 8 kinds of Intra prediction mode.In the present embodiment, adopt MDDT, can further remove the Anisotropic Texture Information correlation in prediction residual, improve compression coding performance.

The present embodiment, by the MB of required coding is carried out to down-sampling division, can make full use of the correlation between pixel, thereby improves precision of prediction, improves compression coding performance; Carry out MDDT by least one prediction residual in the prediction residual of obtaining after spatial domain prediction intraframe coding is processed, can further remove the Anisotropic Texture Information correlation in prediction residual, improve compression coding performance.

Adopt several specific embodiments to be elaborated to the technical scheme of Image Coding processing method of the present invention below.

Fig. 2 is the flow chart of Image Coding processing method embodiment bis-of the present invention, and as shown in Figure 2, the method for the present embodiment can comprise:

Step 201, the macro block of required coding is carried out to down-sampling division, obtain a plurality of.

In the present embodiment, the step 101 shown in step 201 and Fig. 1 to realize principle similar.The MB that in the present embodiment, can suppose required coding is 16 × 16.Fig. 3 is the structural representation that in Image Coding processing method embodiment bis-of the present invention, down-sampling is divided, as shown in Figure 3, image code processing device can carry out 1/2 down-sampling division to 16 × 16 MB, in vertical and horizontal direction every 1 pixel of 1 pixel decimation, and the pixel after extracting is formed to new piece, can form altogether the Block of 48 × 8.In the present embodiment, this Block of 48 × 8 can be designated as respectively Block0, Block1, Block2 and Block3.

Step 202, adopt predictive mode to carry out inter prediction intraframe coding processing to first in described a plurality of, obtain and described first corresponding the first prediction residual, and the inter prediction intraframe coding result of applying described first carries out predicting between pixel intraframe coding processing to all the other pieces in described a plurality of, obtain and all the other pieces in described a plurality of corresponding the second prediction residual respectively.

In the present embodiment, image code processing device can adopt Intra prediction mode to carry out inter prediction intraframe coding processing to 8 × 8 Block3, obtain the inter prediction intraframe coding result of this Block3, and by the processing that subtracts of the original pixels of the Block3 of this inter prediction intraframe coding result and 8 × 8, obtain first prediction residual corresponding with Block3.Then, image code processing device can be with the inter prediction intraframe coding result of this Block3 as a reference, to all the other Block of 38 × 8, be that Block0, Block1, Blcok2 carry out predicting intraframe coding processing between pixel, distinguish the second corresponding prediction residual thereby obtain with Block0, Block1, Blcok2.Specifically, the present embodiment can carry out interpolation to the pixel of Block3, estimates to obtain prediction more accurately.For instance, the present embodiment can adopt a kind of interpolation method of interpolation filter of the horizontal and vertical direction based on 4-tap to carry out interpolation to the pixel of Block3, therefore, pixel in Block0, Block1, Blcok2 can predict with reference to adjacent pixel of having encoded Block3, and between this kind of pixel, prediction is more accurate than the precision of inter prediction.So far, image code processing device can obtain with Block0, Block1, Blcok2 and between corresponding pixel, predict intraframe coding result respectively, then image code processing device can will be predicted original pixels corresponding subtract each other of intraframe coding result with Block0, Block1, Blcok2 respectively between the pixel of Block0, Block1, Blcok2, distinguishes the second corresponding prediction residual thereby obtain with Block0, Block1, Blcok2.

Fig. 4 is the structural representation that Image Coding processing method embodiment bis-of the present invention carries out infra-frame prediction, as shown in Figure 4, wherein I3 represents Block3 to carry out inter prediction intraframe coding processing, obtains the first prediction residual, therefore, this first prediction residual has Intra prediction mode; And P0, P1, P2 represent respectively taking the inter prediction intraframe coding result of Block3 as predicting intraframe coding processing between pixel with reference to Block0, Block1, Blcok2 are carried out, obtain respectively the second prediction residual, therefore, the second prediction residual does not have Intra prediction mode.

It should be noted that, in the present embodiment, the predictive mode using can be previously selected optimum Match pattern, and those skilled in the art can adopt prior art to select optimum Match pattern, repeat no more herein.

Step 203, described the first prediction residual is carried out to the directional transform based on described predictive mode, described the second prediction residual is carried out to the conversion irrelevant with directivity.

Obtain because I3 adopts traditional inter prediction intraframe coding processing, it has the Intra prediction mode of Anisotropic Texture Information of sign, and P0, P1 and P2 adopt and between pixel, predict that intraframe coding processing obtains, and it does not exist Intra prediction mode.And MDDT must have Intra prediction mode, therefore, the present embodiment can adopt MDDT conversion to I3, adopt the transformation matrix corresponding with the Intraprediction mode using in step 202 to carry out MDDT conversion to I3, and P0, P1 and P2 are adopted to the conversion irrelevant with Anisotropic Texture Information, for example dct transform.

It should be noted that, in the present embodiment, for the DC pattern in Intra prediction mode, because it does not characterize Anisotropic Texture Information, thus do not use MDDT conversion, and can use dct transform.

The present embodiment, by the MB of required coding is carried out to down-sampling division, can make full use of the correlation between pixel, thereby improves precision of prediction, improves compression coding performance; By the residual MDDT of carrying out of the first prediction obtaining after inter prediction intraframe coding is processed, can remove the Anisotropic Texture Information correlation in the first prediction residual, improve the compression coding performance of Block3, and adopt the inter prediction intraframe coding result of Block3 as a reference Block0, Block1, Blcok2 to be carried out predicting and intraframe coding processing can further improving compression coding performance between pixel.

Fig. 5 is the flow chart of Image Coding processing method embodiment tri-of the present invention, and as shown in Figure 5, the method for the present embodiment can comprise:

Step 301, the macro block of required coding is carried out to down-sampling division, obtain a plurality of.

Step 302, adopt predictive mode to carry out inter prediction intraframe coding processing to first in described a plurality of, obtain and described first corresponding the first prediction residual, and the inter prediction intraframe coding result of applying described first carries out predicting between pixel intraframe coding processing to all the other pieces in described a plurality of, obtain and all the other pieces in described a plurality of corresponding the second prediction residual respectively.

In the present embodiment, step 301 and step 302 realize principle and the step 201 shown in above-mentioned Fig. 2 and step 202 to realize principle identical, repeat no more herein.

Step 303, described the first prediction residual and the second prediction residual are carried out to the directional transform based on described predictive mode.

The difference of Image Coding processing method embodiment bis-of the present invention shown in the present embodiment and Fig. 2 is, the present embodiment can adopt the predictive mode using in step 302 all to adopt MDDT to process to the first prediction residual and the second prediction residual.It should be noted that, in the present embodiment, the second prediction residual is carried out to the predictive mode that MDDT adopts while processing identical with the predictive mode of the first prediction residual.

For instance, as shown in Figure 4, because I3 adopts traditional inter prediction intraframe coding processing to obtain the first prediction residual, there is the Intra prediction mode that characterizes Anisotropic Texture Information, therefore, it can directly apply MDDT, and P0, P1 and P2 adopt to predict that intraframe coding processes the second prediction residual of obtaining respectively between pixel, there is not Intra prediction mode in it, therefore cannot directly apply MDDT.But, under the framework of the spatial domain prediction intraframe coding processing of dividing based on down-sampling, each down-sampling piece, Block0, Block1, Block2 and Block3 are an epitome of whole MB, and its grain direction information is consistent with the grain direction information of whole MB.That is to say, the grain direction information of each down-sampling piece is same or similar.Therefore, P0, P1 and P2 are in fact same or similar with the grain direction information of I3, therefore, also retain and have similar grain direction information respectively with P0, P1 and P2 in corresponding the second prediction residual.Therefore, P0, P1 and P2 can adopt the Intra prediction mode of I3 to carry out MDDT conversion.That is to say, when P0, P1 and P2 are carried out to MDDT, use the transformation matrix corresponding with the Intra prediction mode of I3.Like this, just, realized all down-sampling pieces in MB have all been used to MDDT conversion.

The present embodiment, by the MB of required coding is carried out to down-sampling division, can make full use of the correlation between pixel, thereby improves precision of prediction, improves compression coding performance; By first prediction of obtaining after inter prediction intraframe coding is processed residual and between pixel, predict intraframe coding process after the second prediction residual of obtaining all carry out MDDT conversion, can remove the Anisotropic Texture Information correlation in the first prediction residual and the second prediction residual, improve the compression coding performance of whole down-sampling pieces.

Fig. 6 is the flow chart of Image Coding processing method embodiment tetra-of the present invention, and as shown in Figure 6, the method for the present embodiment can comprise:

Step 401, macro block is carried out to the first down-sampling division, obtain first and a plurality of second, and carry out the second down-sampling division to described first, obtain the first sub-block and a plurality of the second sub-block.

The present embodiment adopts multilayer dividing mode to the MB of required coding.Specifically, the present embodiment can be divided on basis at the first down-sampling shown in Fig. 4, and I3 is carried out to the second down-sampling division again.For instance, image code processing device can carry out the first down-sampling division to 16 × 16MB of required coding, 1/2 down-sampling is divided, obtain first Block3 of 8 × 8 and 3 s' 8 × 8 second Block0, Block1 and Block2, then image code processing device can carry out the second down-sampling division to first Block3, for example 1/2 down-sampling is divided, and obtains the first sub-block sub-Block3 of 4 × 4 and the second sub-block sub-Block0, sub-Block1 and the sub-Block2 of 34 × 4.

It should be noted that, in the present embodiment, the first down-sampling is divided and 1/2 down-sampling division is all chosen in the division of the second down-sampling, but it will be understood by those skilled in the art that, the first down-sampling is divided and the division of the second down-sampling also can be different, those skilled in the art can choose down-sampling division methods according to the size of MB, are not restricted herein.

Step 402, employing predictive mode carry out inter prediction intraframe coding processing to described the first sub-block, obtain first prediction residual corresponding with described the first sub-block, and the inter prediction intraframe coding result of applying described the first sub-block carries out predicting between pixel intraframe coding processing to described a plurality of second and described a plurality of the second sub-blocks, obtain and described a plurality of second and described a plurality of the second sub-blocks corresponding the second prediction residual respectively.

In the present embodiment, image code processing device can adopt Intra prediction mode to carry out inter prediction intraframe coding processing to 4 × 4 sub-Block3, obtain the inter prediction intraframe coding result of this sub-Block3, and by the processing that subtracts of the original pixels of the sub-Block3 of this inter prediction intraframe coding result and 4 × 4, obtain first prediction residual corresponding with sub-Block3.Then, image code processing device can be with the inter prediction intraframe coding result of this sub-Block3 as a reference, all the other the second sub-block sub-Block0, sub-Block1 of 34 × 4 and sub-Block2 and 3 s' 8 × 8 second Block0, Block1, Blcok2 are carried out predicting between pixel intraframe coding processing, thereby obtain corresponding the second prediction residual respectively with sub-Block0, sub-Block1, sub-Block2, Block0, Block1 and Blcok2.

Fig. 7 is the structural representation that Image Coding processing method embodiment tetra-of the present invention carries out infra-frame prediction, as shown in Figure 7, wherein, I represents that sub-Block3 is carried out to inter prediction intraframe coding processing to be obtained the first prediction residual, p0, p1, p2, P0, P1, P2 and represent that respectively inter prediction intraframe coding result taking sub-Block3 is as reference, sub-Block0, sub-Block1, sub-Block2, Block0, Block1 and Block2 are carried out respectively predicting intraframe coding processing between pixel, obtain respectively the second prediction residual.

Step 403, described the first prediction residual is carried out to the directional transform based on described predictive mode, described the second prediction residual is carried out to the conversion irrelevant with directivity.

Because adopting traditional inter prediction intraframe coding processing, obtains I, it has the Intra prediction mode that characterizes Anisotropic Texture Information, p0, p1, p2, P0, P1 and P2 adopt and between pixel, predict that intraframe coding processing obtains, and it does not exist Intra prediction mode.And MDDT must have Intraprediction mode, therefore, the present embodiment can adopt MDDT conversion to I, adopt the transformation matrix corresponding with the Intra prediction mode using in step 402 to carry out MDDT conversion to I, and p0, p1, p2, P0, P1 and P2 are adopted to the conversion irrelevant with Anisotropic Texture Information, for example dct transform.

The present embodiment, by the MB of required coding being carried out to the division of multilayer down-sampling, can make full use of the correlation between pixel, thereby improves precision of prediction, improves compression coding performance; By the residual MDDT of carrying out of the first prediction obtaining after inter prediction intraframe coding is processed, can remove the Anisotropic Texture Information correlation in the first prediction residual, improve the compression coding performance of sub-Block3, and adopt the inter prediction intraframe coding result of sub-Block3 as a reference sub-Block0, sub-Block1, sub-Block2, Block0, Block1 and Block2 to be carried out predicting and intraframe coding processing can further improving compression coding performance between pixel.

Fig. 8 is the flow chart of Image Coding processing method embodiment five of the present invention, and as shown in Figure 8, the method for the present embodiment can comprise:

Step 501, macro block is carried out to the first down-sampling division, obtain first and a plurality of second, and carry out the second down-sampling division to described first, obtain the first sub-block and a plurality of the second sub-block.

Step 502, employing predictive mode carry out inter prediction intraframe coding processing to described the first sub-block, obtain first prediction residual corresponding with described the first sub-block, and the inter prediction intraframe coding result of applying described the first sub-block carries out predicting between pixel intraframe coding processing to described a plurality of second and described a plurality of the second sub-blocks, obtain and described a plurality of second and described a plurality of the second sub-blocks corresponding the second prediction residual respectively.

The step 501 of the present embodiment and step 502, its realize the step 401 shown in principle and Fig. 6 and step 402 to realize principle similar, repeat no more herein.

Step 503, described the first prediction residual and described the second prediction residual are carried out to the directional transform based on described predictive mode.

The difference of Image Coding processing method embodiment tetra-of the present invention shown in the present embodiment and Fig. 4 is, the present embodiment can adopt the predictive mode using in step 502 all to adopt MDDT to convert to the first prediction residual and the second prediction residual.

For instance, as shown in Figure 7, because I adopts traditional inter prediction intraframe coding processing to obtain the first prediction residual, there is the Intra prediction mode that characterizes Anisotropic Texture Information, therefore, it can directly apply MDDT, and p0, p1, p2, P0, P1 and P2 adopt to predict that intraframe coding processes the second prediction residual of obtaining respectively between pixel, there is not Intra prediction mode in it, therefore cannot directly apply MDDT.But, under the framework of the spatial domain prediction intraframe coding processing of dividing based on down-sampling, each down-sampling piece, be the epitome that sub-Block0, sub-Block 1, sub-Block2, sub-Block3, Block0, Block1 and Block2 are whole MB, its grain direction information is consistent with the grain direction information of whole MB.That is to say, the grain direction information of each down-sampling piece is same or similar.Therefore, p0, p1, p2, P0, P1 and P2 are in fact same or similar with the grain direction information of I, therefore, also retain and have similar grain direction information respectively with p0, p1, p2, P0, P1 and P2 in corresponding the second prediction residual.Therefore, p0, p1, p2, P0, P1 and P2 can adopt the Intra prediction mode of I to carry out MDDT conversion.That is to say, when p0, p1, p2, P0, P1 and P2 are carried out to MDDT, use the transformation matrix corresponding with the Intra prediction mode of I.Like this, just, realized all down-sampling pieces in MB have all been used to MDDT conversion.

The present embodiment, by the MB of required coding being carried out to the division of multilayer down-sampling, can make full use of the correlation between pixel, thereby improves precision of prediction, improves compression coding performance; By the first prediction residual of obtaining after inter prediction intraframe coding is processed and to predicting between pixel that the second prediction residual of obtaining after intraframe coding is processed all carries out MDDT, can remove the Anisotropic Texture Information correlation in the first prediction residual and the second prediction residual, further improve the compression coding performance of whole MB.

The flow chart of Fig. 9 Image Coding processing method of the present invention embodiment six, as shown in Figure 9, the method for the present embodiment can comprise:

Step 601, described macro block is carried out to down-sampling division, obtain first and a plurality of second, and carry out spatial domain to described first and divide and process, obtain a plurality of the first sub-blocks.

The difference of the step 501 shown in the present embodiment and Fig. 8 is, the present embodiment is first Block3 of 8 × 8 to be divided on spatial domain to the first sub-block sub-Block3 of 44 × 4, is 4 I sub-blocks corresponding to the form of expression of predictive mode.Figure 10 is the structural representation that Image Coding processing method embodiment six of the present invention carries out infra-frame prediction, as shown in figure 10, in the present embodiment, the sub-Block3 of 44 × 4 is the I sub-block that need to adopt inter prediction coding, all the other 3 second Block0, Block1 and Block2 are the P piece that need to adopt predictive coding between pixel, respectively corresponding P0, P1 and P2.

Step 602, employing predictive mode carry out respectively inter prediction intraframe coding processing to described a plurality of the first sub-blocks, obtain first prediction residual corresponding with described a plurality of the first sub-blocks difference, and the inter prediction intraframe coding result of applying described a plurality of the first sub-blocks carries out predicting between pixel intraframe coding processing to described a plurality of second, obtain and described a plurality of second corresponding the second prediction residual respectively.

In the present embodiment, image code processing device can adopt Intra prediction mode respectively the sub-Block3 of 44 × 4 to be carried out to inter prediction intraframe coding processing, obtain respectively the inter prediction intraframe coding result of these 4 sub-Block3, thereby obtain and 4 the first prediction residual that sub-Block3 difference is corresponding, and, this image code processing device can be looked as a reference as a whole by 4 inter prediction intraframe coding results, to all the other Block of 38 × 8, be Block0, Block1, Blcok2 carries out predicting between pixel intraframe coding processing, thereby obtain and Block0, Block1, Blcok2 is the second prediction residual of correspondence respectively.

Step 603, described the first prediction residual is carried out to the directional transform based on described predictive mode, described the second prediction residual is carried out to the conversion irrelevant with directivity.

In the present embodiment, due to the sub-Block3 of 44 × 4, to combine be the equal of an epitome of whole MB to 4 I sub-blocks, and single 4 × 4 sub-Block3, single I sub-block is not an epitome of whole MB.Therefore, in this case, P0, P1 and P2 cannot utilize the Intra prediction mode of I sub-block to carry out MDDT conversion.Therefore, the present embodiment adopts MDDT conversion to I sub-block, adopt the transformation matrix corresponding with the Intra prediction mode using in step 402 to carry out MDDT conversion to I sub-block, and P0, P1 and P2 are adopted to the conversion irrelevant with Anisotropic Texture Information, for example dct transform.

The present embodiment, by the MB of required coding being carried out to the division of multilayer down-sampling, can make full use of the correlation between pixel, thereby improves precision of prediction, improves compression coding performance; By the residual MDDT of carrying out of the first prediction obtaining after inter prediction intraframe coding is processed, can remove the Anisotropic Texture Information correlation in the first prediction residual, improve the compression coding performance of Block3, and adopt the inter prediction intraframe coding result of Block3 as a reference Block0, Block1 and Block2 to be carried out predicting and intraframe coding processing can further improving compression coding performance between pixel.

Figure 11 is the flow chart of Image Coding processing method embodiment seven of the present invention, and as shown in figure 11, the method for the present embodiment can comprise:

Step 701, described macro block is carried out to down-sampling division, obtain first and a plurality of second, and carry out spatial domain to described first and divide and process, obtain a plurality of the first sub-blocks.

Step 702, employing predictive mode carry out respectively inter prediction intraframe coding processing to described a plurality of the first sub-blocks, obtain first prediction residual corresponding with described a plurality of the first sub-blocks difference, and the inter prediction intraframe coding result of applying described a plurality of the first sub-blocks carries out predicting between pixel intraframe coding processing to described a plurality of second, obtain and described a plurality of second corresponding the second prediction residual respectively.

The step 701 of the present embodiment and step 702, its realize the step 601 shown in principle and Fig. 9 and step 602 to realize principle similar, repeat no more herein.

Step 703, described the first prediction residual and described the second prediction residual are carried out to the directional transform based on described predictive mode.

The difference of Image Coding processing method embodiment six of the present invention shown in the present embodiment and Fig. 9 is, the present embodiment can adopt the predictive mode using in step 702 all to adopt MDDT to convert to the first prediction residual and the second prediction residual.

Specifically, in the present embodiment, 4 I sub-blocks can be considered to have identical Intraprediction mode, so, the Anisotropic Texture Information of each I sub-block has still characterized the Anisotropic Texture Information of whole MB, and these Anisotropic Texture Informations can further be utilized by P0, P1 and P2, thereby makes P0, P1 and P2 all can use MDDT, thereby further remove the Anisotropic Texture Information correlation of retaining in the second prediction residual of P0, P1 and P2 piece, thereby promote compression coding performance.

Figure 12 is the flow chart of decoding processing method embodiment mono-of the present invention, and as shown in figure 12, the method for the present embodiment can comprise:

Step 104, to carrying out the directivity inverse transformation based on predictive mode with a plurality of prediction residual coefficients that at least one piece is corresponding in corresponding prediction residual coefficient respectively, obtain prediction residual.

Step 105, described prediction residual is carried out to the processing of spatial domain prediction intraframe decoder, obtain described a plurality of, wherein said spatial domain prediction intraframe decoder is processed and is at least comprised and adopt described predictive mode to carry out inter prediction intraframe decoder.

Step 106, to carry out up-sampling to described a plurality of synthetic, obtains decoded macro block.

The method of the present embodiment is the decoding process corresponding with the processing method of Image Coding shown in Fig. 1 embodiment mono-, it just adopts and the contrary process of processing procedure of the encoding processing of decoding in implementation, it realizes principle and the embodiment illustrated in fig. 1 effect that realizes principle and reach is similar, repeats no more herein.

Figure 13 is the flow chart of decoding processing method embodiment bis-of the present invention, and as shown in figure 13, the method for the present embodiment can comprise:

Step 204, to carrying out the directivity inverse transformation based on predictive mode with first corresponding the first prediction residual coefficient in described a plurality of, obtain the first prediction residual, the second prediction residual coefficient corresponding with all the other pieces in described a plurality of carried out to the inverse transformation irrelevant with directivity, obtain the second prediction residual.

Step 205, adopt described predictive mode to carry out the processing of inter prediction intraframe decoder to described the first prediction residual, obtain the inter prediction intraframe decoder result of described first, and the inter prediction intraframe decoder result of applying described first carries out predicting between pixel intraframe decoder processing to described the second prediction residual, obtain between the pixel of all the other pieces in described a plurality of decoding process result in predictive frame.

Step 206, that decoding process result in predictive frame between described inter prediction intraframe decoder result and described pixel is carried out to up-sampling is synthetic, obtains decoded macro block.

The method of the present embodiment is the decoding process corresponding with the processing method of Image Coding shown in Fig. 2 embodiment bis-, it just adopts and the contrary process of processing procedure of the encoding processing of decoding in implementation, it realizes principle and the embodiment illustrated in fig. 2 effect that realizes principle and reach is similar, repeats no more herein.

Figure 14 is the flow chart of decoding processing method embodiment tri-of the present invention, and as shown in figure 14, the method for the present embodiment can comprise:

Step 304, to first corresponding the first prediction residual coefficient in described a plurality of and with described a plurality of in the second prediction residual coefficient corresponding to all the other pieces carry out the directivity inverse transformation based on same prediction mode, obtain respectively the first prediction residual and the second prediction residual.

Step 305, adopt described predictive mode to carry out the processing of inter prediction intraframe decoder to described the first prediction residual, obtain the inter prediction intraframe decoder result of described first, and the inter prediction intraframe decoder result of applying described first carries out predicting between pixel intraframe decoder processing to described the second prediction residual, obtain between the pixel of all the other pieces in described a plurality of decoding process result in predictive frame.

Step 306, that decoding process result in predictive frame between described inter prediction intraframe decoder result and described pixel is carried out to up-sampling is synthetic, obtains decoded macro block.

The method of the present embodiment is the decoding process corresponding with the processing method of Image Coding shown in Fig. 5 embodiment tri-, it just adopts and the contrary process of processing procedure of the encoding processing of decoding in implementation, it realizes principle and the embodiment illustrated in fig. 5 effect that realizes principle and reach is similar, repeats no more herein.

Figure 15 is the flow chart of decoding processing method embodiment tetra-of the present invention, and as shown in figure 15, the method for the present embodiment can comprise:

Step 404, the first prediction residual coefficient corresponding with the first sub-block in described a plurality of carried out to the directivity inverse transformation based on predictive mode, obtain the first prediction residual, and to a plurality of second and a plurality of the second sub-blocks in described a plurality of respectively corresponding the second prediction residual coefficient carry out the inverse transformation irrelevant with directivity, obtain the second prediction residual.

Step 405, adopt described predictive mode to carry out the processing of inter prediction intraframe decoder to described the first prediction residual, obtain the inter prediction intraframe decoder result of described the first sub-block, and the inter prediction intraframe decoder result of applying described the first sub-block carries out predicting between pixel intraframe decoder processing to described the second prediction residual, obtain between the pixel of a plurality of second and a plurality of the second sub-blocks in described a plurality of decoding process result in predictive frame.

Between the pixel of step 406, inter prediction intraframe decoder result to described the first sub-block and described a plurality of the second sub-blocks, in predictive frame, decoding process result carries out that the second up-sampling is synthetic obtains described first, and it is synthetic that decoding process result in predictive frame between described first and described a plurality of second pixel is carried out to the first up-sampling, obtains decoded macro block.

The method of the present embodiment is the decoding process corresponding with the processing method of Image Coding shown in Fig. 6 embodiment tetra-, it just adopts and the contrary process of processing procedure of the encoding processing of decoding in implementation, it realizes principle and the embodiment illustrated in fig. 6 effect that realizes principle and reach is similar, repeats no more herein.

Figure 16 is the flow chart of decoding processing method embodiment five of the present invention, and as shown in figure 16, the method for the present embodiment can comprise:

Step 504, to the first prediction residual coefficient corresponding with the first sub-block in described a plurality of and with described a plurality of in a plurality of second and the second prediction residual coefficients corresponding to a plurality of the second sub-block carry out the directivity inverse transformation based on same prediction mode, obtain respectively the first prediction residual and the second prediction residual.

Step 505, adopt described predictive mode to carry out the processing of inter prediction intraframe decoder to described the first prediction residual, obtain the inter prediction intraframe decoder result of described the first sub-block, and the inter prediction intraframe decoder result of applying described the first sub-block carries out predicting between pixel intraframe decoder processing to described the second prediction residual, obtain between the pixel of a plurality of second and a plurality of the second sub-blocks in described a plurality of decoding process result in predictive frame.

Between the pixel of step 506, inter prediction intraframe decoder result to described the first sub-block and described a plurality of the second sub-blocks, in predictive frame, decoding process result carries out that the second up-sampling is synthetic obtains described first, and it is synthetic that decoding process result in predictive frame between described first and described a plurality of second pixel is carried out to the first up-sampling, obtains decoded macro block.

The method of the present embodiment is the decoding process corresponding with the processing method of Image Coding shown in Fig. 8 embodiment five, it just adopts and the contrary process of processing procedure of the encoding processing of decoding in implementation, it realizes principle and the embodiment illustrated in fig. 8 effect that realizes principle and reach is similar, repeats no more herein.

Figure 17 is the flow chart of decoding processing method embodiment six of the present invention, and as shown in figure 17, the method for the present embodiment can comprise:

Step 604, the first prediction residual coefficient corresponding with a plurality of the first sub-blocks in described a plurality of carried out respectively to the directivity inverse transformation based on predictive mode, obtain the first prediction residual, and to carrying out the inverse transformation irrelevant with directivity with a plurality of second corresponding the second prediction residual coefficient in described a plurality of, obtain the second prediction residual.

Step 605, adopt described predictive mode to carry out the processing of inter prediction intraframe decoder to described the first prediction residual, obtain the inter prediction intraframe decoder result of described a plurality of the first sub-blocks, and the inter prediction intraframe decoder result of applying described a plurality of the first sub-blocks carries out predicting between pixel intraframe decoder processing to described the second prediction residual, obtain in described a plurality of between the pixel of a plurality of second decoding process result in predictive frame.

Step 606, the inter prediction intraframe decoder result of described a plurality of the first sub-blocks is carried out to spatial domain is synthetic processes, obtain described first, and it is synthetic that decoding process result in predictive frame between described first and described a plurality of second pixel is carried out to up-sampling, obtains decoded macro block.

The method of the present embodiment is the decoding process corresponding with the processing method of Image Coding shown in Fig. 9 embodiment six, it just adopts and the contrary process of processing procedure of the encoding processing of decoding in implementation, it realizes principle and the embodiment illustrated in fig. 9 effect that realizes principle and reach is similar, repeats no more herein.

Figure 18 is the flow chart of decoding processing method embodiment seven of the present invention, and as shown in figure 18, the method for the present embodiment can comprise:

Step 704, to the first prediction residual coefficient corresponding with a plurality of the first sub-blocks in described a plurality of and with described a plurality of in a plurality of second corresponding the second prediction residual coefficient carry out the directivity inverse transformation based on same prediction mode, obtain respectively the first prediction residual and the second prediction residual.

Step 705, adopt described predictive mode to carry out the processing of inter prediction intraframe decoder to described the first prediction residual, obtain the inter prediction intraframe decoder result of described a plurality of the first sub-blocks, and the inter prediction intraframe decoder result of applying described plural the first sub-block carries out predicting between pixel intraframe decoder processing to described the second prediction residual, obtain in described a plurality of between the pixel of a plurality of second decoding process result in predictive frame.

Step 706, the inter prediction intraframe decoder result of described a plurality of the first sub-blocks is carried out to spatial domain is synthetic processes, obtain described first, and it is synthetic that decoding process result in predictive frame between described first and described a plurality of second pixel is carried out to up-sampling, obtains decoded macro block.

The method of the present embodiment is the decoding process corresponding with the processing method of Image Coding shown in Figure 11 embodiment seven, it just adopts and the contrary process of processing procedure of the encoding processing of decoding in implementation, it realizes principle and the embodiment illustrated in fig. 11 effect that realizes principle and reach is similar, repeats no more herein.

Figure 19 is the structural representation of image code processing device embodiment mono-of the present invention, as shown in figure 19, the device of the present embodiment can comprise: coding is divided module 11, coded prediction module 12 and transcoding, coding transform module 13, wherein, coding is divided module 11 for the macro block of required coding is carried out to down-sampling division, obtains a plurality of; Coded prediction module 12 is for carrying out spatial domain prediction intraframe coding to described a plurality of, obtains and described a plurality of corresponding prediction residual respectively, and wherein said spatial domain prediction intraframe coding at least comprises that employing predictive mode carries out inter prediction intraframe coding; Transcoding, coding transform module 13 is for carrying out the directional transform based on described predictive mode at least one prediction residual of described prediction residual.

The device of the present embodiment can be for the technical scheme of the embodiment of the method one shown in execution graph 1, and it is similar with the technique effect reaching that it realizes principle, repeats no more herein.

Figure 20 is the structural representation of image code processing device embodiment bis-of the present invention, as shown in figure 20, the present embodiment is on the basis of the image code processing device embodiment mono-shown in Figure 19, further, coded prediction module 12 comprises: the first coded prediction unit 121 and the second coded prediction unit 122, wherein, the first coded prediction unit 121 is for adopting predictive mode to carry out inter prediction intraframe coding processing to described a plurality of first, obtains and described first corresponding the first prediction residual; The second coded prediction unit 122 carries out predicting intraframe coding processing between pixel to all the other pieces of described a plurality of for the inter prediction intraframe coding result of applying described first, obtains with all the other pieces in described a plurality of and distinguish the second corresponding prediction residual; Transcoding, coding transform module 13 comprises: the first transcoding, coding transform unit 131 and the second transcoding, coding transform unit 132, and wherein, the first transcoding, coding transform unit 131 is for carrying out the directional transform based on described predictive mode to described the first prediction residual; The second transcoding, coding transform unit 132 is for carrying out the conversion irrelevant with directivity to described the second prediction residual.

The device of the present embodiment can be for the technical scheme of the embodiment of the method two shown in execution graph 2, and it is similar with the technique effect reaching that it realizes principle, repeats no more herein.

Figure 21 is the structural representation of image code processing device embodiment tri-of the present invention, as shown in figure 21, the present embodiment is on the basis of the image code processing device embodiment mono-shown in Figure 19, further, coded prediction module 12 comprises: the 3rd coded prediction unit 123 and the 4th coded prediction unit 124, wherein, the 3rd coded prediction unit 123 is for adopting predictive mode to carry out inter prediction intraframe coding processing to described a plurality of first, obtains and described first corresponding the first prediction residual; The 4th coded prediction unit 124 carries out predicting intraframe coding processing between pixel to all the other pieces of described a plurality of for the inter prediction intraframe coding result of applying described first, obtains with all the other pieces in described a plurality of and distinguish the second corresponding prediction residual; Transcoding, coding transform module 13 is specifically for carrying out the directional transform based on described predictive mode to described the first prediction residual and the second prediction residual.

The device of the present embodiment can be for the technical scheme of the embodiment of the method three shown in execution graph 5, and it is similar with the technique effect reaching that it realizes principle, repeats no more herein.

Figure 22 is the structural representation of image code processing device embodiment tetra-of the present invention, as shown in figure 22, the present embodiment is on the basis of the image code processing device embodiment mono-shown in Figure 19, further, coding is divided module 11 and is comprised: the first coded sample unit 111 and the second coded sample unit 112, wherein, the first coded sample unit 111, for described macro block being carried out to the first down-sampling division, obtains first and a plurality of second; The second coded sample unit 112, for carrying out the second down-sampling division to described first, obtains the first sub-block and a plurality of the second sub-block; Coded prediction module 12 comprises: the 5th coded prediction unit 125 and the 6th coded prediction unit 126, wherein the 5th coded prediction unit 125, for adopting predictive mode to carry out inter prediction intraframe coding processing to described the first sub-block, obtains first prediction residual corresponding with described the first sub-block; The 6th coded prediction unit 126 for and the inter prediction intraframe coding result of applying described the first sub-block described a plurality of second and described a plurality of the second sub-blocks are carried out predicting between pixel intraframe coding processing, obtain and described a plurality of second and described a plurality of the second sub-blocks corresponding the second prediction residual respectively; Transcoding, coding transform module 13 comprises: the 3rd transcoding, coding transform unit 133 and the 4th transcoding, coding transform unit 134, and wherein, the 3rd transcoding, coding transform unit 133 is for carrying out the directional transform based on described predictive mode to described the first prediction residual; Or described the first prediction residual and described the second prediction residual are carried out to the directional transform based on described predictive mode; The 4th transcoding, coding transform unit 134 is for carrying out the conversion irrelevant with directivity to described the second prediction residual.

The device of the present embodiment can be for the technical scheme of the embodiment of the method five shown in the embodiment of the method four shown in execution graph 6 or Fig. 8, and it is similar with the technique effect reaching that it realizes principle, repeats no more herein.

Figure 23 is the structural representation of image code processing device embodiment five of the present invention, as shown in figure 23, the present embodiment is on the basis of the image code processing device embodiment mono-shown in Figure 19, further, coding is divided module 11 and is comprised: the 3rd coded sample unit 113 and the 4th coded sample unit 114, wherein, the 3rd coded sample unit 113, for described macro block is carried out to down-sampling division, obtains first and a plurality of second; Processing is divided for carrying out spatial domain to described first in the 4th coded sample unit 114, obtains a plurality of the first sub-blocks; Coded prediction module 12 comprises: the 7th coded prediction unit 127 and the 8th coded prediction unit 128, wherein, the 7th coded prediction unit 127, for adopting predictive mode to carry out respectively inter prediction intraframe coding processing to described a plurality of the first sub-blocks, obtains first prediction residual corresponding with described a plurality of the first sub-blocks difference; The 8th coded prediction unit 128 carries out predicting intraframe coding processing between pixel to described a plurality of second for the inter prediction intraframe coding result of applying described a plurality of the first sub-blocks, obtains and described a plurality of second the second prediction residual that difference is corresponding; Transcoding, coding transform module 13, specifically for described the first prediction residual is carried out to the directional transform based on described predictive mode, is carried out the conversion irrelevant with directivity to described the second prediction residual; Or described the first prediction residual and described the second prediction residual are carried out to the directional transform based on described predictive mode.

The device of the present embodiment can be for the technical scheme of the embodiment of the method seven shown in the embodiment of the method six shown in execution graph 9 or Figure 11, and it is similar with the technique effect reaching that it realizes principle, repeats no more herein.

Figure 24 is the structural representation of image decoding processing unit embodiment mono-of the present invention, as shown in figure 24, the image decoding processing unit of the present embodiment, can comprise: decoding conversion module 21, decoding prediction module 22 and decoding synthesis module 23, wherein, decoding conversion module 21, for to carrying out the directivity inverse transformation based on predictive mode with a plurality of prediction residual coefficients that corresponding at least one piece of prediction residual coefficient is corresponding respectively, obtains prediction residual; Decoding prediction module 22, for described prediction residual being carried out to the processing of spatial domain prediction intraframe decoder, is obtained described a plurality of, and wherein said spatial domain prediction intraframe decoder is processed and at least comprised that the described predictive mode of employing carries out inter prediction intraframe decoder; Decoding synthesis module 23 synthesizes for carrying out up-sampling to described a plurality of, obtains decoded macro block.

The device of the present embodiment can be for the technical scheme of the embodiment of the method one shown in execution graph 12, and it realizes principle and the embodiment illustrated in fig. 12 effect that realizes principle and reach is similar, repeats no more herein.

Figure 25 is the structural representation of image decoding processing unit embodiment bis-of the present invention, as shown in figure 25, the image decoding processing unit of the present embodiment is on the basis of the image decoding processing unit embodiment mono-shown in Figure 24, further, decoding conversion module 21 comprises: the first decoding converter unit 211 and the second decoding converter unit 212, wherein, the first decoding converter unit 211, for to carrying out the directivity inverse transformation based on predictive mode with described a plurality of first corresponding the first prediction residual coefficient, obtains the first prediction residual; The second decoding converter unit 212, for the second prediction residual coefficient corresponding with described a plurality of all the other pieces carried out to the inverse transformation irrelevant with directivity, obtains the second prediction residual; Decoding prediction module 22 comprises: the first decoding predicting unit 221 and the second decoding predicting unit 222, wherein, the first decoding predicting unit 221, for adopting described predictive mode to carry out the processing of inter prediction intraframe decoder to described the first prediction residual, is obtained the inter prediction intraframe decoder result of described first; The second decoding predicting unit 222 carries out predicting between pixel intraframe decoder processing for the inter prediction intraframe decoder result of applying described first to described the second prediction residual, obtains between the pixel of all the other pieces in described a plurality of decoding process result in predictive frame; Decoding synthesis module 23 is synthetic specifically for decoding process result in predictive frame between described inter prediction intraframe decoder result and described pixel being carried out to up-sampling, obtains decoded macro block.

The device of the present embodiment can be for the technical scheme of the embodiment of the method two shown in execution graph 13, and it realizes principle and the embodiment illustrated in fig. 13 effect that realizes principle and reach is similar, repeats no more herein.

Figure 26 is the structural representation of image decoding processing unit embodiment tri-of the present invention, as shown in figure 26, the image decoding processing unit of the present embodiment is on the basis of the image decoding processing unit embodiment mono-shown in Figure 24, further, decoding conversion module 21 specifically for to first corresponding the first prediction residual coefficient in described a plurality of and with described a plurality of in the second prediction residual coefficient corresponding to all the other pieces carry out the directivity inverse transformation based on same prediction mode, obtain respectively the first prediction residual and the second prediction residual; Decoding prediction module 22 comprises: the 3rd decoding predicting unit 223 and the 4th decoding predicting unit 224, wherein, the 3rd decoding predicting unit 223, for adopting described predictive mode to carry out the processing of inter prediction intraframe decoder to described the first prediction residual, is obtained the inter prediction intraframe decoder result of described first; The 4th decoding predicting unit 224 carries out predicting between pixel intraframe decoder processing for the inter prediction intraframe decoder result of applying described first to described the second prediction residual, obtains between the pixel of all the other pieces in described a plurality of decoding process result in predictive frame; Decoding synthesis module 23 is synthetic specifically for decoding process result in predictive frame between described inter prediction intraframe decoder result and described pixel being carried out to up-sampling, obtains decoded macro block.

The device of the present embodiment can be for the technical scheme of the embodiment of the method three shown in execution graph 14, and it realizes principle and the embodiment illustrated in fig. 14 effect that realizes principle and reach is similar, repeats no more herein.

Figure 27 is the structural representation of image decoding processing unit embodiment tetra-of the present invention, as shown in figure 27, the image decoding processing unit of the present embodiment is on the basis of the image decoding processing unit embodiment mono-shown in Figure 24, further, decoding conversion module 21 comprises: the 3rd decoding converter unit 213 and the 4th decoding converter unit 214, wherein, the 3rd decoding converter unit 213, for the first prediction residual coefficient corresponding with described a plurality of the first sub-blocks carried out to the directivity inverse transformation based on predictive mode, obtains the first prediction residual; The 4th decoding converter unit 214 for to described a plurality of a plurality of second and a plurality of the second sub-blocks respectively corresponding the second prediction residual coefficient carry out the inverse transformation irrelevant with directivity, obtain the second prediction residual; Decoding prediction module 22 comprises: the 5th decoding predicting unit 225 and the 6th decoding predicting unit 226, wherein, the 5th decoding predicting unit 225, for adopting described predictive mode to carry out the processing of inter prediction intraframe decoder to described the first prediction residual, is obtained the inter prediction intraframe decoder result of described the first sub-block; The 6th decoding predicting unit 226 carries out predicting between pixel intraframe decoder processing for the inter prediction intraframe decoder result of applying described the first sub-block to described the second prediction residual, obtains between the pixel of a plurality of second and a plurality of the second sub-blocks in described a plurality of decoding process result in predictive frame; Decoding synthesis module 23 comprises: the first decoding sampling unit 231 and the second decoding sampling unit 232, wherein, the first decoding sampling unit 231 carries out for decoding process result in predictive frame between the pixel of the inter prediction intraframe decoder result to described the first sub-block and described a plurality of the second sub-blocks that the second up-sampling is synthetic obtains described first; The second decoding sampling unit 232 is synthetic for decoding process result in predictive frame between described first and described a plurality of second pixel being carried out to the first up-sampling, obtains decoded macro block.

The device of the present embodiment can be for the technical scheme of the embodiment of the method four shown in execution graph 15, and it realizes principle and the embodiment illustrated in fig. 15 effect that realizes principle and reach is similar, repeats no more herein.

Figure 28 is the structural representation of image decoding processing unit embodiment five of the present invention, as shown in figure 28, the image decoding processing unit of the present embodiment is on the basis of the image decoding processing unit embodiment mono-shown in Figure 24, further, decoding conversion module 21 specifically for the first prediction residual coefficient to corresponding with the first sub-block in described a plurality of and with described a plurality of in a plurality of second and the second prediction residual coefficients corresponding to a plurality of the second sub-blocks carry out the directivity inverse transformation based on same prediction mode, obtain respectively the first prediction residual and the second prediction residual, decoding prediction module 22 comprises: the 7th decoding predicting unit 227 and the 8th decoding predicting unit 228, wherein, the 7th decoding predicting unit 227, for adopting described predictive mode to carry out the processing of inter prediction intraframe decoder to described the first prediction residual, is obtained the inter prediction intraframe decoder result of described the first sub-block, the 8th decoding predicting unit 228 carries out predicting between pixel intraframe decoder processing for the inter prediction intraframe decoder result of applying described the first sub-block to described the second prediction residual, obtains between the pixel of a plurality of second and a plurality of the second sub-blocks in described a plurality of decoding process result in predictive frame, decoding synthesis module 23 comprises: the 3rd decoding sampling unit 233 and the 4th decoding sampling unit 234, wherein, the 3rd decoding sampling unit 233 carries out for decoding process result in predictive frame between the pixel of the inter prediction intraframe decoder result to described the first sub-block and described a plurality of the second sub-blocks that the second up-sampling is synthetic obtains described first, the 4th decoding sampling unit 234 is synthetic for decoding process result in predictive frame between described first and described a plurality of second pixel being carried out to the first up-sampling, obtains decoded macro block.

The device of the present embodiment can be for the technical scheme of the embodiment of the method five shown in execution graph 16, and it realizes principle and the embodiment illustrated in fig. 16 effect that realizes principle and reach is similar, repeats no more herein.

Figure 29 is the structural representation of image decoding processing unit embodiment six of the present invention, as shown in figure 29, the image decoding processing unit of the present embodiment is on the basis of the image decoding processing unit embodiment mono-shown in Figure 24, further, decoding conversion module 21 comprises: the 5th decoding converter unit 215 and the 6th decoding converter unit 216, wherein, the 5th decoding converter unit 215, for the first prediction residual coefficient corresponding with described a plurality of a plurality of the first sub-blocks carried out respectively to the directivity inverse transformation based on predictive mode, obtains the first prediction residual; The 6th decoding converter unit 216, for the second prediction residual coefficient corresponding with described a plurality of a plurality of second carried out to the inverse transformation irrelevant with directivity, obtains the second prediction residual; Decoding prediction module 22 comprises: the 9th decoding predicting unit 229 and the tenth decoding predicting unit 230, wherein, the 9th decoding predicting unit 229, for adopting described predictive mode to carry out the processing of inter prediction intraframe decoder to described the first prediction residual, is obtained the inter prediction intraframe decoder result of described a plurality of the first sub-blocks; The tenth decoding predicting unit 230 carries out predicting between pixel intraframe decoder processing for the inter prediction intraframe decoder result of applying described a plurality of the first sub-blocks to described the second prediction residual, obtains in described a plurality of between the pixel of a plurality of second decoding process result in predictive frame; Decoding synthesis module 23 comprises: the 6th decoding sampling unit 236 and the 7th decoding sampling unit 237, wherein, the 6th decoding sampling unit 236 carries out the synthetic processing in spatial domain for the inter prediction intraframe decoder result to described a plurality of the first sub-blocks, obtains described first; The 7th decoding sampling unit 237 is synthetic for decoding process result in predictive frame between described first and described a plurality of second pixel being carried out to up-sampling, obtains decoded macro block.

The device of the present embodiment can be for the technical scheme of the embodiment of the method six shown in execution graph 17, and it realizes principle and the embodiment illustrated in fig. 17 effect that realizes principle and reach is similar, repeats no more herein.

One of ordinary skill in the art will appreciate that: all or part of step that realizes said method embodiment can complete by the relevant hardware of program command, aforesaid program can be stored in a computer read/write memory medium, this program, in the time carrying out, is carried out the step that comprises said method embodiment; And aforesaid storage medium comprises: various media that can be program code stored such as ROM, RAM, magnetic disc or CDs.

Finally it should be noted that: above embodiment only, in order to technical scheme of the present invention to be described, is not intended to limit; Although the present invention is had been described in detail with reference to previous embodiment, those of ordinary skill in the art is to be understood that: its technical scheme that still can record aforementioned each embodiment is modified, or part technical characterictic is wherein equal to replacement; And these amendments or replacement do not make the essence of appropriate technical solution depart from the spirit and scope of various embodiments of the present invention technical scheme.

Claims

1. an Image Coding processing method, is characterized in that, comprising:

At least one prediction residual in described prediction residual is carried out to the directional transform based on described predictive mode;

Wherein, the macro block of required coding is carried out to down-sampling division, obtains a plurality of, comprising:

Pixel is extracted at interval, forms described a plurality of;

Describedly carry out spatial domain prediction intraframe coding to described a plurality of, obtain and described a plurality of corresponding prediction residual respectively, wherein said spatial domain prediction intraframe coding at least comprises that employing predictive mode carries out inter prediction intraframe coding, comprising:

Adopt predictive mode to carry out inter prediction intraframe coding processing to first in described a plurality of, obtain and described first corresponding the first prediction residual, and the inter prediction intraframe coding result of applying described first carries out predicting between pixel intraframe coding processing to all the other pieces in described a plurality of, obtain and all the other pieces in described a plurality of corresponding the second prediction residual respectively.

2. Image Coding processing method according to claim 1, is characterized in that,

Described at least one prediction residual in described prediction residual is carried out to the directional transform based on described predictive mode, comprising:

Described the first prediction residual is carried out to the directional transform based on described predictive mode, described the second prediction residual is carried out to the conversion irrelevant with directivity.

3. Image Coding processing method according to claim 1, is characterized in that,

Described the first prediction residual and the second prediction residual are carried out to the directional transform based on described predictive mode.

4. Image Coding processing method according to claim 1, is characterized in that, described the macro block of required coding is carried out to down-sampling division, obtains a plurality of, comprising:

Described macro block is carried out to the first down-sampling division, obtain first and a plurality of second, and carry out the second down-sampling division to described first, obtain the first sub-block and a plurality of the second sub-block;

Adopt predictive mode to carry out inter prediction intraframe coding processing to described the first sub-block, obtain first prediction residual corresponding with described the first sub-block, and the inter prediction intraframe coding result of applying described the first sub-block carries out predicting between pixel intraframe coding processing to described a plurality of second and described a plurality of the second sub-blocks, obtain and described a plurality of second and described a plurality of the second sub-blocks corresponding the second prediction residual respectively;

Described the first prediction residual is carried out to the directional transform based on described predictive mode, described the second prediction residual is carried out to the conversion irrelevant with directivity; Or described the first prediction residual and described the second prediction residual are carried out to the directional transform based on described predictive mode.

5. Image Coding processing method according to claim 1, is characterized in that, described the macro block of required coding is carried out to down-sampling division, obtains a plurality of, comprising:

Described macro block is carried out to down-sampling division, obtain first and a plurality of second, and carry out spatial domain to described first and divide processing, obtain a plurality of the first sub-blocks;

Adopt predictive mode to carry out respectively inter prediction intraframe coding processing to described a plurality of the first sub-blocks, obtain first prediction residual corresponding with described a plurality of the first sub-blocks difference, and the inter prediction intraframe coding result of applying described a plurality of the first sub-blocks carries out predicting between pixel intraframe coding processing to described a plurality of second, obtain and described a plurality of second corresponding the second prediction residual respectively;

6. an image decoding processing method, is characterized in that, comprising:

Carry out up-sampling to described a plurality of and synthesize, obtain decoded macro block;

Described described prediction residual is carried out to the processing of spatial domain prediction intraframe decoder, obtain described a plurality of, wherein said spatial domain prediction intraframe decoder is processed and is at least comprised and adopt described predictive mode to carry out inter prediction intraframe decoder, comprising:

Adopt described predictive mode to carry out the processing of inter prediction intraframe decoder to the first prediction residual, obtain in described a plurality of the inter prediction intraframe decoder result of first, and the inter prediction intraframe decoder result of applying described first carries out predicting between pixel intraframe decoder processing to the second prediction residual, obtain between the pixel of all the other pieces in described a plurality of decoding process result in predictive frame.

7. image decoding processing method according to claim 6, it is characterized in that, describedly obtain prediction residual to carrying out the directivity inverse transformation based on predictive mode with a plurality of prediction residual coefficients that at least one piece is corresponding in corresponding prediction residual coefficient respectively, comprising:

To carrying out the directivity inverse transformation based on predictive mode with first corresponding the first prediction residual coefficient in described a plurality of, obtain the first prediction residual, the second prediction residual coefficient corresponding with all the other pieces in described a plurality of carried out to the inverse transformation irrelevant with directivity, obtain the second prediction residual;

Described to carry out up-sampling to described a plurality of synthetic, obtains decoded macro block, comprising:

Decoding process result in predictive frame between described inter prediction intraframe decoder result and described pixel is carried out to up-sampling and synthesize, obtain decoded macro block.

8. image decoding processing method according to claim 6, it is characterized in that, describedly obtain prediction residual to carrying out the directivity inverse transformation based on predictive mode with a plurality of prediction residual coefficients that at least one piece is corresponding in corresponding prediction residual coefficient respectively, comprising:

To with first corresponding the first prediction residual coefficient in described a plurality of and with described a plurality of in the second prediction residual coefficient corresponding to all the other pieces carry out the directivity inverse transformation based on same prediction mode, obtain respectively the first prediction residual and the second prediction residual;

9. image decoding processing method according to claim 6, it is characterized in that, describedly obtain prediction residual to carrying out the directivity inverse transformation based on predictive mode with a plurality of prediction residual coefficients that at least one piece is corresponding in corresponding prediction residual coefficient respectively, comprising:

The first prediction residual coefficient corresponding with the first sub-block in described a plurality of carried out to the directivity inverse transformation based on predictive mode, obtain the first prediction residual, and to a plurality of second and a plurality of the second sub-blocks in described a plurality of respectively corresponding the second prediction residual coefficient carry out the inverse transformation irrelevant with directivity, obtain the second prediction residual;

Adopt described predictive mode to carry out the processing of inter prediction intraframe decoder to described the first prediction residual, obtain the inter prediction intraframe decoder result of described the first sub-block, and the inter prediction intraframe decoder result of applying described the first sub-block carries out predicting between pixel intraframe decoder processing to described the second prediction residual, obtain between the pixel of a plurality of second and a plurality of the second sub-blocks in described a plurality of decoding process result in predictive frame;

Between the inter prediction intraframe decoder result to described the first sub-block and the pixel of described a plurality of the second sub-blocks, in predictive frame, decoding process result carries out that the second up-sampling is synthetic obtains described first, and it is synthetic that decoding process result in predictive frame between described first and described a plurality of second pixel is carried out to the first up-sampling, obtains decoded macro block.

10. image decoding processing method according to claim 6, is characterized in that, described to carrying out the directivity inverse transformation based on predictive mode with a plurality of prediction residual coefficients that at least one piece is corresponding in corresponding prediction residual coefficient respectively, comprising:

To the first prediction residual coefficient corresponding with the first sub-block in described a plurality of and with described a plurality of in a plurality of second and the second prediction residual coefficients corresponding to a plurality of the second sub-blocks carry out the directivity inverse transformation based on same prediction mode, obtain respectively the first prediction residual and the second prediction residual;

11. image decoding processing methods according to claim 6, is characterized in that, described to carrying out the directivity inverse transformation based on predictive mode with a plurality of prediction residual coefficients that at least one piece is corresponding in corresponding prediction residual coefficient respectively, comprising:

The first prediction residual coefficient corresponding with a plurality of the first sub-blocks in described a plurality of carried out respectively to the directivity inverse transformation based on predictive mode, obtain the first prediction residual, and to carrying out the inverse transformation irrelevant with directivity with a plurality of second corresponding the second prediction residual coefficient in described a plurality of, obtain the second prediction residual;

Adopt described predictive mode to carry out the processing of inter prediction intraframe decoder to described the first prediction residual, obtain the inter prediction intraframe decoder result of described a plurality of the first sub-blocks, and the inter prediction intraframe decoder result of applying described a plurality of the first sub-blocks carries out predicting between pixel intraframe decoder processing to described the second prediction residual, obtain in described a plurality of between the pixel of a plurality of second decoding process result in predictive frame;

The inter prediction intraframe decoder result of described a plurality of the first sub-blocks is carried out to the synthetic processing in spatial domain, obtain described first, and it is synthetic that decoding process result in predictive frame between described first and described a plurality of second pixel is carried out to up-sampling, obtains decoded macro block.

12. image decoding processing methods according to claim 6, is characterized in that, described to carrying out the directivity inverse transformation based on predictive mode with a plurality of prediction residual coefficients that at least one piece is corresponding in corresponding prediction residual coefficient respectively, comprising:

To the first prediction residual coefficient corresponding with a plurality of the first sub-blocks in described a plurality of and with described a plurality of in a plurality of second corresponding the second prediction residual coefficient carry out the directivity inverse transformation based on same prediction mode, obtain respectively the first prediction residual and the second prediction residual;

Adopt described predictive mode to carry out the processing of inter prediction intraframe decoder to described the first prediction residual, obtain the inter prediction intraframe decoder result of described a plurality of the first sub-blocks, and the inter prediction intraframe decoder result of applying described plural the first sub-block carries out predicting between pixel intraframe decoder processing to described the second prediction residual, obtain in described a plurality of between the pixel of a plurality of second decoding process result in predictive frame;

13. 1 kinds of image code processing devices, is characterized in that, comprising:

Transcoding, coding transform module, for carrying out the directional transform based on described predictive mode at least one prediction residual of described prediction residual;

Wherein, described coding is divided module, extracts pixel specifically for interval, forms described a plurality of;

Described coded prediction module, comprising:

The first coded prediction unit, for adopting predictive mode to carry out inter prediction intraframe coding processing to described a plurality of first, obtains and described first corresponding the first prediction residual;

The second coded prediction unit, for the inter prediction intraframe coding result of applying described first, all the other pieces of described a plurality of are carried out predicting intraframe coding processing between pixel, obtain with all the other pieces in described a plurality of and distinguish the second corresponding prediction residual.

14. image code processing devices according to claim 13, is characterized in that,

Described transcoding, coding transform module, comprising:

The first transcoding, coding transform unit, for carrying out the directional transform based on described predictive mode to described the first prediction residual;

The second transcoding, coding transform unit, for carrying out the conversion irrelevant with directivity to described the second prediction residual.

15. image code processing devices according to claim 13, is characterized in that, described transcoding, coding transform module is specifically for carrying out the directional transform based on described predictive mode to described the first prediction residual and the second prediction residual.

16. image code processing devices according to claim 13, is characterized in that, described coding is divided module, comprising:

The first coded sample unit, for described macro block being carried out to the first down-sampling division, obtains first and a plurality of second;

The second coded sample unit, for carrying out the second down-sampling division to described first, obtains the first sub-block and a plurality of the second sub-block;

Described coded prediction module, comprising:

The 5th coded prediction unit, for adopting predictive mode to carry out inter prediction intraframe coding processing to described the first sub-block, obtains first prediction residual corresponding with described the first sub-block;

The 6th coded prediction unit, for the inter prediction intraframe coding result of applying described the first sub-block, described a plurality of second and described a plurality of the second sub-blocks are carried out predicting between pixel intraframe coding processing, obtain and described a plurality of second and described a plurality of the second sub-blocks corresponding the second prediction residual respectively;

Described transcoding, coding transform module, comprising:

The 3rd transcoding, coding transform unit, for carrying out the directional transform based on described predictive mode to described the first prediction residual; Or described the first prediction residual and described the second prediction residual are carried out to the directional transform based on described predictive mode;

The 4th transcoding, coding transform unit, for carrying out the conversion irrelevant with directivity to described the second prediction residual.

17. image code processing devices according to claim 13, is characterized in that, described coding is divided module, comprising:

The 3rd coded sample unit, for described macro block is carried out to down-sampling division, obtains first and a plurality of second;

The 4th coded sample unit, divides processing for carrying out spatial domain to described first, obtains a plurality of the first sub-blocks;

Described coded prediction module, comprising:

The 7th coded prediction unit, for adopting predictive mode to carry out respectively inter prediction intraframe coding processing to described a plurality of the first sub-blocks, obtains first prediction residual corresponding with described a plurality of the first sub-blocks difference;

The 8th coded prediction unit, for the inter prediction intraframe coding result of applying described a plurality of the first sub-blocks, described a plurality of second is carried out predicting intraframe coding processing between pixel, obtain and described a plurality of second the second prediction residual that difference is corresponding;

Described transcoding, coding transform module, specifically for described the first prediction residual is carried out to the directional transform based on described predictive mode, is carried out the conversion irrelevant with directivity to described the second prediction residual; Or described the first prediction residual and described the second prediction residual are carried out to the directional transform based on described predictive mode.

18. 1 kinds of image decoding processing unit, is characterized in that, comprising:

Decoding synthesis module, synthesizes for carrying out up-sampling to described a plurality of, obtains decoded macro block;

Described decoding prediction module, comprising:

The first decoding predicting unit, for adopting described predictive mode to carry out the processing of inter prediction intraframe decoder to the first prediction residual, obtains in described a plurality of the inter prediction intraframe decoder result of first;

The second decoding predicting unit, carries out predicting between pixel intraframe decoder processing for the inter prediction intraframe decoder result of applying described first to the second prediction residual, obtains between the pixel of all the other pieces in described a plurality of decoding process result in predictive frame.

19. image decoding processing unit according to claim 18, is characterized in that, described decoding conversion module, comprising:

The first decoding converter unit, for to carrying out the directivity inverse transformation based on predictive mode with described a plurality of first corresponding the first prediction residual coefficient, obtains the first prediction residual;

The second decoding converter unit, for the second prediction residual coefficient corresponding with described a plurality of all the other pieces carried out to the inverse transformation irrelevant with directivity, obtains the second prediction residual;

Described decoding synthesis module is synthetic specifically for decoding process result in predictive frame between described inter prediction intraframe decoder result and described pixel being carried out to up-sampling, obtains decoded macro block.

20. image decoding processing unit according to claim 18, it is characterized in that, described decoding conversion module specifically for to first corresponding the first prediction residual coefficient in described a plurality of and with described a plurality of in the second prediction residual coefficient corresponding to all the other pieces carry out the directivity inverse transformation based on same prediction mode, obtain respectively the first prediction residual and the second prediction residual;

21. image decoding processing unit according to claim 18, is characterized in that, described decoding conversion module, comprising:

The 3rd decoding converter unit, for the first prediction residual coefficient corresponding with described a plurality of the first sub-blocks carried out to the directivity inverse transformation based on predictive mode, obtains the first prediction residual;

The 4th decoding converter unit, for to described a plurality of a plurality of second and a plurality of the second sub-blocks respectively corresponding the second prediction residual coefficient carry out the inverse transformation irrelevant with directivity, obtain the second prediction residual;

Described decoding prediction module, comprising:

The 5th decoding predicting unit, for adopting described predictive mode to carry out the processing of inter prediction intraframe decoder to described the first prediction residual, obtains the inter prediction intraframe decoder result of described the first sub-block;

The 6th decoding predicting unit, for the inter prediction intraframe decoder result of applying described the first sub-block, described the second prediction residual is carried out predicting between pixel intraframe decoder processing, obtain between the pixel of a plurality of second and a plurality of the second sub-blocks in described a plurality of decoding process result in predictive frame;

Described decoding synthesis module, comprising:

The first decoding sampling unit, carries out for decoding process result in predictive frame between the pixel of the inter prediction intraframe decoder result to described the first sub-block and described a plurality of the second sub-blocks that the second up-sampling is synthetic obtains described first;

The second decoding sampling unit, synthetic for decoding process result in predictive frame between described first and described a plurality of second pixel being carried out to the first up-sampling, obtain decoded macro block.

22. image decoding processing unit according to claim 18, it is characterized in that, described decoding conversion module specifically for the first prediction residual coefficient to corresponding with the first sub-block in described a plurality of and with described a plurality of in a plurality of second and the second prediction residual coefficients corresponding to a plurality of the second sub-blocks carry out the directivity inverse transformation based on same prediction mode, obtain respectively the first prediction residual and the second prediction residual;

Described decoding prediction module, comprising:

The 7th decoding predicting unit, for adopting described predictive mode to carry out the processing of inter prediction intraframe decoder to described the first prediction residual, obtains the inter prediction intraframe decoder result of described the first sub-block;

The 8th decoding predicting unit, for the inter prediction intraframe decoder result of applying described the first sub-block, described the second prediction residual is carried out predicting between pixel intraframe decoder processing, obtain between the pixel of a plurality of second and a plurality of the second sub-blocks in described a plurality of decoding process result in predictive frame;

Described decoding synthesis module, comprising:

The 3rd decoding sampling unit, carries out for decoding process result in predictive frame between the pixel of the inter prediction intraframe decoder result to described the first sub-block and described a plurality of the second sub-blocks that the second up-sampling is synthetic obtains described first;

The 4th decoding sampling unit, synthetic for decoding process result in predictive frame between described first and described a plurality of second pixel being carried out to the first up-sampling, obtain decoded macro block.

23. image decoding processing unit according to claim 18, is characterized in that, described decoding conversion module, comprising:

The 5th decoding converter unit, for the first prediction residual coefficient corresponding with described a plurality of a plurality of the first sub-blocks carried out respectively to the directivity inverse transformation based on predictive mode, obtains the first prediction residual;

The 6th decoding converter unit, for the second prediction residual coefficient corresponding with described a plurality of a plurality of second carried out to the inverse transformation irrelevant with directivity, obtains the second prediction residual;

Described decoding prediction module, comprising:

The 9th decoding predicting unit, for adopting described predictive mode to carry out the processing of inter prediction intraframe decoder to described the first prediction residual, obtains the inter prediction intraframe decoder result of described a plurality of the first sub-blocks;

The tenth decoding predicting unit, for the inter prediction intraframe decoder result of applying described a plurality of the first sub-blocks, described the second prediction residual is carried out predicting between pixel intraframe decoder processing, obtain in described a plurality of between the pixel of a plurality of second decoding process result in predictive frame;

Described decoding synthesis module, comprising:

The 6th decoding sampling unit, carries out the synthetic processing in spatial domain for the inter prediction intraframe decoder result to described a plurality of the first sub-blocks, obtains described first;

The 7th decoding sampling unit, synthetic for decoding process result in predictive frame between described first and described a plurality of second pixel being carried out to up-sampling, obtain decoded macro block.