CN102984540A - Video quality assessment method estimated on basis of macroblock domain distortion degree - Google Patents

Abstract

The invention discloses a video quality assessment method estimated on the basis of a macroblock domain distortion degree. The method comprises the following steps that (1), a video image is divided into a plurality of macroblocks; (2), a blocking effect distortion degree is calculated; (3), a blurring effect distortion degree is calculated; (4), a luminance contrast degree is calculated; (5), texture complexity is calculated; (6), an exercise intensity contrast degree is calculated; (7), exercise direction consistency is calculated; (8), visual perceptibility is calculated; and (9), a video quality assessment value is calculated. During the course of video quality assessment, a model of the video quality assessment method is simple; the objective quality of a video can be obtained only when the to-be-assessed video is provided; the method is very high in flexibility; in addition, relatively accurate assessment results of various video scenes can be obtained; and the method is preferable in universality.

Description

A kind of method for evaluating video quality of estimating based on the macro block territory distortion factor

Technical field

The invention belongs to the video quality evaluation technical field, be specifically related to a kind of method for evaluating video quality of estimating based on the macro block territory distortion factor.

Background technology

Along with the develop rapidly of the computer and network communication technology, people are day by day vigorous to the demand of obtaining multimedia messages.In recent years, every field is contained in the application relevant with video, such as video conference, video monitoring and mobile TV etc.In these were used, video information all needed through overcompression and transmission before arriving the recipient, and these processes tend to cause video quality loss.In order to obtain better subjective effect, be necessary video quality is estimated, adjust the parameter of encoder and transmission channel according to the result.The final receptor of video is human eyes, and eye-observation is considered to the method for accurate evaluates video quality.Yet, because the amount of information of video is very large, rely on the subjective method of manual observation that video quality is estimated manpower and the time that needs consumption a large amount of, be not suitable for large-scale practical application.Therefore, how to set up the video quality evaluation model according to human visual system (HVS) characteristic, automatically finished on this basis the quality evaluation of video by computer, become a very significant problem.

Video objective quality evaluation method (Video Objective Quality Assessment) refers to by the design mathematic model video be carried out intelligent analysis, and video is carried out the method for objectively evaluating of automatic scoring by the yardstick of setting.According to the degree of dependence to original video, video objective quality evaluation method can be divided into full reference type, partial reference type and without reference type three classes.Because full reference type and partial reference type evaluation method all need extra bandwidth to transmit original video and relevant information, its practical value is very limited.By contrast, reference-free quality evaluation method does not need to rely on any information relevant with original video, directly according to the information calculations video quality of video to be evaluated, has better flexibility and adaptability, and using value widely.Particularly in the Video Applications relevant with network multimedia, without the reference video evaluating objective quality at quality of server (Quality of Service, QoS) detection and terminal quality are experienced (Quality of Experience, QoE) play an important role above, according to the video quality evaluation feedback information, video server can dynamic regulating video coder parameters and transport channel parameters, to guarantee transmission stability, improves the receiving terminal video quality.In addition, can replace human eye without the reference video evaluating objective quality, compare equitably the video quality of different video codec output, for video receiver provides reference, make optimal selection.

Although existing method for evaluating video quality has been obtained certain effect, formed the model of some comparative maturities; Be Image quality assessment:From error visibility to structural similarity (IEEE Transactios on Image Processing based on method for evaluating video quality and the Wang Zhou of PSNR (Y-PSNR) at title such as tradition, a kind of method for evaluating video quality based on SSIM (structural similarity) has been proposed in the document 2004,13 (4)); But these methods are not considered the effect of HVS in video quality evaluation, have ignored the impact of video content features on video quality, and accuracy is still waiting to improve; And be difficult to be applicable to the video of different scenes, universality is not high.

Summary of the invention

For the existing above-mentioned technological deficiency of prior art, the invention provides a kind of method for evaluating video quality of estimating based on the macro block territory distortion factor, its index that obtains has higher accuracy, and can satisfy the needs of various different scene videos.

A kind of method for evaluating video quality of estimating based on the macro block territory distortion factor comprises the steps:

(1) the every two field picture of video to be evaluated is divided into several macro blocks;

(2) calculate the blocking effect distortion factor of macro block;

(3) calculate the blurring effect distortion factor of macro block;

(4) calculate the luminance contrast of macro block;

(5) calculate the Texture complication of macro block;

(6) calculate the exercise intensity contrast of macro block;

(7) direction of motion that calculates macro block is unanimously spent;

(8) according to described luminance contrast, Texture complication, the exercise intensity contrast is consistent with the direction of motion spends, and calculates the visually-perceptible degree of macro block;

(9) according to the described blocking effect distortion factor, the blurring effect distortion factor and visually-perceptible degree, calculate the quality evaluation value of the every two field picture of video to be evaluated; Quality evaluation value to all images of video to be evaluated is averaging, and the mean value that obtains is the quality evaluation value of video to be evaluated.

In the described step (2), the method for the blocking effect distortion factor of computing macro block is as follows:

A. calculate the horizontal block effect distortion factor of current macro according to following formula:

S _BH(i)＝|A _(16，i)-B _(1，i)|

S _IH(1，i)＝|A _(14，i)-A _(15，i)|

S _IH(2，i)＝|A _(15，i)-A _(16，i)|

S _IH(3，i)＝|B _(1，i)-B _(2，i)|

S _IH(4，i)＝|B _(2，i)-B _(3，i)|

$S_{\mathrm{IH}\_\mathrm{AVG}\left(i\right)}=\frac{S_{\mathrm{IH}(1,i)}+S_{\mathrm{IH}(2,i)}+S_{\mathrm{IH}(3,i)}+S_{\mathrm{IH}(4,i)}}{4}$

$J_{H\left(i\right)}=\left\{\begin{array}{cc}{S}_{\mathrm{BH}\left(i\right)}-S_{\mathrm{IH}\_\mathrm{AVG}\left(i\right)}& \mathrm{if}(S_{\mathrm{BH}\left(i\right)}>S_{\mathrm{IH}\_\mathrm{AVG}\left(i\right)})\\ 0& \mathrm{otherwise}\end{array}\right.$

$D_{\mathrm{BLOCK}\_H}=\frac{1}{16}\underset{i=1}{\overset{16}{\mathrm{\Σ}}}{J}_{H\left(i\right)}$

Wherein, D _{BLOCK_H}The horizontal block effect distortion factor for current macro, A (16, i) be the brightness value of current macro the 16th row i row pixel, A (15, i) be the brightness value of current macro the 15th row i row pixel, A (14, i) be the brightness value of current macro the 14th row i row pixel, B (1, i) be and the brightness value of following adjacent macro block the 1st row i row pixel of current macro that B (2, i) be brightness value with following adjacent macro block the 2nd row i row pixel of current macro, B (3, i) be and the brightness value of following adjacent macro block the 3rd row i row pixel of current macro that i is natural number and 1≤i≤16;

B. calculate the vertical blocks effect distortion factor of current macro according to following formula:

S _BV(i)＝|A _(i，16)-C _(i，1)|

S _IV(i，1)＝|A _(i，14)-A _(i，15)|

S _IV(i，2)＝|A _(i，15)-A _(i，16)|

S _IV(i，3)＝|C _(i，1)-C _(i，2)|

S _IV(i，4)＝|C _(i，2)-C _(i，3)|

$S_{\mathrm{IV}\_\mathrm{AVG}\left(i\right)}=\frac{S_{\mathrm{IV}(i,1)}+S_{\mathrm{IV}(i,2)}+S_{\mathrm{IV}(i,3)}+S_{\mathrm{IV}(i,4)}}{4}$

$J_{V\left(i\right)}=\left\{\begin{array}{cc}{S}_{\mathrm{BV}\left(i\right)}-S_{\mathrm{IV}\_\mathrm{AVG}\left(i\right)}& \mathrm{if}(S_{\mathrm{BV}\left(i\right)}>S_{\mathrm{IV}\_\mathrm{AVG}\left(i\right)})\\ 0& \mathrm{otherwise}\end{array}\right.$

$D_{\mathrm{BLOCK}\_V}=\frac{1}{16}\underset{i=1}{\overset{16}{\mathrm{\Σ}}}{J}_{V\left(i\right)}$

Wherein, D _{BLOCK_v}The vertical blocks effect distortion factor for current macro, A (i, 16) be the brightness value of capable the 16th row pixel of current macro i, A (i, 15) be the brightness value of capable the 15th row pixel of current macro i, A (i, 14) be the brightness value of capable the 14th row pixel of current macro i, C (i, 1) be the brightness value of the macro block i capable 1st row pixel adjacent with current macro the right, C (i, 2) be the brightness value of capable the 2nd row pixel of macro block i adjacent on the right of current macro, C (i, 3) is the brightness value of the macro block i capable 3rd row pixel adjacent with current macro the right;

C. the described horizontal block effect distortion factor and the vertical blocks effect distortion factor are averaging, obtain the blocking effect distortion factor of current macro.

In the described step (3), the method for the blurring effect distortion factor of computing macro block is as follows:

A. utilize the Sobel operator that macro block is carried out rim detection, determine edge pixel and gradient direction thereof in the macro block;

B. the gradient direction of edge pixel is ranged a kind of in the middle of { 0 °, 45 °, 90 °, 135 °, 180 °, 225 °, 270 °, 315 ° } these eight kinds of directions according to nearby principle, and as the approximate gradient direction of edge pixel;

C. according to the clear-cut margin degree of following formula edge calculation pixel:

$S=\frac{1}{4}\underset{l=1}{\overset{4}{\mathrm{\Σ}}}\frac{|I_{(\mathrm{\θ},l)}-I|}{l}$

Wherein: S is the clear-cut margin degree of edge pixel, and I is the brightness value of edge pixel, I (θ, l) be with edge pixel along the approximate gradient direction θ distance brightness value for the pixel of l, l is natural number and 1≤l≤4;

D. the clear-cut margin degree of all edge pixels in the macro block is averaging, obtains the blurring effect distortion factor of macro block.

In the described step (4), the method for the luminance contrast of computing macro block is: according to the luminance contrast of each pixel in the following formula computing macro block, get the maximum of all pixel intensity contrasts in the macro block as the luminance contrast of macro block;

$C_{\mathrm{Luma}\_P}(\mathrm{\θ},l)=\frac{|I_{(\mathrm{\θ},l)}-I|}{l^{\mathrm{\η}}}$

$C_{\mathrm{Luma}}=\underset{\mathrm{\θ}\∈M}{\overset{M}{\mathrm{\Σ}}}\underset{l=1}{\overset{L}{\mathrm{\Σ}}}{C}_{\mathrm{Luma}\_P}(l,N)$

Wherein, C _LumaLuminance contrast for current pixel, I is the brightness value of current pixel, I (θ, l) be along the brightness value of direction θ distance for the pixel of l with current pixel, l is natural number and 1≤l≤L, L is default ultimate range, and η is pixel distance decline coefficient, and M is the set of 0 °, 90 °, 180 ° and 270 ° these four kinds of directions.

In the described step (5), the method for the Texture complication of computing macro block is as follows:

A. utilize the Sobel operator that macro block is carried out rim detection, determine total number of edge pixel in the macro block and the gradient direction of each pixel;

B. the gradient direction of pixel is ranged a kind of in the middle of { 0 °, 45 °, 90 °, 135 °, 180 °, 225 °, 270 °, 315 ° } these eight kinds of directions according to nearby principle, and as the approximate gradient direction of pixel;

C. the approximate gradient direction with pixel is divided into four classes: 0 ° and 180 ° is classified as a class, 45 ° and 225 ° and is classified as a class, 90 ° and 270 ° and is classified as a class, 135 ° and 315 ° and is classified as a class;

D. according to the Texture complication of following formula computing macro block:

$T_{\mathrm{block}}=\left\{\begin{array}{cc}0.5& \mathrm{if}(k_{\mathrm{\θ}}=1)\\ 1& \mathrm{if}(k_{\mathrm{\θ}}=2)\\ (2-c_e)/2& \mathrm{if}(k_{\mathrm{\θ}}=3)\\ (1-c_e)/2& \mathrm{if}(k_{\mathrm{\θ}}=4)\end{array}\right.$

$c_e=\left\{\begin{array}{cc}1& \mathrm{if}(n_{\mathrm{edge}}>T_{\mathrm{edge}})\\ 0& \mathrm{otherwise}\end{array}\right.$

Wherein, T _BlockBe the Texture complication of macro block, k _θBe the species number of all pixel approximate gradient directions in the macro block, n _EdgeBe total number of edge pixel in the macro block, T _EdgeBe given edge pixel number threshold value.

In the described step (6), the method for the exercise intensity contrast of computing macro block is as follows:

A. calculate horizontal motion vector and the vertical motion vector of macro block by inter prediction; Described horizontal motion vector and vertical motion vector root mean square are obtained the exercise intensity of macro block;

B. centered by current macro, set up the reference windows that is formed by 7 * 7 macro blocks;

C. the exercise intensity difference distance of calculating each macro block in current macro and the reference windows according to following formula compares M _{I_diff}, and get wherein that maximum is M _{Id_max}:

$M_{I\_\mathrm{diff}}=\frac{|M_I-M_I\left(D\right)|}{d^{\mathrm{\δ}}}$

Wherein: M _IBe the exercise intensity of current macro, M _I(D) be the exercise intensity of macro block D, macro block D is the arbitrary macro block in the reference windows, and d is the distance of current macro and macro block D, and δ is that macro block is apart from the decline coefficient;

D. determine the exercise intensity contrast of current macro according to following relational expression:

$C_{\mathrm{Motion}\_\mathrm{block}}=\left\{\begin{array}{cc}\frac{M_{\mathrm{Id}\_\max }}{M_{I\_\max }}& \mathrm{if}(M_{I\_\max }\≠0)\\ 0& \mathrm{otherwise}\end{array}\right.$

Wherein: C _{Motion_block}Be the exercise intensity contrast of current macro, M _{I_max}Maximum for all macro block exercise intensities in the reference windows.

In the described step (7), the method for the consistent degree of the direction of motion of computing macro block is as follows:

A. calculate the horizontal motion vector mv of macro block by inter prediction _xWith vertical motion vector mv _yDescribed horizontal motion vector and vertical motion vector root mean square are obtained the exercise intensity of macro block, and according to formula θ _Mv=arctan (mv _y/ mv _x) calculate the direction of motion θ of macro block _Mv

B. 0 ° to 360 ° circumference equal dividing is become 12 sectors, make the direction of motion of macro block map to corresponding sector;

C. centered by current macro, set up the reference windows that is formed by 21 * 21 macro blocks;

D. unanimously spend according to the direction of motion of following formula calculating current macro:

$M_{\mathrm{Con}\_\mathrm{block}}=-\underset{j=1}{\overset{12}{\mathrm{\Σ}}}p\left(j\right)\log \left[p\left(j\right)\right]\frac{M_{I\_\mathrm{avg}}\left(j\right)}{M_{I\_\max }}$ $p\left(j\right)=\frac{k\left(j\right)}{{21}^2}$

Wherein: M _{Con_block}For the direction of motion of current macro is unanimously spent, k (j) is for belonging to the total number of macro block of j sector, M in the reference windows _{I_max}Be the maximum of all macro block exercise intensities in the reference windows, M _{I_avg}(j) for belonging to the mean value of all macro block exercise intensities of j sector in the reference windows, j is natural number and 1≤j≤12.

In the described step (8), according to the visually-perceptible degree of following formula computing macro block:

VP _S＝log(α ₁+C _{Luma_block})×(α ₂+T _block) ²

VP _T＝α ₃C _{Motion_block}+α ₄M _{con_block}

VM＝λ×VP _S×VP _T

Wherein: VM is the visually-perceptible degree of macro block, M _{Con_block}For the direction of motion of macro block is unanimously spent C _{Motion_block}Be the exercise intensity contrast of macro block, T _BlockBe the Texture complication of macro block, C _{Luma_block}Be the luminance contrast of macro block, λ, α ₁, α ₂, α ₃And α ₄Be given weight coefficient.

In the described step (9), calculate the quality evaluation value of the every two field picture of video to be evaluated according to following formula:

$Q=\frac{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}[Q_{\mathrm{MB}}\left(n\right)\×\mathrm{VM}\left(n\right)]}{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\mathrm{VM}\left(n\right)}$

Q _MB(n)＝γ ₁×D _{BLOCK_MB}(n)+γ ₂×D _{BLUR_MB}(n)

Wherein: Q is the quality evaluation value of the arbitrary two field picture of video to be evaluated, D _{BLOCK_MB}(n) be the blocking effect distortion factor of n macro block in the arbitrary two field picture of video to be evaluated, D _{BLUR_MB}(n) be the blurring effect distortion factor of n macro block in the arbitrary two field picture of video to be evaluated, VM (n) is the visually-perceptible degree of n macro block in the arbitrary two field picture of video to be evaluated, γ ₁And γ ₂Be given weight coefficient, n is natural number and 1≤n≤N, and N is total number of macro block in the arbitrary two field picture of video to be evaluated.

Useful technique effect of the present invention is as follows:

(1) the present invention only needs video to be evaluated just can obtain the video quality evaluation result, need not reference information, has good flexibility and adaptability.

(2) evaluation result accuracy of the present invention is higher, meets human eye to the subjective perception of video.

(3) the present invention can both obtain more accurately evaluation result to various video scene, has preferably universality.

Description of drawings

Fig. 1 is the schematic flow sheet of the inventive method.

Fig. 2 is the position view of macroblock boundaries pixel.

Fig. 3 is the classification schematic diagram of the direction of motion.

Embodiment

In order more specifically to describe the present invention, below in conjunction with the drawings and the specific embodiments technical scheme of the present invention is elaborated.

We at first are divided into several macro blocks with the every two field picture of video to be evaluated; As shown in Figure 1, a kind of method for evaluating video quality of estimating based on the macro block territory distortion factor comprises the steps:

(1) the computing block effect distortion factor.

Blocking effect generally appears in the compression algorithm based on discrete cosine transform (DCT), and coarse quantizing process also can aggravate blocking effect.Different macro blocks are subjected to the impact of DCT and quantification different, and loss of detail is also variant, therefore, often have significantly discontinuously in macroblock boundaries, form the blocking effect distortion.From the blocking effect distortion level of horizontal and vertical direction calculating macroblock edges, detailed process is as follows respectively for present embodiment:

A. calculate the horizontal block effect distortion factor of current macro according to following formula; As shown in Figure 2, the square of solid black lines above and below represents respectively current macro A and its following adjacent macro block B.

S _BH(i)＝|A _(16，i)-B _(1，i)|

S _IH(1，i)＝|A _(14，i)-A _(15，i)|

S _IH(2，i)＝|A _(15，i)-A _(16，i)|

S _IH(3，i)＝|B _(1，i)-B _(2，i)|

S _IH(4，i)＝|B _(2，i)-B _(3，i)|

$S_{\mathrm{IH}\_\mathrm{AVG}\left(i\right)}=\frac{S_{\mathrm{IH}(1,i)}+S_{\mathrm{IH}(2,i)}+S_{\mathrm{IH}(3,i)}+S_{\mathrm{IH}(4,i)}}{4}$

$J_{H\left(i\right)}=\left\{\begin{array}{cc}{S}_{\mathrm{BH}\left(i\right)}-S_{\mathrm{IH}\_\mathrm{AVG}\left(i\right)}& \mathrm{if}(S_{\mathrm{BH}\left(i\right)}>S_{\mathrm{IH}\_\mathrm{AVG}\left(i\right)})\\ 0& \mathrm{otherwise}\end{array}\right.$

$D_{\mathrm{BLOCK}\_H}=\frac{1}{16}\underset{i=1}{\overset{16}{\mathrm{\Σ}}}{J}_{H\left(i\right)}$

Wherein, D _{BLOCK_H}The horizontal block effect distortion factor for current macro, A (16, i) be the brightness value of current macro the 16th row i row pixel, A (15, i) be the brightness value of current macro the 15th row i row pixel, A (14, i) be the brightness value of current macro the 14th row i row pixel, B (1, i) be and the brightness value of following adjacent macro block the 1st row i row pixel of current macro that B (2, i) be brightness value with following adjacent macro block the 2nd row i row pixel of current macro, B (3, i) be and the brightness value of following adjacent macro block the 3rd row i row pixel of current macro that i is natural number and 1≤i≤16;

B. calculate the vertical blocks effect distortion factor of current macro according to following formula:

S _BV(i)＝|A _(i，16)-C _(i，1)|

S _IV(i，1)＝|A _(i，14)-A _(i，15)|

S _IV(i，2)＝|A _(i，15)-A _(i，16)|

S _IV(i，3)＝|C _(i，1)-C _(i，2)|

S _IV(i，4)＝|C _(i，2)-C _(i，3)|

$S_{\mathrm{IV}\_\mathrm{AVG}\left(i\right)}=\frac{S_{\mathrm{IV}(i,1)}+S_{\mathrm{IV}(i,2)}+S_{\mathrm{IV}(i,3)}+S_{\mathrm{IV}(i,4)}}{4}$

$J_{V\left(i\right)}=\left\{\begin{array}{cc}{S}_{\mathrm{BV}\left(i\right)}-S_{\mathrm{IV}\_\mathrm{AVG}\left(i\right)}& \mathrm{if}(S_{\mathrm{BV}\left(i\right)}>S_{\mathrm{IV}\_\mathrm{AVG}\left(i\right)})\\ 0& \mathrm{otherwise}\end{array}\right.$

$D_{\mathrm{BLOCK}\_V}=\frac{1}{16}\underset{i=1}{\overset{16}{\mathrm{\Σ}}}{J}_{V\left(i\right)}$

Wherein, D _{BLOCK_v}The vertical blocks effect distortion factor for current macro, A (i, 16) be the brightness value of capable the 16th row pixel of current macro i, A (i, 15) be the brightness value of capable the 15th row pixel of current macro i, A (i, 14) be the brightness value of capable the 14th row pixel of current macro i, C (i, 1) be the brightness value of the macro block i capable 1st row pixel adjacent with current macro the right, C (i, 2) be the brightness value of capable the 2nd row pixel of macro block i adjacent on the right of current macro, C (i, 3) is the brightness value of the macro block i capable 3rd row pixel adjacent with current macro the right;

C. the horizontal block effect distortion factor and the vertical blocks effect distortion factor are averaging, obtain the blocking effect distortion factor D of current macro _{BLOCK_MB}

Because quantizing process carries out take macro block as unit, adjacent macro block can adopt different quantization parameters.Therefore, the difference between the macroblock edges pixel is often larger than the difference between interior pixels, and along with the increase of difference, the discontinuous saltus step between the macroblock edges is just more obvious.The blocking effect distortion factor D of macro block _{BLOCK_MB}Be the mean value of the horizontal and vertical distortion factor; D _{BLOCK_MB}Larger, it is more serious to represent blocking effect.

(2) calculate the blurring effect distortion factor.

Blurring effect is a kind of common video distortion effect, and from subjective vision, blurring effect is mainly reflected near the details degeneration object edge in the video, and sharp keen degree descends.Pixel on the object edge if be subject to the impact of blurring effect, generally all can have maximum partial gradient value.Present embodiment is carried out rim detection with the Sobel operator to video, to each pixel that is positioned at object edge, detects the clear-cut margin degree S of its gradient direction, assesses the distortion level that video is subject to the blurring effect impact with this; Detailed process is as follows:

A. utilize the Sobel operator that macro block is carried out rim detection, determine edge pixel and gradient direction thereof in the macro block;

B. the gradient direction of edge pixel is ranged a kind of in the middle of { 0 °, 45 °, 90 °, 135 °, 180 °, 225 °, 270 °, 315 ° } these eight kinds of directions according to nearby principle, and as the approximate gradient direction of edge pixel;

C. according to the clear-cut margin degree of following formula edge calculation pixel:

$S=\frac{1}{4}\underset{l=1}{\overset{4}{\mathrm{\Σ}}}\frac{|I_{(\mathrm{\θ},l)}-I|}{l}$

Wherein: S is the clear-cut margin degree of edge pixel, and I is the brightness value of edge pixel, I (θ, l) be with edge pixel along the approximate gradient direction θ distance brightness value for the pixel of l, l is natural number and 1≤l≤4;

D. the clear-cut margin degree of all edge pixels in the macro block is averaging, obtains the blurring effect distortion factor D of macro block _{BLUR_MB}

Present embodiment is chosen four points at gradient direction, calculates S with their average gradual change speed, and S is less, and it is less to represent the edge sharpness, and blurring effect is more serious; At last, add up S value corresponding to each edge pixel point in the whole macro block, get its mean value as the blurring effect distortion factor D of this macro block _{BLUR_MB}, D _{BLUR_MB}Less, it is more serious to represent blurring effect.

(3) calculate luminance contrast.

The HVS characteristic research is shown, human when watching video, generally also insensitive to the absolute brightness of object in the video, and the relative size of the brightness of localized region and peripheral region is more responsive, therefore, luminance contrast can simply be described as the poor of brightness value between pixel.Research shows that simultaneously human eye increases along with the distance between pixel the susceptibility of luminance difference and reduces.In conjunction with above 2 points, present embodiment defines two luminance contrasts between pixel with pixel intensity difference and distance, and concrete grammar is as follows:

According to the luminance contrast of each pixel in the following formula computing macro block, get the maximum of all pixel intensity contrasts in the macro block as the luminance contrast C of macro block _{Luma_block}

$C_{\mathrm{Luma}\_P}(\mathrm{\θ},l)=\frac{|I_{(\mathrm{\θ},l)}-I|}{l^{\mathrm{\η}}}$

$C_{\mathrm{Luma}}=\underset{\mathrm{\θ}\∈M}{\overset{M}{\mathrm{\Σ}}}\underset{l=1}{\overset{L}{\mathrm{\Σ}}}{C}_{\mathrm{Luma}\_P}(l,N)$

Wherein, C _LumaLuminance contrast for current pixel, I is the brightness value of current pixel, I (θ, l) be along the brightness value of direction θ distance for the pixel of l with current pixel, l is natural number and 1≤l≤L, present embodiment is preset L=5, and η is pixel distance decline coefficient (this coefficient gets 2 in the present embodiment), and M is the set of 0 °, 90 °, 180 ° and 270 ° these four kinds of directions.

Luminance contrast has certain regionality, and some point of contrast maximum is the most responsive in the human eye localized region.Therefore, present embodiment is divided into the macro block of the non-overlapping copies of 16 * 16 pixels with whole frame video image, with the pixel C in each piece _LumaMaximum is as the luminance contrast C of this macro block _{Luma_block}C _{Luma_block}This regional luminance contrast of the larger expression of value higher, namely should the zone and the peripheral region between luminance difference larger, the easier attention that causes human eye.

(4) calculate Texture complication.

According to the different qualities of texture, video image can be divided into structural texture zone and random grain zone.The texture in structural texture zone is comparatively simple, and with the image relevance is lower on every side, and the texture in random grain zone is than horn of plenty, and the spatial contrast degree is low, with the image relevance is higher on every side.The HVS characteristic research shows, the easier attention that attracts the mankind of the image fault in the structural texture zone.

Present embodiment is at first carried out convolution with the Sobel operator to entire image, extracts the edge pixel point, calculates respectively the horizontal and vertical gradient (G of each pixel according to following formula _Hor, G _Ver), and then total number n of edge pixel in definite macro block _EdgeAnd the gradient direction θ (i, j) of each pixel;

G _hor(i，j)＝I(i，j)*S _hor

G _ver(i，j)＝I(i，j)*S _ver

$\mathrm{\θ}(i,j)=\arctan \frac{G_{\mathrm{ver}}(i,j)}{G_{\mathrm{hor}}(i,j)}$

Then, the gradient direction of pixel is ranged a kind of in the middle of { 0 °, 45 °, 90 °, 135 °, 180 °, 225 °, 270 °, 315 ° } these eight kinds of directions according to nearby principle, and as the approximate gradient direction of pixel;

The approximate gradient direction of pixel is divided into four classes: 0 ° and 180 ° is classified as a class, 45 ° and 225 ° and is classified as a class, 90 ° and 270 ° and is classified as a class, 135 ° and 315 ° and is classified as a class;

At last, according to the Texture complication of following formula computing macro block:

$T_{\mathrm{block}}=\left\{\begin{array}{cc}0.5& \mathrm{if}(k_{\mathrm{\θ}}=1)\\ 1& \mathrm{if}(k_{\mathrm{\θ}}=2)\\ (2-c_e)/2& \mathrm{if}(k_{\mathrm{\θ}}=3)\\ (1-c_e)/2& \mathrm{if}(k_{\mathrm{\θ}}=4)\end{array}\right.$

$c_e=\left\{\begin{array}{cc}1& \mathrm{if}(n_{\mathrm{edge}}>T_{\mathrm{edge}})\\ 0& \mathrm{otherwise}\end{array}\right.$

Wherein, T _BlockBe the Texture complication of macro block, k _θBe the species number of all pixel approximate gradient directions in the macro block, n _EdgeBe total number of edge pixel in the macro block, T _EdgeBe edge pixel number threshold value (this threshold value gets 16 in the present embodiment).

T _BlockSpan be [0,1], work as T _BlockTended to 0 o'clock, it is abundant to represent this zone-texture, belongs to the random grain zone, and its distortion effect is not easy to be discovered by human eye; Work as T _BlockTended to 1 o'clock, it is simple to represent this zone-texture, belongs to the random grain zone, and human eye is relatively more responsive to distortion effect wherein.

(5) calculate the exercise intensity contrast.

In the video quality evaluation field, movable information is widely used in reflection HVS to the perception of visual signal as one of most important primary vision information.Exercise intensity contrast and direction of motion consistency are the characteristic values that best embodies the video motion characteristic.HVS studies show that, human eye will be higher than the absolute movement of single body to the susceptibility of relative motion of object, and the exercise intensity contrast has just in time reflected and the movement velocity difference of different objects in the video reflected the whole perception of HVS to video.The process of the exercise intensity contrast of present embodiment computing macro block is as follows:

A. calculate horizontal motion vector and the vertical motion vector of macro block by inter prediction; Horizontal motion vector and vertical motion vector root mean square are obtained the exercise intensity of macro block;

B. centered by current macro, set up the reference windows that is formed by 7 * 7 macro blocks;

C. the exercise intensity difference distance of calculating each macro block in current macro and the reference windows according to following formula is than MI_diff, and gets wherein that maximum is M _{Id_max}:

$M_{I\_\mathrm{diff}}=\frac{|M_I-M_I\left(D\right)|}{d^{\mathrm{\δ}}}$

Wherein: M _IBe the exercise intensity of current macro, M _I(D) be the exercise intensity of macro block D, macro block D is the arbitrary macro block in the reference windows, and d is the distance of current macro and macro block D, and δ is that macro block is apart from decline coefficient (this coefficient gets 2 in the present embodiment);

Consider that the exercise intensity maximum in the regional area also can have influence on HVS to the perception of exercise intensity difference, determine at last the exercise intensity contrast of current macro according to following relational expression:

$C_{\mathrm{Motion}\_\mathrm{block}}=\left\{\begin{array}{cc}\frac{M_{\mathrm{Id}\_\max }}{M_{I\_\max }}& \mathrm{if}(M_{I\_\max }\≠0)\\ 0& \mathrm{otherwise}\end{array}\right.$

Wherein: C _{Motion_block}Be the exercise intensity contrast of current macro, M _{I_max}Maximum for all macro block exercise intensities in the reference windows.C _{Motion_block}Be worth greatlyr, the exercise intensity contrast that represents macro block is larger.

(6) calculating the direction of motion unanimously spends.

In video scene, the motor behavior of each object is diversified, and the motion consistency has reflected the similitude of these motor behaviors, wherein pays close attention to the easiest HVS that causes of the direction of motion again.In the present embodiment, by analyzing the probability distribution of the various directions of motion in the regional area, in conjunction with exercise intensity, represent direction of motion consistency.The method of the consistent degree of the direction of motion of present embodiment computing macro block is as follows:

A. calculate the horizontal motion vector mv of macro block by inter prediction _xWith vertical motion vector mv _yTo horizontal motion vector mv _xWith vertical motion vector mv _yRoot mean square obtains the exercise intensity of macro block, and according to formula θ _Mv=arctan (mv _y/ mv _x) calculate the direction of motion θ of macro block _Mv

B. 0 ° to 360 ° circumference equal dividing is become 12 sectors, make the direction of motion of macro block map to corresponding sector; As shown in Figure 3, the direction of motion roughly is divided into 12 kinds, be about to [0,360 °) be divided into 12 kinds of different direction Angle zones, adopt histogram method the direction of motion θ of each macro block _MvBe mapped to corresponding Angle zone;

C. centered by current macro, set up the reference windows that is formed by 21 * 21 macro blocks;

D. unanimously spend according to the direction of motion of following formula calculating current macro:

$M_{\mathrm{Con}\_\mathrm{block}}=-\underset{j=1}{\overset{12}{\mathrm{\Σ}}}p\left(j\right)\log \left[p\left(j\right)\right]\frac{M_{I\_\mathrm{avg}}\left(j\right)}{M_{I\_\max }}$ $p\left(j\right)=\frac{k\left(j\right)}{{21}^2}$

Wherein: M _{Con_block}For the direction of motion of current macro is unanimously spent, k (j) is for belonging to the total number of macro block of j sector, M in the reference windows _{I_max}Be the maximum of all macro block exercise intensities in the reference windows, M _{I_avg}(j) for belonging to the mean value of all macro block exercise intensities of j sector in the reference windows, j is natural number and 1≤j≤12; M _{Con_block}Be worth greatlyr, the motion consistency that represents macro block is higher.

(7) computation vision perceptibility.

According to luminance contrast, Texture complication, the exercise intensity contrast is consistent with the direction of motion spends, and calculates the visually-perceptible degree of every each macro block of two field picture of video to be evaluated according to following formula:

VP _S＝log(α ₁+C _{Luma_block})×(α ₂+T _Block) ²

VP _T＝α ₃C _{motion_block}+α ₄M _{con_block}

VM＝λ×VP _S×VP _T

Wherein: VM is the visually-perceptible degree of macro block, M _{Con_block}For the direction of motion of macro block is unanimously spent C _{Motion_block}Be the exercise intensity contrast of macro block, T _BlockBe the Texture complication of macro block, C _{Luma_block}Be the luminance contrast of macro block, λ, α ₁, α ₂, α ₃And α ₄Be weight coefficient, λ gets 1.2, α in the present embodiment ₁, α ₂And α ₄All get 1, α ₃Get 2.

(8) the quality evaluation value of calculating video.

According to the blocking effect distortion factor, the blurring effect distortion factor and visually-perceptible degree, calculate the quality evaluation value of the every two field picture of video to be evaluated according to following formula;

$Q=\frac{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}[Q_{\mathrm{MB}}\left(n\right)\×\mathrm{VM}\left(n\right)]}{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\mathrm{VM}\left(n\right)}$

Q _MB(n)＝γ ₁×D _{BLOCK_MB}(n)+γ ₂×D _{BLUR_MB}(n)

Wherein: Q is the quality evaluation value of the arbitrary two field picture of video to be evaluated, D _{BLOCK_MB}(n) be the blocking effect distortion factor of n macro block in the arbitrary two field picture of video to be evaluated, D _{BLUR_MB}(n) be the blurring effect distortion factor of n macro block in the arbitrary two field picture of video to be evaluated, VM (n) is the visually-perceptible degree of n macro block in the arbitrary two field picture of video to be evaluated, γ ₁And γ ₂Be weight coefficient; N is natural number and 1≤n≤N, and N is total number of macro block in the arbitrary two field picture of video to be evaluated.Q _MBThis piece distortion of larger representative is more serious, and video quality is poorer.Present embodiment is introduced the visual attention model based on visually-perceptible, adjusts the weight of video regional in quality evaluation system, γ in the present embodiment with this model ₁Get 1, γ ₂Get-0.5.

At last, the quality evaluation value of all images of video to be evaluated is averaging the mean value Q that obtains _AVGBe the quality evaluation value of video to be evaluated; Q _AVGSpan be [0,100], 0 to represent video quality best, 100 to represent video quality the poorest.

Below we come the effect of verification algorithm by 100 JM compressed videos, weigh respectively monotonicity and the accuracy of present embodiment with Spearman coefficient correlation and Pearson correlation coefficient, and compare based on PSNR with based on the SSIM method with tradition that (Spearman coefficient and Pearson coefficient are larger, monotonicity and the accuracy of illustration method are better, the accuracy that is method is higher), as shown in table 1:

Table 1

Quality evaluating method	The Spearman coefficient correlation	Pearson correlation coefficient
			PSNR	0.5389	0.6135
SSIM	0.7023	0.7321
			Present embodiment	0.7754	0.7912

Can see that from table 1 result the as a result accuracy rate of present embodiment will be higher than other 2 kinds of existing method for evaluating video quality.

Claims (9)

1. a method for evaluating video quality of estimating based on the macro block territory distortion factor comprises the steps:

(1) the every two field picture of video to be evaluated is divided into several macro blocks;

(2) calculate the blocking effect distortion factor of macro block;

(3) calculate the blurring effect distortion factor of macro block;

(4) calculate the luminance contrast of macro block;

(5) calculate the Texture complication of macro block;

(6) calculate the exercise intensity contrast of macro block;

(7) direction of motion that calculates macro block is unanimously spent;

(8) according to described luminance contrast, Texture complication, the exercise intensity contrast is consistent with the direction of motion spends, and calculates the visually-perceptible degree of macro block;

(9) according to the described blocking effect distortion factor, the blurring effect distortion factor and visually-perceptible degree, calculate the quality evaluation value of the every two field picture of video to be evaluated; Quality evaluation value to all images of video to be evaluated is averaging, and the mean value that obtains is the quality evaluation value of video to be evaluated.

2. method for evaluating video quality according to claim 1, it is characterized in that: in the described step (2), the method for the blocking effect distortion factor of computing macro block is as follows:

A. calculate the horizontal block effect distortion factor of current macro according to following formula:

S _BH(i)＝|A _(16，i)-B _(1，i)|

S _IH(1，i)＝|A _(14，i)-A _(15，i)|

S _IH(2，i)＝|A _(15，i)-A _(16，i)|

S _IH(3，i)＝|B _(1，i)-B _(2，i)|

S _IH(4，i)＝|B _(2，i)-B _(3，i)|

$S_{\mathrm{IH}\_\mathrm{AVG}\left(i\right)}=\frac{S_{\mathrm{IH}(1,i)}+S_{\mathrm{IH}(2,i)}+S_{\mathrm{IH}(3,i)}+S_{\mathrm{IH}(4,i)}}{4}$

$J_{H\left(i\right)}=\left\{\begin{array}{cc}{S}_{\mathrm{BH}\left(i\right)}-S_{\mathrm{IH}\_\mathrm{AVG}\left(i\right)}& \mathrm{if}(S_{\mathrm{BH}\left(i\right)}>S_{\mathrm{IH}\_\mathrm{AVG}\left(i\right)})\\ 0& \mathrm{otherwise}\end{array}\right.$

$D_{\mathrm{BLOCK}\_H}=\frac{1}{16}\underset{i=1}{\overset{16}{\mathrm{\Σ}}}{J}_H\left(i\right)$

Wherein, D _{BLOCK_H}The horizontal block effect distortion factor for current macro, A (16, i) be the brightness value of current macro the 16th row i row pixel, A (15, i) be the brightness value of current macro the 15th row i row pixel, A (14, i) be the brightness value of current macro the 14th row i row pixel, B (1, i) be and the brightness value of following adjacent macro block the 1st row i row pixel of current macro that B (2, i) be brightness value with following adjacent macro block the 2nd row i row pixel of current macro, B (3, i) be and the brightness value of following adjacent macro block the 3rd row i row pixel of current macro that i is natural number and 1≤i≤16;

B. calculate the vertical blocks effect distortion factor of current macro according to following formula:

S _BV(i)＝|A _(i，16)-C _(i，1)|

S _IV(i，1)＝|A _(i，14)-A _(i，15)|

S _IV(i，2)＝|A _(i，15)-A _(i，16)|

S _IV(i，3)＝|C _(i，1)-C _(i，2)|

S _IV(i，4)＝|C _(i，2)-C _(i，3)|

$S_{\mathrm{IV}\_\mathrm{AVG}\left(i\right)}=\frac{S_{\mathrm{IV}(i,1)}+S_{\mathrm{IV}(i,2)}+S_{\mathrm{IV}(i,3)}+S_{\mathrm{IV}(i,4)}}{4}$

$J_{V\left(i\right)}=\left\{\begin{array}{cc}{S}_{\mathrm{BV}\left(i\right)}-S_{\mathrm{IV}\_\mathrm{AVG}\left(i\right)}& \mathrm{if}(S_{\mathrm{BV}\left(i\right)}>S_{\mathrm{IV}\_\mathrm{AVG}\left(i\right)})\\ 0& \mathrm{otherwise}\end{array}\right.$

$D_{\mathrm{BLOCK}\_V}=\frac{1}{16}\underset{i=1}{\overset{16}{\mathrm{\Σ}}}{J}_{V\left(i\right)}$

Wherein, D _{BLOCK_v}The vertical blocks effect distortion factor for current macro, A (i, 16) be the brightness value of capable the 16th row pixel of current macro i, A (i, 15) be the brightness value of capable the 15th row pixel of current macro i, A (i, 14) be the brightness value of capable the 14th row pixel of current macro i, C (i, 1) be the brightness value of the macro block i capable 1st row pixel adjacent with current macro the right, C (i, 2) be the brightness value of capable the 2nd row pixel of macro block i adjacent on the right of current macro, C (i, 3) is the brightness value of the macro block i capable 3rd row pixel adjacent with current macro the right;

C. the described horizontal block effect distortion factor and the vertical blocks effect distortion factor are averaging, obtain the blocking effect distortion factor of current macro.

3. method for evaluating video quality according to claim 1, it is characterized in that: in the described step (3), the method for the blurring effect distortion factor of computing macro block is as follows:

A. utilize the Sobel operator that macro block is carried out rim detection, determine edge pixel and gradient direction thereof in the macro block;

B. the gradient direction of edge pixel is ranged a kind of in the middle of { 0 °, 45 °, 90 °, 135 °, 180 °, 225 °, 270 °, 315 ° } these eight kinds of directions according to nearby principle, and as the approximate gradient direction of edge pixel;

C. according to the clear-cut margin degree of following formula edge calculation pixel:

$S=\frac{1}{4}\underset{l=1}{\overset{4}{\mathrm{\Σ}}}\frac{|I_{(\mathrm{\θ},l)}-I|}{l}$

Wherein: S is the clear-cut margin degree of edge pixel, and I is the brightness value of edge pixel, I (θ, l) be with edge pixel along the approximate gradient direction θ distance brightness value for the pixel of l, l is natural number and 1≤l≤4;

D. the clear-cut margin degree of all edge pixels in the macro block is averaging, obtains the blurring effect distortion factor of macro block.

4. method for evaluating video quality according to claim 1, it is characterized in that: in the described step (4), the method of the luminance contrast of computing macro block is: according to the luminance contrast of each pixel in the following formula computing macro block, get the maximum of all pixel intensity contrasts in the macro block as the luminance contrast of macro block;

$C_{\mathrm{Luma}\_P}(\mathrm{\θ},l)=\frac{|I_{(\mathrm{\θ},l)}-I|}{l^{\mathrm{\η}}}$

$C_{\mathrm{Luma}}=\underset{\mathrm{\θ}\∈M}{\overset{M}{\mathrm{\Σ}}}\underset{l=1}{\overset{L}{\mathrm{\Σ}}}{C}_{\mathrm{Luma}\_P}(l,N)$

Wherein, C _LumaLuminance contrast for current pixel, I is the brightness value of current pixel, I (θ, l) be along the brightness value of direction θ distance for the pixel of l with current pixel, l is natural number and 1≤l≤L, L is default ultimate range, and η is pixel distance decline coefficient, and M is the set of 0 °, 90 °, 180 ° and 270 ° these four kinds of directions.

5. method for evaluating video quality according to claim 1, it is characterized in that: in the described step (5), the method for the Texture complication of computing macro block is as follows:

A. utilize the Sobel operator that macro block is carried out rim detection, determine total number of edge pixel in the macro block and the gradient direction of each pixel;

B. the gradient direction of pixel is ranged a kind of in the middle of { 0 °, 45 °, 90 °, 135 °, 180 °, 225 °, 270 °, 315 ° } these eight kinds of directions according to nearby principle, and as the approximate gradient direction of pixel;

C. the approximate gradient direction with pixel is divided into four classes: 0 ° and 180 ° is classified as a class, 45 ° and 225 ° and is classified as a class, 90 ° and 270 ° and is classified as a class, 135 ° and 315 ° and is classified as a class;

D. according to the Texture complication of following formula computing macro block:

$T_{\mathrm{block}}=\left\{\begin{array}{cc}0.5& \mathrm{if}(k_{\mathrm{\θ}}=1)\\ 1& \mathrm{if}(k_{\mathrm{\θ}}=2)\\ (2-c_e)/2& \mathrm{if}(k_{\mathrm{\θ}}=3)\\ (1-c_e)/2& \mathrm{if}(k_{\mathrm{\θ}}=4)\end{array}\right.$

$c_e=\left\{\begin{array}{cc}1& \mathrm{if}(n_{\mathrm{edge}}>T_{\mathrm{edge}})\\ 0& \mathrm{otherwise}\end{array}\right.$

Wherein, T _BlockBe the Texture complication of macro block, k _θBe the species number of all pixel approximate gradient directions in the macro block, n _EdgeBe total number of edge pixel in the macro block, T _EdgeBe given edge pixel number threshold value.

6. method for evaluating video quality according to claim 1, it is characterized in that: in the described step (6), the method for the exercise intensity contrast of computing macro block is as follows:

A. calculate horizontal motion vector and the vertical motion vector of macro block by inter prediction; Described horizontal motion vector and vertical motion vector root mean square are obtained the exercise intensity of macro block;

B. centered by current macro, set up the reference windows that is formed by 7 * 7 macro blocks;

C. the exercise intensity difference distance of calculating each macro block in current macro and the reference windows according to following formula compares M _{I_diff}, and get wherein that maximum is M _{Id_max}:

$M_{I\_\mathrm{diff}}=\frac{|M_I-M_I\left(D\right)|}{d^{\mathrm{\δ}}}$

Wherein: M _IBe the exercise intensity of current macro, M _I(D) be the exercise intensity of macro block D, macro block D is the arbitrary macro block in the reference windows, and d is the distance of current macro and macro block D, and δ is that macro block is apart from the decline coefficient;

D. determine the exercise intensity contrast of current macro according to following relational expression:

$C_{\mathrm{Motion}\_\mathrm{block}}=\left\{\begin{array}{cc}\frac{M_{\mathrm{Id}\_\max }}{M_{I\_\max }}& \mathrm{if}(M_{I\_\max }\≠0)\\ 0& \mathrm{otherwise}\end{array}\right.$

Wherein: C _{Motion_block}Be the exercise intensity contrast of current macro, M _{I_max}Maximum for all macro block exercise intensities in the reference windows.

7. method for evaluating video quality according to claim 1 is characterized in that: in the described step (7), the method for the consistent degree of the direction of motion of computing macro block is as follows:

A. calculate the horizontal motion vector mv of macro block by inter prediction _xWith vertical motion vector mv _yDescribed horizontal motion vector and vertical motion vector root mean square are obtained the exercise intensity of macro block, and according to formula θ _Mv=arctan (mv _y/ mv _x) calculate the direction of motion θ of macro block _Mv

B. 0 ° to 360 ° circumference equal dividing is become 12 sectors, make the direction of motion of macro block map to corresponding sector;

C. centered by current macro, set up the reference windows that is formed by 21 * 21 macro blocks;

D. unanimously spend according to the direction of motion of following formula calculating current macro:

$M_{\mathrm{Con}\_\mathrm{block}}=-\underset{j=1}{\overset{12}{\mathrm{\Σ}}}p\left(j\right)\log \left[p\left(j\right)\right]\frac{M_{I\_\mathrm{avg}}\left(j\right)}{M_{I\_\max }}$ $p\left(j\right)=\frac{k\left(j\right)}{{21}^2}$

Wherein: M _{Con_block}For the direction of motion of current macro is unanimously spent, k (j) is for belonging to the total number of macro block of j sector, M in the reference windows _{I_max}Be the maximum of all macro block exercise intensities in the reference windows, M _{I_avg}(j) for belonging to the mean value of all macro block exercise intensities of j sector in the reference windows, j is natural number and 1≤j≤12.

8. method for evaluating video quality according to claim 1 is characterized in that: in the described step (8), according to the visually-perceptible degree of following formula computing macro block:

VP _S＝log(α ₁+C _{Luma_block})×(α ₂+T _block) ²

VP _T＝α ₃C _{Motion_block}+α ₄M _{com_block}

VM＝λ×VP _S×VP _T

Wherein: VM is the visually-perceptible degree of macro block, M _{Con_block}For the direction of motion of macro block is unanimously spent C _{Motion_block}Be the exercise intensity contrast of macro block, T _BlockBe the Texture complication of macro block, C _{Luma_block}Be the luminance contrast of macro block, λ, α ₁, α ₂, α ₃And α ₄Be given weight coefficient.

9. method for evaluating video quality according to claim 1 is characterized in that: in the described step (9), calculate the quality evaluation value of the every two field picture of video to be evaluated according to following formula:

$Q=\frac{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}[Q_{\mathrm{MB}}\left(n\right)\×\mathrm{VM}\left(n\right)]}{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\mathrm{VM}\left(n\right)}$

Q _MB(n)＝γ1×D _{BLOCK_MB}(n)+γ ²×D _{BLUR_MB}(n)

Wherein: Q is the quality evaluation value of the arbitrary two field picture of video to be evaluated, D _{BLOCK_MB}(n) be the blocking effect distortion factor of n macro block in the arbitrary two field picture of video to be evaluated, D _{BLUR_MB}(n) be the blurring effect distortion factor of n macro block in the arbitrary two field picture of video to be evaluated, VM (n) is the visually-perceptible degree of n macro block in the arbitrary two field picture of video to be evaluated, γ ₁And γ ₂Be given weight coefficient, n is natural number and 1≤n≤N, and N is total number of macro block in the arbitrary two field picture of video to be evaluated.

Images