CN102780885B - Rate distortion optimization method - Google Patents

Abstract

The invention relates to a rate distortion optimization method, which comprises the following steps: obtaining a block layer Lagrange parameter of each basic coding unit in a current frame and a frame layer Lagrange parameter of the current frame; acquiring a final Lagrange parameter of each basic coding unit in the current frame according to the Lagrange parameter of the block layer and the Lagrange parameter of the frame layer; according to the final Lagrange parameter, rate distortion optimization coding is carried out on each basic coding unit in the current frame to obtain a code rate value and a distortion value of each basic coding unit in the current frame, and then the code rate value and the distortion value of the current frame are obtained; and establishing a frame layer rate-distortion model according to the code rate values and the distortion values of at least two basic coding units in the current frame, and estimating the frame layer Lagrange parameters of the next frame according to the frame layer rate-distortion model and the code rate values and the distortion values of the current frame. The rate distortion optimization method realizes the optimal rate distortion optimization of video contents by integrating the rate distortion optimization of a frame layer and the rate distortion optimization of a basic coding unit layer.

Description

A kind of Rate-distortion optimization method

Technical field

The present invention relates to image/video codec domain, more particularly, relate to a kind of Rate-distortion optimization method.

Background technology

With international standard H264 for reference, when basic coding unit is encoded, different coding modes can be selected.The selection of coding mode comprises infra-frame prediction mode or inter prediction way choice, also partitioning scheme (the routine INTRA-4x4 to basic coding unit can be comprised, INTRA-8x8, INTRA-16x16, SKIP, DIRECT, INTER-16x16, INTER-16x8, INTER-8x16, INTER-8x8, INTER-8x8 can be further divided into INTER-8x8, INTER-8x4, INTER-4x8, INTER-4x4) selection, can also comprise prediction block position (routine Intra_4x4_Vertical, Intra_4x4_Horizontal, Intra_4x4_Diagonal_Down_Left, Intra_4x4_Diagonal_Down_Right, Intra4x4_Vertical_Right, Intra_4x4_Hori_zontal_Down, Intra_4x4_Vertical_Left, Intra_4x4_Horizontal_Up, Intra_4x4_DC) selection.The determination of coding mode is that percent of pass aberration optimizing realizes, and wherein rate-distortion optimization is the process of minimizing to following cost function J,

J(s,c,mode|QP)＝D(s,c,mode|QP)+λ _modeR(s,c,mode|QP)，

Wherein D is distortion value, R is code check value, s and c represents former figure and the corresponding basic coding unit building image again by encoding and decoding process respectively, mode represents the coding mode selected of basic coding unit, QP is quantization parameter, and λ mode is used to the LaGrange parameter of compromise distortion value and code check value.

Rate-distortion optimization is under the condition that quantization parameter QP is fixed, determines the mode that above-mentioned cost function J can be made minimum.In H264 standard, λ mode determines namely have by quantization parameter QP

λ _mode＝0.85×2 ^(QP-12)/3，

Above-mentioned rate-distortion optimization mode is a statistical approximation result, lacks the adaptive ability to video content, does not consider the continuity of interframe content, to the rate distortion relation between successive frame without any restriction.Therefore likely cause the rate distortion between successive frame uneven, cause the distortion between successive frame or code check to rise and fall excessive.

Therefore, be necessary to provide a kind of Rate-distortion optimization method, to solve the problem existing for prior art.

Summary of the invention

The technical problem to be solved in the present invention is, the defect of adaptive optimization can not be carried out to video content for Rate-distortion optimization method of the prior art, the rate-distortion optimization of a kind of rate-distortion optimization by comprehensive frame-layer and basic coding elementary layer is provided, realizes the Rate-distortion optimization method of the optimal ratio aberration optimizing to video content.

The technical solution adopted for the present invention to solve the technical problems is: the present invention relates to a kind of Rate-distortion optimization method, it comprises step:

The block layer LaGrange parameter of each basic coding unit and the frame-layer LaGrange parameter of present frame in A, acquisition present frame;

B, according to the block layer LaGrange parameter of described basic coding unit and the frame-layer LaGrange parameter of described present frame, obtain the final LaGrange parameter of each basic coding unit in described present frame;

C, according to described final LaGrange parameter, rate-distortion optimal coding is carried out to basic coding unit each in described present frame, obtain code check value that in described present frame, each basic coding unit is corresponding and distortion value, and then obtain code check value corresponding to present frame and distortion value;

D, according to code check value corresponding at least two basic coding unit in described present frame and distortion value, set up frame-layer rate-distortion model, according to described frame-layer rate-distortion model and code check value corresponding to described present frame and distortion value, the frame-layer LaGrange parameter of estimation next frame;

Wherein said step B is specially: obtain described final LaGrange parameter by following formula,

λ _mode,n＝w ₁λ _B,n+w ₂λ _F,0+w ₃，

Wherein λ _{mode, n}be the final LaGrange parameter of the n-th basic coding unit, λ _{f, 0}for the frame-layer LaGrange parameter of present frame, λ _b,nbe the block layer LaGrange parameter of the n-th basic coding unit, w ₁, w ₂, w ₃for corresponding weight coefficient, described weight coefficient w ₁, w ₂, w ₃obtain according to the coding result of described present frame and the coding result of nearest M frame, M is positive integer.

In Rate-distortion optimization method of the present invention, obtain described weight coefficient w by following steps ₁, w ₂, w ₃:

Store the frame-layer LaGrange parameter λ of described present frame _{f, 0}, nearest M frame frame-layer LaGrange parameter λ _f,mand intermediate parameters w _{23, m}, wherein w _{23, m}obtained by following formula:

$w_{23,m}=\frac{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n^2\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{y}_n-\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\left(x_n{y}_n\right)\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n}{N\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n^2-{\left(\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n\right)}^2},$

Wherein x _nbe the block layer LaGrange parameter of the n-th basic coding unit of m frame, y _nbe the final LaGrange parameter of the n-th basic coding unit of m frame:

x _n＝λ _B,n，

y _n=λ _mode,n，

According to the block layer LaGrange parameter x of the n-th basic coding unit of described m frame _nwith the final LaGrange parameter y of the n-th basic coding unit of described m frame _n, calculate w ₁:

$w_1=\frac{N\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\left(x_n{y}_n\right)-\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{y}_n}{N\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n^2-{\left(\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n\right)}^2},$

According to the frame-layer LaGrange parameter λ of present frame _{f, 0}, nearest M frame frame-layer LaGrange parameter λ _f,mwith described intermediate parameters w _{23, m}, calculate w ₂and w ₃,

u _m＝λ _F,m，

v _m＝w _23,m，

$w_2=\frac{M\underset{m=0}{\overset{M}{\mathrm{\Σ}}}\left(u_m{v}_m\right)-\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{v}_m}{M\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m^2-{\left(\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m\right)}^2},$

$w_3=\frac{\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m^2\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{v}_m-\underset{n=0}{\overset{N}{\mathrm{\Σ}}}\left(u_m{v}_m\right)\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m}{M\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m^2-{\left(\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m\right)}^2},$

Wherein, m is the integer being more than or equal to 0, being less than or equal to M, and N is the number of the described basic coding unit selected.

In Rate-distortion optimization method of the present invention, the block layer LaGrange parameter of described each basic coding unit is:

λ _B＝0.85×2 ^(QP-12)/3，

Wherein λ _bfor the block layer LaGrange parameter of described each basic coding unit, QP is the quantization parameter of described basic coding unit.

In Rate-distortion optimization method of the present invention, described step C comprises: the minimum code cost being obtained the rate-distortion optimization of the described basic coding unit of described present frame by following formula,

J(s,c,mode|QP)＝min[D(s,c,mode|QP)+λ _modeR(s,c,mode|QP)]，

Wherein J is minimum code cost, D is distortion value, R is code check value, s represents the basic coding unit of former figure, c represents the corresponding basic coding unit building image again by encoding and decoding process, mode represents the coding mode selected of described basic coding unit, and QP is the quantization parameter of described basic coding unit, λ _modefor corresponding final LaGrange parameter;

According to described minimum code cost, obtain corresponding forced coding pattern, rate-distortion optimal coding is carried out to the described basic coding unit of described present frame.

In Rate-distortion optimization method of the present invention, described step D comprises: according to code check value and the distortion value of at least two described basic coding unit of described present frame, adopts following formula to set up described frame-layer rate-distortion model,

R＝aln(b/D)，

Wherein R is code check value, and D is corresponding distortion value, and a, b are model parameter.

In Rate-distortion optimization method of the present invention, when code check value and the distortion value of the described basic coding unit of use two, when setting up described frame-layer rate-distortion model, described model parameter is obtained by following formula,

$a=\frac{R_1-R_2}{\ln D_2-\ln D_1},$

$b=e^{\frac{R_1\ln D_2-R_2\ln D_1}{R_1-R_2}},$

Wherein R ₁and R ₂for the code check value of described two basic coding unit, D ₁and D ₂for corresponding distortion value.

In Rate-distortion optimization method of the present invention, when using code check value and the distortion value of N number of described basic coding unit, when setting up described frame-layer rate-distortion model, described model parameter is obtained by following formula,

a＝-p，

b＝e ^-q/p，

$p=\frac{N\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\left(x_n{y}_n\right)-\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{y}_n}{N\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n^2-{\left(\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n\right)}^2},$

$q=\frac{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n^2\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{y}_n-\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\left(x_n{y}_n\right)\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n}{N\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n^2-{\left(\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n\right)}^2},$

x _n＝lnD _n，y _n=R _n，

N＞2，

Wherein R _nfor the code check value of described n-th basic coding unit, D _nfor corresponding distortion value.

In Rate-distortion optimization method of the present invention, described step D also comprises: by the frame-layer LaGrange parameter of following formula estimation next frame,

λ _F＝-(D(A ₁)-D(A ₂))/(R(A ₁)-R(A ₂))，

Wherein R (A ₁) be the code check value of the first intersection point, R (A ₂) be the code check value of the second intersection point, D (A ₁) be the distortion value of the first intersection point, D (A ₂) be the distortion value of the second intersection point, described first intersection point is R=R in R ~ D plane _awith described rate-distortion model determine the intersection point of curve, described second intersection point is D=D _awith described rate-distortion model determine the intersection point of curve, R _afor frame-layer code check value, D _afor frame-layer distortion value.

In Rate-distortion optimization method of the present invention, described frame-layer code check value R _afor the code check value sum of basic coding unit all in described frame, described frame-layer distortion value D _aby asking mean square error to obtain to the former figure of described frame and the image of building again of described frame.

Implement Rate-distortion optimization method of the present invention, there is following beneficial effect: by the rate-distortion optimization of comprehensive frame-layer and the rate-distortion optimization of basic coding elementary layer, realize the optimal ratio aberration optimizing to video content.The Rate-distortion optimization method avoiding prior art can not carry out the technical problem of adaptive optimization to video content.

Accompanying drawing explanation

Below in conjunction with drawings and Examples, the invention will be further described, in accompanying drawing:

Fig. 1 is the flow chart of the first preferred embodiment of Rate-distortion optimization method of the present invention;

Fig. 2 is the idiographic flow block diagram of the first preferred embodiment of Rate-distortion optimization method of the present invention;

Fig. 3 is the schematic diagram of the frame-layer LaGrange parameter obtaining next frame in Rate-distortion optimization method of the present invention.

Embodiment

In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.

Please refer to Fig. 1, Fig. 1 is the flow chart of the first preferred embodiment of Rate-distortion optimization method of the present invention.This Rate-distortion optimization method starts from:

Step 101, obtains the block layer LaGrange parameter of each basic coding unit and the frame-layer LaGrange parameter of present frame in present frame;

Step 102, according to the block layer LaGrange parameter of described basic coding unit and the frame-layer LaGrange parameter of described present frame, obtains the final LaGrange parameter of each basic coding unit in described present frame;

Step 103, according to described final LaGrange parameter, rate-distortion optimal coding is carried out to described basic coding unit each in described present frame, obtains code check value corresponding to each basic coding unit and distortion value, and then obtain code check value corresponding to present frame and distortion value;

Step 104, the code check value corresponding according at least two basic coding unit in present frame and distortion value, set up frame-layer rate-distortion model, the code check value corresponding according to described present frame and distortion value, and the frame-layer rate-distortion model of present frame, the frame-layer LaGrange parameter of estimation next frame, and weight coefficient used when calculating the final LaGrange parameter of next frame.

Please refer to Fig. 1 and Fig. 2, Fig. 2 is the idiographic flow block diagram of the first preferred embodiment of Rate-distortion optimization method of the present invention.Below by the specific implementation process of Fig. 2 detailed description Rate-distortion optimization method of the present invention.

First, before step 101, input an absolute coding part of video to be encoded, such as GOP(picture group, Group of Pictures) in a frame to code device, wherein in GOP, each frame can have different coded systems, and according to existing video encoding standard, coded system can be divided into I frame mode, P frame mode and B frame mode.Rate-distortion optimization method of the present invention can be applicable to I frame, P frame and B frame respectively, in order to the simplicity illustrated no longer distinguishes I frame, P frame and B frame here, but uses unified mode to be described.

After image frame is input to code device, by basic coding unit acquiring unit, this image frame is divided into basic coding unit, frame can be divided into the macro block of 16x16 pixel size simply, using each macro block as a basic coding unit.Here the dividing method adopting other is not got rid of, as adopted band as basic coding unit.Then all coding modes of selecting of basic coding unit are selected one by one, coding mode can be selected for each, complete corresponding Code And Decode process, obtain basic coding unit corresponding build image again, the processing method of Code And Decode can be carried out according to various video encoding standard.Such as, coded treatment can perform following steps successively: determine prediction block, residual computations, dct transform (discrete cosine transform, Discrete Cosine Transform), quantification, entropy code.Decoding process is then the reverse operating to above steps.The concrete grammar of Rate-distortion optimization method of the present invention to this Code And Decode process does not limit.

Come step 101 subsequently, in a step 101, obtain the block layer LaGrange parameter λ of each basic coding unit in present frame _band the frame-layer LaGrange parameter λ of present frame _f.Wherein the block layer LaGrange parameter of basic coding unit is:

λ _B＝0.85×2 ^(QP-12)/3，

Wherein λ _bfor the block layer LaGrange parameter of described basic coding unit, QP is the quantization parameter of described basic coding unit.It is pointed out that each basic coding unit can have different λ _bwith QP value, but in order to interest of clarity, use λ with not making any distinction between here _bblock layer LaGrange parameter and the quantization parameter of a basic coding unit is represented with QP.The frame-layer LaGrange parameter λ of present frame _fdetermine by the coding result of previous frame, concrete grammar hereafter elaborates.If present frame is the initial frame of a coded system (i.e. I frame, P frame or B frame mode), then desirable λ _fbe a designation number, such as 0.

Come step 102 subsequently, in a step 102, according to the block layer LaGrange parameter λ of basic coding unit each in present frame _{b, n}with the frame-layer LaGrange parameter λ of present frame _f, obtain the final LaGrange parameter λ of each described basic coding unit _{mode, n}.The final LaGrange parameter of each basic coding unit is obtained specifically by following formula,

λ _mode,n＝w ₁λ _B,n+w ₂λ _F,0+w ₃，

Wherein λ _{mode, n}be the final LaGrange parameter of the n-th basic coding unit, λ _fthe frame-layer LaGrange parameter of present frame, λ _b,nbe the block layer LaGrange parameter of the n-th basic coding unit, w ₁, w ₂, w ₃for corresponding weight coefficient, described weight coefficient w ₁, w ₂, w ₃obtain according to the coding result of present frame and the coding result of nearest M frame, M is positive integer, and concrete grammar elaborates below.If present frame is initial frame, then can be taken as designation number, such as w ₁be 1, w ₂be 0, w ₃be 0 etc.

Come step 103 subsequently, in step 103, according to each basic coding unit final LaGrange parameter separately, rate-distortion optimal coding is carried out to basic coding unit each in present frame, obtain corresponding code check value and distortion value, and obtain code check value corresponding to present frame and distortion value further, set up the frame-layer rate-distortion model of present frame simultaneously.Specifically comprise:

The minimum code cost of the rate-distortion optimization of a basic coding unit in described present frame is obtained by following formula,

J(s,c,mode|QP)＝min[D(s,c,mode|QP)+λ _modeR(s,c,mode|QP)]，

Wherein J is minimum code cost, D is distortion value, R is code check value, s represents the basic coding unit of former figure, c represents the corresponding basic coding unit building image again by encoding and decoding process, mode represents the coding mode selected of described basic coding unit, and QP is the quantization parameter of described basic coding unit, λ _modefor corresponding final LaGrange parameter;

Found the minimum code cost of rate-distortion optimization by above formula, the forced coding pattern of each basic coding unit can be found like this from optional coding mode, then adopt this forced coding pattern to encode to this basic coding unit.When all basic coding unit in a frame all complete above-mentioned coding, then the coding of this frame completes.

Meanwhile, adopt above-mentioned forced coding pattern to basic coding cell encoding, code check value corresponding to this basic coding unit and distortion value can be obtained.

The distortion value of above-mentioned basic coding unit by the MSE(mean square error between the corresponding basic coding unit that calculates former figure and build image again, Mean Squared Error) obtain.

Subsequently, according to the above-mentioned coding to basic coding unit each in present frame, code check value corresponding to present frame and distortion value can be obtained.Wherein, code check value is the code check value sum of all basic coding unit in present frame, and distortion value is by the former figure of whole frame with build image again and ask MSE to obtain.

Come step 104 subsequently, at step 104, according to code check value R and the distortion value D of at least two basic coding unit in present frame, adopt following formula to set up the frame-layer rate-distortion model describing frame-layer rate distortion relation,

R＝aln(b/D)，

Wherein R represents code check value, and D represents distortion value.A, b are model parameter.

Be specially, when the code check value of use two basic coding unit and distortion value set up frame-layer rate-distortion model, available following formula obtains the model parameter of frame-layer rate-distortion model

$a=\frac{R_1-R_2}{\ln D_2-\ln D_1},$

$b=e^{\frac{R_1\ln D_2-R_2\ln D_1}{R_1-R_2}},$

Wherein R ₁and R ₂for the code check value that described two basic coding unit are corresponding, D ₁and D ₂for the distortion value that described two basic coding unit are corresponding.

When the code check value and distortion value that use N number of basic coding unit set up frame-layer rate-distortion model, available following formula obtains the model parameter of frame-layer rate-distortion model

a＝-p，

b＝e ^-q/p，

$p=\frac{N\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\left(x_n{y}_n\right)-\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{y}_n}{N\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n^2-{\left(\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n\right)}^2},$

$q=\frac{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n^2\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{y}_n-\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\left(x_n{y}_n\right)\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n}{N\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n^2-{\left(\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n\right)}^2},$

x _n＝lnD _n，y _n=R _n，

N＞2，

Wherein R _nbe the code check value that n-th (n=1 ~ N) individual basic coding unit is corresponding, D _nfor the distortion value that this basic coding unit is corresponding.

According to the frame-layer rate-distortion model of present frame, and code check value corresponding to present frame and distortion value, the frame-layer LaGrange parameter of estimation next frame and weight coefficient w corresponding to next frame ₁, w ₂, w ₃.

As shown in Figure 3, Fig. 3 is the schematic diagram of the frame-layer LaGrange parameter obtaining next frame in Rate-distortion optimization method of the present invention.Curve in Fig. 3 is the determined curve of described frame-layer rate-distortion model, and wherein the coordinate of A point is the code check value R that present frame is corresponding _awith distortion value D _acoordinate in the coordinate system, as previously mentioned, code check value R _afor the code check value sum of basic coding unit all in present frame, distortion value D _aby to the former figure of whole frame with build image again and ask mean square error to obtain.Supposed that the longitudinal direction of A point and horizontal straight line were A1 point and A2 point with the intersection point of curve respectively, wherein the coordinate of A1 point is (R (A ₁), D (A ₁)), the coordinate of A2 point is (R (A ₂), D (A ₂)), the straight slope that A1 point and A2 point are formed just can be used as the frame-layer LaGrange parameter λ of next frame _f.Here also can to the nearest point of A point on calculated curve, namely curve can be used as the frame-layer LaGrange parameter λ of next frame at the slope of this point _f.

Particularly, by the frame-layer LaGrange parameter of following formula estimation next frame,

λ _F＝-(D(A ₁)-D(A ₂))/(R(A ₁)-R(A ₂))，

Wherein R (A ₁) be the code check value of the first intersection point, R (A ₂) be the code check value of the second intersection point, D (A ₁) be the distortion value of the first intersection point, D (A ₂) be the distortion value of the second intersection point, the first intersection point is R=R in R ~ D plane _awith the frame-layer rate-distortion model of above-mentioned present frame determine the intersection point of curve, the second intersection point is D=D _awith above-mentioned rate-distortion model determine the intersection point of curve, R _afor the code check value that above-mentioned present frame is corresponding, D _afor the distortion value that above-mentioned present frame is corresponding.

Weight coefficient w ₁, w ₂, w ₃computational methods as follows,

First by the λ of N number of basic coding unit in present frame _{b, n}and λ _{mode, n}obtain w ₁with intermediate variable w _{23, m},

$w_1=\frac{N\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\left(x_n{y}_n\right)-\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{y}_n}{N\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n^2-{\left(\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n\right)}^2},$

$w_{23,m}=\frac{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n^2\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{y}_n-\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\left(x_n{y}_n\right)\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n}{N\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n^2-{\left(\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n\right)}^2},$

x _n＝λ _B,n，

y _n=λ _mode,n，

Subsequently, then by the frame-layer LaGrange parameter λ of present frame _{f, 0}, nearest M frame frame-layer LaGrange parameter λ _f,mthe frame-layer LaGrange parameter w corresponding with every frame _{23, m}(m=0 ~ M), calculates w ₂and w ₃, specific as follows:

u _m＝λ _F，m，

v _m＝w _23,m，

$w_2=\frac{M\underset{m=0}{\overset{M}{\mathrm{\Σ}}}\left(u_m{v}_m\right)-\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{v}_m}{M\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m^2-{\left(\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m\right)}^2},$

$w_3=\frac{\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m^2\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{v}_m-\underset{n=0}{\overset{N}{\mathrm{\Σ}}}\left(u_m{v}_m\right)\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m}{M\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m^2-{\left(\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m\right)}^2},$

In above calculating formula, λ _{mode, n}for the final LaGrange parameter of the n-th basic coding unit of present frame, λ _f,mbe the frame-layer LaGrange parameter of m frame, λ _{f, 0}for the frame-layer LaGrange parameter of present frame, λ _b,nbe the block layer LaGrange parameter of the n-th basic coding unit in a frame, w ₁, w ₂, w ₃for corresponding weight coefficient, N is the basic coding unit number selected from a frame, and M is the quantity of the nearest frame selected.

The weight coefficient will used when can obtain by above formula the final LaGrange parameter calculating each basic coding unit of next frame like this.

In sum, the rate-distortion optimization of Rate-distortion optimization method of the present invention by comprehensive frame-layer and the rate-distortion optimization of basic coding elementary layer, ensure that the continuity of the rate-distortion characteristic between successive frame, realizes the optimal ratio aberration optimizing to video content.Simultaneity factor aberration optimizing can realize according to the type (i.e. I frame, P frame or B frame) of frame coding is independent separately; The Rate-distortion optimization method that Rate-distortion optimization method of the present invention well avoids prior art can not carry out the technical problem of adaptive optimization to video content.

The foregoing is only embodiments of the invention; not thereby the scope of the claims of the present invention is limited; every equivalent structure transformation utilizing specification of the present invention and accompanying drawing content to do, or be directly or indirectly used in other relevant technical fields, be all in like manner included in scope of patent protection of the present invention.

Claims (7)

1. a Rate-distortion optimization method, is characterized in that, comprises step:

The block layer LaGrange parameter of each basic coding unit and the frame-layer LaGrange parameter of present frame in A, acquisition present frame;

B, according to the block layer LaGrange parameter of described basic coding unit and the frame-layer LaGrange parameter of described present frame, obtain the final LaGrange parameter of each basic coding unit in described present frame;

C, according to described final LaGrange parameter, rate-distortion optimal coding is carried out to basic coding unit each in described present frame, obtain code check value that in described present frame, each basic coding unit is corresponding and distortion value, and then obtain code check value corresponding to present frame and distortion value;

D, according to code check value corresponding at least two basic coding unit in described present frame and distortion value, set up frame-layer rate-distortion model, according to described frame-layer rate-distortion model and code check value corresponding to described present frame and distortion value, the frame-layer LaGrange parameter of estimation next frame;

Wherein said step B is specially: obtain described final LaGrange parameter by following formula,

λ _mode,n＝w ₁λ _B,n+w ₂λ _F，0+w ₃，

Wherein λ _{mode, n}be the final LaGrange parameter of the n-th basic coding unit, λ _{f, 0}for the frame-layer LaGrange parameter of present frame, λ _b,nbe the block layer LaGrange parameter of the n-th basic coding unit, w ₁, w ₂, w ₃for corresponding weight coefficient, described weight coefficient w ₁, w ₂, w ₃obtain according to the coding result of described present frame and the coding result of nearest M frame, M is positive integer;

Wherein, described weight coefficient w is obtained by following steps ₁, w ₂, w ₃:

Store the frame-layer LaGrange parameter λ of described present frame _{f, 0}, nearest M frame frame-layer LaGrange parameter λ _f,mand intermediate parameters w _{23, m}, wherein w _{23, m}obtained by following formula:

$w_{23,m}=\frac{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n^2\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{y}_n-\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\left(x_n{y}_n\right)\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n}{N\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n^2-{\left(\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n\right)}^2},$

Wherein x _nbe the block layer LaGrange parameter of the n-th basic coding unit of m frame, y _nbe the final LaGrange parameter of the n-th basic coding unit of m frame:

x _n＝λ _B,n，

y _n＝λ _mode,n，

According to the block layer LaGrange parameter x of the n-th basic coding unit of described m frame _nwith the final LaGrange parameter y of the n-th basic coding unit of described m frame _n, calculate w ₁:

$w_1=\frac{N\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\left(x_n{y}_n\right)-\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{y}_n}{N\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n^2-{\left(\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n\right)}^2},$

According to the frame-layer LaGrange parameter λ of present frame _{f, 0}, nearest M frame frame-layer LaGrange parameter λ _f,mwith described intermediate parameters w _{23, m}, calculate w ₂and w ₃,

u _m＝λ _F,m，

v _m＝w _23,m，

$w_2=\frac{M\underset{m=0}{\overset{M}{\mathrm{\Σ}}}\left(u_m{v}_m\right)-\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{v}_m}{M\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m^2-{\left(\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m\right)}^2},$

$w_3=\frac{\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m^2\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{v}_m-\underset{m=0}{\overset{M}{\mathrm{\Σ}}}\left(u_m{v}_m\right)\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m}{M\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m^2-{\left(\underset{m=0}{\overset{M}{\mathrm{\Σ}}}{u}_m\right)}^2},$

Wherein, m is the integer being more than or equal to 0, being less than or equal to M, and N is the number of the described basic coding unit selected;

Wherein, described step D comprises: according to code check value and the distortion value of at least two described basic coding unit of described present frame, adopts following formula to set up described frame-layer rate-distortion model,

R＝aln(b/D)，

Wherein R is code check value, and D is corresponding distortion value, and a, b are model parameter.

2. Rate-distortion optimization method according to claim 1, is characterized in that, the block layer LaGrange parameter of described each basic coding unit is:

λ _B＝0.85×2 ^(QP-12)/3，

Wherein λ _bfor the block layer LaGrange parameter of described each basic coding unit, QP is the quantization parameter of described basic coding unit.

3. Rate-distortion optimization method according to claim 1, is characterized in that, described step C comprises: the minimum code cost being obtained the rate-distortion optimization of the described basic coding unit of described present frame by following formula,

J(s,c,mode|QP)＝min[D(s,c,mode|QP)+λ _modeR(s,c,mode|QP)]，

Wherein J is minimum code cost, D is distortion value, R is code check value, s represents the basic coding unit of former figure, c represents the corresponding basic coding unit building image again by encoding and decoding process, mode represents the coding mode selected of described basic coding unit, and QP is the quantization parameter of described basic coding unit, λ _modefor corresponding final LaGrange parameter;

According to described minimum code cost, obtain corresponding forced coding pattern, rate-distortion optimal coding is carried out to the described basic coding unit of described present frame.

4. Rate-distortion optimization method according to claim 1, is characterized in that, when code check value and the distortion value of the described basic coding unit of use two, when setting up described frame-layer rate-distortion model, described model parameter is obtained by following formula,

$a=\frac{R_1-R_2}{\ln D_2-\ln D_1},$

$b=e^{\frac{R_1\ln D_2-R_2\ln D_1}{R_1-R_2}},$

Wherein R ₁and R ₂for the code check value of described two basic coding unit, D ₁and D ₂for corresponding distortion value.

5. Rate-distortion optimization method according to claim 1, is characterized in that, when using code check value and the distortion value of N number of described basic coding unit, when setting up described frame-layer rate-distortion model, described model parameter is obtained by following formula,

a＝-p，

$b=e^{\frac{-q}{p}},$

$p=\frac{N\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\left(x_n{y}_n\right)-\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{y}_n}{N\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n^2-{\left(\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n\right)}^2},$

$q=\frac{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n^2\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{y}_n-\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\left(x_n{y}_n\right)\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n}{N\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n^2-{\left(\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{x}_n\right)}^2},$

x _n＝lnD _n，y _n＝R _n，

N>2，

Wherein R _nfor the code check value of described n-th basic coding unit, D _nfor corresponding distortion value.

6. Rate-distortion optimization method according to claim 1, is characterized in that, described step D also comprises: by the frame-layer LaGrange parameter of following formula estimation next frame,

λ _F＝-(D(A ₁)-D(A ₂))/(R(A ₁)-R(A ₂))，

Wherein R (A ₁) be the code check value of the first intersection point, R (A ₂) be the code check value of the second intersection point, D (A ₁) be the distortion value of the first intersection point, D (A ₂) be the distortion value of the second intersection point, described first intersection point is R=R in R ~ D plane _awith described rate-distortion model determine the intersection point of curve, described second intersection point is D=D _awith described rate-distortion model determine the intersection point of curve, R _afor frame-layer code check value, D _afor frame-layer distortion value.

7. Rate-distortion optimization method according to claim 6, is characterized in that, described frame-layer code check value R _afor the code check value sum of basic coding unit all in described frame, described frame-layer distortion value D _aby asking mean square error to obtain to the former figure of described frame and the image of building again of described frame.