CN1157059C - Movement character combined video quality evaluation method - Google Patents

Abstract

The present invention relates to a video quality evaluating method combining motion characteristics, which belongs to the field of computer digital video processing. The evaluating method is a comprehensive evaluating method combining the motion characteristics of the sequence of a video image, considering the visual characteristics of human eyes and organically combining the time domain characteristics and the space domain characteristics of a video. The method provides a describing method of the intense degree of motion and provides a measuring method of video definition and fluency based on the describing method, and comprehensive evaluating results are given combining the definition with the fluency. The method orderly comprises: video signals are read in from a video capture card by a computer and a video capture interval T is stored; a subprogram of video image compressing processing is carried out; a reconstruct subprogram of the video image is carried out; a subprogram for calculating the intense degree of the motion of a video sequence is carried out; comprehensive evaluating results are calculated after a subprogram for video quality comprehensive evaluation is carried out; the results can be compared with the visual feelings of the human eyes when the video image is watched from a display. Experiments prove that the results comprehensively evaluated are basically consistent with the visual feelings of the human eyes.

Description

A kind of movement character combined method for evaluating video quality

Technical field

The present invention relates to the computer digit field of video processing, designed a kind of movement character combined method for evaluating video quality.

Background technology

Video quality evaluation is one of basis of image/video information engineering, as in video communication, the video information of subject is transferred to receiving terminal, recover acceptable video, wherein will pass through processes such as light-to-current inversion, compression processing, transmission, record and other conversion, the quality of all these technology all can collect the evaluation of video quality.

For being the image/video information of the final stay of two nights with human eye, should be consistent to its evaluation with the subjective feeling of human eye.The mechanism that human vision is handled is very complicated, up to the present, is not also got a real idea of and grasps.But people have also found some visual phenomenons, and these phenomenons have produced influence to the evaluation that people study video quality.Particularly, these visual characteristics have multi-channel structure, vision threshold value and shelter.

1. multi-channel structure and vision threshold value

People's vision system is a multi-channel structure, and it becomes different sensation components to the picture breakdown of input.Each sensory channel all has its oneself threshold value, is referred to as the vision threshold value.Give the excitation in the routing, if its value is lower than the pairing vision threshold value of passage, human eye should encourage with regard to imperceptible.

2. masking effect

When having a plurality of excitation, will interfere with each other between them, cause the vision threshold value to change, be referred to as the visual masking effect.To the video image of coding transmission, original image is the person of sheltering, and the coding transmission damage is a target.Real video is ever-changing, and the background of the damage component behind the coding is also of all kinds, so the observability of damage component also changes.

The existence of vision threshold value makes that the damage that is lower than threshold value is not perceiveed, and the existence of sheltering then makes the vision threshold value promote, thereby sightless damage is increased.This is the important starting point of research based on human visual system's video quality metric.In other words, if manage to make damage to appear at place invisible to the human eye, also just improved the quality of image.

Sheltered various forms.Human eye descends in very bright or very dark zone to the susceptibility of damage, is called contrast and shelters; Human eye is little regional lower than changing to the zone that the susceptibility of damage changes in the image space territory greatly, is called texture to shelter; Content changes big image block in time, and human eye is low to the susceptibility of its damage, and this is that motion is sheltered; Moment after scene is switched, human eye descends to the observability of damage, and this is to switch to shelter.

At present people mainly are based on the evaluation of video quality the evaluation of rest image are carried out, and measure by average peak signal to noise ratio and frame frequency.Frame frequency is the frame number of transmission each second; Average peak signal to noise ratio is defined as:

$\stackrel{\‾}{\mathrm{PSNR}}=\frac{1}{P}\underset{k=0}{\overset{P-1}{\mathrm{\Σ}}}\mathrm{PSN}{R}_k$

Wherein

${\mathrm{PSNR}}_k=10\·{\log }_{10}\frac{255*255*N*M}{\underset{i=0}{\overset{N-1}{\mathrm{\Σ}}}\underset{j=0}{\overset{M-1}{\mathrm{\Σ}}}{(f_k(i,j)-f_k^{\′}(i,j))}^2},$

f _k(i, j) and f ' _k(i j) is the gray scale of k frame original image and reconstructed image respectively, and M, N are respectively the wide and high of image.

Average peak signal to noise ratio is the distortion level of reflecting video objectively, but do not consider the visual characteristic of human eye, and frame frequency and average peak signal to noise ratio be separate reflecting video time-domain and spatial domain characteristic, and this evaluation method is isolated the time domain characteristic of video image and spatial domain characteristic and come.Thereby this evaluation result is often inconsistent with people's visual experience.

Summary of the invention

In order to overcome the deficiency of present evaluation method, the present invention has designed a kind of movement character combined method for evaluating video quality.This evaluation method has been considered the visual characteristic of human eye, and the spatial domain characteristic and the time domain characteristic of video integrated consideration, and its evaluation result can be consistent with the visual experience of human eye.Technical thought of the present invention is characterised in that:

1, two kinds of definition of video definition and video fluency have been proposed, and with its two kinds of measures as video quality evaluation.So-called definition is meant the degree of knowing of each two field picture; Fluency is meant the continuous degree of one section video.

2, the describing method of video image motion severe degree has been proposed.Its description is divided into two parts: the motion severe degree of each image block distributes in (1) one two field picture, and the spatial distribution characteristic that we are referred to as to move is applied to the video definition evaluation; (2) the motion severe degree of whole frame video image, the time distribution characteristics that we are referred to as to move is applied to the evaluation of video fluency.

3, considered the visual characteristic of human eye.Human eye is subjected to the influence of video pictures motion severe degree to the impression of video definition and fluency.For a certain frame video image, the violent part of its motion often causes the attention of human eye, belongs to the human eye region of interest.And when the whole two field picture motion of one section video was violent, image definition was goaded into action the thorn of human eye and is weakened relatively, and the effect of picture fluency will strengthen relatively.

4, definition and the fluency with video image organically combines, and proposed to calculate the method for video quality comprehensive evaluation index.

Technical scheme of the present invention is referring to Fig. 1, Fig. 2.It combines the visual characteristic of the motion feature and the human eye of video image, to carrying out objective appraisal through the video image after the Computer Processing.This scheme includes camera (1) image light signals of the object sequence of video images that obtains is converted into the signal of telecommunication, to come from the video sequence digitlization of camera and be input in the computer processor by capture card (2), it is characterized in that it also is included in the method for the calculating video quality comprehensive evaluation index that definition and the fluency with the video image that are provided with in the computer processor (3) organically combine, this method comprises the steps: successively

1), computer reads in vision signal from video frequency collection card, and preserves the video acquisition interval T;

2), enter and finish the original video frame that collection is entered computer and carry out processing such as motion compensation, conversion and coding, handle subprogram with the video image compression of compression active redundancy;

3), enter the video code flow after the compression is decoded, the video image reconstruction subprogram of reconstruction video sequence image;

4), according to video definition and two kinds of definition of video fluency, and with its two kinds of measures as video quality evaluation:

Be introduced into and calculate video sequence motion severe degree subprogram, each frame video image is divided into 16 * 16 image block, calculate the motion severe degree of each image block and the motion severe degree of a whole two field picture;

Enter video quality overall merit subprogram again,, calculate the definition and the fluency of video sequence earlier, calculate comprehensive evaluation result then according to aforesaid motion severe degree;

5), from considering human eye the impression of video definition and fluency is subjected to the influence of video pictures motion severe degree, to 4) comprehensive evaluation result output, can compare with the visual experience that human eye is watched from display.

Calculate the motion severe degree of video in two steps according to aforementioned the present invention, the first step: calculate the motion severe degree of each image block, its calculating can be passed through two kinds of methods:

First kind is that motion vector according to image block calculates, and its computing formula is:

${\mathrm{MA}}_k(i,j)=\sqrt{{\mathrm{\Δx}}_k^2(i,j)+{\mathrm{\Δy}}_k^2(i,j)}$

MA in the formula _k(i, j) (i, j) the motion severe degree of individual image block, (Δ x in the expression k frame video image _k, Δ y _k) be the motion vector of this piece.Second kind is utilized the pixel grey scale difference of adjacent two two field pictures to calculate, and its computing formula is:

${\mathrm{MA}}_k(i,j)=\mathrm{EXP}\left(\frac{1}{255\·N_p}\underset{(m,n)\∈\mathrm{block}(i,j)}{\mathrm{\Σ}}\right|L_k(m,n)-L_{k-1}(m,n)\left|\right)$

Block in the formula (i, j) (i, j) individual piece, L in expression one two field picture _k(m, n) and L _K-1(m n) represents that respectively (m n) locates the gray value of pixel, N for k frame and k-1 two field picture _pThe number of representing pixel in the image block.

Second step: calculate the motion severe degree of a whole frame video image, its computing formula is:

${\mathrm{MA}}_k=\frac{1}{N_{\mathrm{MB}}}\underset{i}{\mathrm{\Σ}}\underset{j}{\mathrm{\Σ}}{\mathrm{MA}}_k(i,j)$

MA in the formula _kThe motion severe degree of representing the k frame video image, N _MBThe number of representing image block in the two field picture, MA _k(i, meaning j) with aforementioned in identical.

When calculating video image clarity, the video sequence of supposing to be input to computer is to collect with constant high speed frame frequency according to aforementioned the present invention, and acquisition interval is T; Calculate video image clarity generally speaking.Not all frame all be encoded (processing) output.K represent the to be encoded image frame number of (processing), N represent the to be encoded totalframes of (processing) image.Interval time is FT between k frame and the k-1 frame _kT.We consider the visual characteristic of human eye based on the Y-PSNR of rest image, compose with higher weight for the Y-PSNR of the more violent image block of motion, and the less macro block that moves is with less weight.At first calculate the definition of a frame video image, its computing formula is:

${\mathrm{SS}}_k=10\·{\log }_{10}\frac{255\·255}{\frac{1}{N_{\mathrm{MB}}}\underset{i,j}{\mathrm{\Σ}}{\mathrm{MA}}_k(i,j)\·{\mathrm{Diff}}_k(i,j)}$

SS in the formula _kRepresent a frame video image definition, N _MBAnd MA _k(i, identical in meaning j) and 7,

${\mathrm{Diff}}_k(i,j)=\underset{(m,n)\∈\mathrm{block}(i,j)}{\mathrm{\Σ}}{(L_{k,\mathrm{Input}}(m,n)-L_{k,\mathrm{Output}}(m,n))}^2,$

L wherein _{K, Input}(m, n) and L _{K, Output}(m, n) be respectively the video of acquired original and handle the back reconstruction video (m n) locates grey scale pixel value, calculates the definition of one section video then, and its computing formula is:

$\mathrm{PS}=\frac{1}{N}\underset{k=1}{\overset{N}{\mathrm{\Σ}}}(\frac{1}{{\mathrm{MA}}_k}\·{\mathrm{FT}}_k\·{\mathrm{SS}}_k)$

PS represents the definition of one section (N frame) video in the formula, and PS is big more, and the explanation video image picture is clear more.This computing formula comprises the implication of two aspects: on the one hand, and a little less than the effect relatively of the definition of the comparatively violent picture frame that moves in whole section video definition estimated; On the other hand, the effect of definition in whole section video definition estimated of long picture frame is stronger relatively for holding time in the video playback, and this is the visual experience that meets human eye.

According to aforementioned the present invention when calculating the fluency of one section video, consider that human eye is for the more violent picture frame of motion, detailed information that can't resolution image, at this moment the definition of image diminishes to the stimulation of human eye, and the continuity of motion is big to the thorn violent change of human eye, thereby need improve frame frequency when motion is violent.For the frame of video that motion relaxes, at this moment the detailed information that human eye can resolution image need improve the definition of every frame video image, does not need higher frame frequency.This visual characteristic according to human eye is calculated the video fluency, and its computing formula is:

$\mathrm{PT}=\frac{1}{N}\underset{k=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{MA}}_k\·{\mathrm{FT}}_k)$

PT represents the fluency of one section (N frame) video in the formula, and PT novel photopic vision more is frequently smooth more.

When calculating the comprehensive evaluation index value of one section video image, its computing formula is according to aforementioned the present invention:

$P=W_S\·\mathrm{PS}+W_T\·\frac{1}{\mathrm{PT}}$

P represents the quality overall evaluation value of one section video image in the formula.PS is a definition, and PT is a fluency, W _SAnd W _TThe weight in comprehensive evaluation value for definition and fluency.

According to a large amount of experiments, we get W _SBe 0.55, W _TBe 0.45, the visual experience basically identical of its comprehensive evaluation result of such value and human eye.

Description of drawings

Fig. 1 is a video quality overall evaluation system block diagram.

Fig. 2 is a video quality overall evaluation system main program flow chart.

Fig. 3 is that video compression is handled subroutine flow chart.

Fig. 4 is that video is rebuild subroutine flow chart.

Fig. 5 calculates video motion severe degree subroutine flow chart.

Fig. 6 calculates video synthesis quality evaluation index value subroutine flow chart.

Fig. 7 is the 150 frame video sequence motion severe degree distribution maps that utilize the motion vector method to calculate, and wherein transverse axis is represented the frame of video sequence number, and the longitudinal axis is represented the motion severe degree of frame of video.

Fig. 8 is the 150 frame video sequence motion severe degree distribution maps that utilize the absolute difference method to calculate, and wherein transverse axis is represented the frame of video sequence number, and the longitudinal axis is represented the motion severe degree of frame of video.

Fig. 9 is each frame definition distribution map of video sequence in two different examples, and wherein transverse axis is represented the sequence number of frame of video, and the longitudinal axis is represented the definition of frame of video.The example that [-■-Variable Frame Rate] expression is play with variable frame frequency among the figure, the example that [-*-Constant Frame Rate] expression is play with constant frame frequency.

Among the figure 1, camera, 2, video frequency collection card, 3, the computer digit video processor, 4, output buffers, 5, video quality evaluation, 6, display, 7, evaluation result.

Table 1: be compressed the frame of video sequence number of encoding process in two different instances, represent for first group to play, represent for second group to play with variable frame frequency with constant frame frequency.

Embodiment

In the video quality overall evaluation system block diagram of Fig. 1, camera and video frequency collection card all are commercially available, mainly finish the collection video sequence, the optical imagery of target object is converted to signal of telecommunication image is input to computer, be convenient to operations such as Computer Processing, transmission; The computer video processing mainly is that the sequence of video images of importing is carried out processing such as compressed encoding; The processed video image outputs to buffer, is convenient to show; Display is the output equipment of video sequence, and human eye can be watched sequence of video images by display, and the visual experience of human eye can be used for and the video quality comprehensive evaluation result is compared; Video quality overall merit is that the reconstruction video image after the Computer Processing is estimated, output objective appraisal result, and its result can compare from the visual experience of video sequence that display is seen with human eye.

Video quality evaluation is mainly realized by software.Describe the process of video quality evaluation in detail below in conjunction with example.

We have obtained 150 frame video sequence images by camera and video frequency collection card, are kept at the hard disk of computer.At first this section video is carried out compressed encoding and handle, the video code flow after the compression is rebuild the back and is compared with original video, and certain distortion is necessarily arranged.Now the video sequence of rebuilding again after the compression is carried out overall merit by the method for the present invention's design.

The first step: the motion severe degree of calculating each frame of video.Two kinds of methods of describing the motion severe degree that propose with the present invention are calculated respectively, Fig. 7 and Fig. 8 are respectively the motion severe degree distribution maps of the 150 frame video sequences that obtain with two kinds of methods, wherein transverse axis is represented the frame of video sequence number, and the longitudinal axis is represented the motion severe degree of frame of video.

As can be seen, Fig. 7 is similar with Fig. 8, this explanation with these two kinds of methods can both the objective description image the motion severe degree.And can find out that from the 60th frame to 110 frames, image motion gets violent relatively, other each frames motions relax relatively, this and eye-observation to actual conditions match.

Second step: calculate the video sequence definition.For the validity of designed method of the present invention to video evaluation is described, design two groups of examples here so that compare.

Illustrate that at first human eye treats the impression of video sequence image.When our human eye when watching one section video, for video pictures motion time period more slowly, the detailed information that human eye can the resolution image picture, also can be interested in the details of image, because motion relaxes, the difference of front and back frame is little, so lower frame frequency human eye is an acceptable; And concerning violent time period of video pictures motion, the detailed information that human eye can't resolution image, and the continuity of motion is had higher requirement, give the visual experience of human eye more better so at this moment improve frame frequency.Therefore, if the broadcast frame frequency of video sequence can change according to the motion severe degree of picture, the visual experience of human eye is better when playing with constant frame frequency so.

Owing to such fact is arranged, so design following two groups of examples.To same section video sequence, limiting bandwidth is identical in two groups of examples, and other conditions are also identical, just the frame frequency difference.

First group: frame frequency is constant to be 10 frame/seconds;

Second group: frame frequency changes according to the motion severe degree, but average frame was 10 frame/seconds frequently.Why average frame frequency is identical with first group, is because the requirement of doing bandwidth like this is the same.

The actual frame number of handling that is encoded is shown in Table 1 in two groups of examples, therefrom as can be seen, first framing is frequently constant, and second framing changes frequently, in the time period of 60 frame to 110 frames, frame frequency is higher, and frame frequency is lower in the other times section, and the motion severe degree of the frame of video of this and previous calculations matches.

Fig. 7 and Fig. 8 as can be seen, to 110 frames, image motion gets violent relatively from the 60th frame, thereby in second group of experiment, the frame number that is encoded in this time period is more relatively, and the frame number that is encoded in the other times section is less, this meets the visual characteristic of human eye.

Fig. 9 shows each frame definition statistics of video in two groups of examples, and wherein transverse axis is represented the sequence number of frame of video, and the longitudinal axis is represented the definition of frame of video.[-■-Variable Frame Rate] represents variable frame frequency among the figure, and [-*-Constant Frame Rate] represents constant frame frequency.As can be seen from Figure 9, in second group of example, in the video-frequency band that motion relatively relaxes, definition is higher relatively, and in the violent video-frequency band of motion, definition is relatively low.The video sequence definition that the method that designs according to the present invention is calculated two groups of examples is:

First group: 28.2;

Second group: 32.5.

The video sequence definition is high more, gives people's visual experience good more, and the result that the method for utilizing the present invention to design obtains is consistent with the visual experience of human eye.

The 3rd step: calculate the video sequence fluency.The example that designs with second step is identical, calculates two groups of instance video sequence fluency according to the method for the present invention's design to be:

First group: 10.5;

Second group: 9.0

The video sequence fluency is low more, gives the visual experience of human eye good more, and the result of this evaluation method is consistent with the visual experience of human eye.

The 4th step: calculate the video sequence comprehensive evaluation value.According to overall merit formula of the present invention, we get W _SBe 0.55, W _TThe video sequence comprehensive evaluation value that is 0.45, two group of experiment is:

First group: 15.6

Second group: 17.9

The high more explanation video sequence of comprehensive evaluation value gives the visual experience of human eye good more.The evaluation method of the present invention's design is consistent with the subjective vision impression of human eye.

Table 1

First group of example	Second group of example
		0，12，18，24，30，36， 42，48，54，60，66，72， 78，84，90，96，102， 108，114，120，126， 132，138，144，147	0，4，6，20，26，38，50， 56，64，68，72，76，78， 82，86，90，96，100， 104，110，114，122， 126，138，144，148，149

Claims (5)

1, a kind of movement character combined method for evaluating video quality, comprise by camera (1) image light signals of the sequence of video images of the object that obtains is converted into the signal of telecommunication, to come from the video sequence digitlization of camera and be input in the computer processor by capture card (2), it is characterized in that it also is included in the method for the calculating video quality comprehensive evaluation index that definition and the fluency with the video image that are provided with in the computer processor (3) organically combine, this method comprises the steps: successively

1), computer reads in vision signal from video frequency collection card, and preserves the video acquisition interval T;

2), enter and finish the original video frame that collection is entered computer and carry out processing such as motion compensation, conversion and coding, handle subprogram with the video image compression of compression active redundancy;

3), enter the video code flow after the compression is decoded, the video image reconstruction subprogram of reconstruction video sequence image;

4), according to video definition and two kinds of definition of video fluency, and be with its step after as two kinds of measures of video quality evaluation:

Be introduced into and calculate video sequence motion severe degree subprogram, each frame video image is divided into 16 * 16 image block, calculate the motion severe degree of each image block and the motion severe degree of a whole two field picture;

Enter video quality overall merit subprogram again,, calculate the definition and the fluency of video sequence earlier, calculate comprehensive evaluation result then according to aforesaid motion severe degree;

5), from considering human eye the impression of looking video definition and fluency is subjected to the influence of video pictures motion severe degree, to 4) comprehensive evaluation result output, the visual experience that is used for watching from display with human eye is compared.

2, a kind of movement character combined method for evaluating video quality according to claim 1, it is characterized in that calculating in two steps the motion severe degree of video, the first step: the motion severe degree of calculating each image block, its calculating can be by two kinds of methods: first kind is that motion vector according to image block calculates, and its computing formula is:

${\mathrm{MA}}_k(i,j)=\sqrt{{\mathrm{\Δx}}_k^2(i,j)+{\mathrm{\Δy}}_k^2(i,j)}$

MA in the formula _k(i, j) (i, j) the motion severe degree of individual image block, (Δ x in the expression k frame video image _k, Δ y _k) be the motion vector of this piece, second kind is utilized the pixel grey scale difference of adjacent two two field pictures to calculate, and its computing formula is:

${\mathrm{MA}}_k(i,j)=\mathrm{EXP}\left(\frac{1}{255\·N_p}\underset{(m,n)\∈\mathrm{block}(i,j)}{\mathrm{\Σ}}\right|L_k(m,n)-L_{k-1}(m,n)\left|\right)$

Block in the formula (i, j) (i, j) individual piece, L in expression one two field picture _k(m, n) and L _K-1(m n) represents that respectively (m n) locates the gray value of pixel, N for k frame and k-1 two field picture _pRepresent the number of pixels in the image block; Second step: calculate the motion severe degree of a whole frame video image, its computing formula is:

${\mathrm{MA}}_k=\frac{1}{N_{\mathrm{MB}}}\underset{i}{\mathrm{\Σ}}\underset{j}{\mathrm{\Σ}}{\mathrm{MA}}_k(i,j)$

MA in the formula _kThe motion severe degree of representing the k frame video image, N _MBThe number of representing image block in the two field picture.

3, a kind of movement character combined method for evaluating video quality according to claim 1, when it is characterized in that calculating video image clarity, the video sequence of supposing to be input to computer is to collect with constant high speed frame frequency, acquisition interval is T; Generally speaking, not all frame all be encoded (processing) output; K represent the to be encoded image frame number of (processing), N represent the to be encoded totalframes of (processing) image; Interval time is FT between k frame and the k-1 frame _kT; We consider the visual characteristic of human eye based on the Y-PSNR of rest image, compose with higher weight for the Y-PSNR of the more violent image block of motion, and the less macro block that moves is with less weight; At first calculate the definition of a frame video image, its computing formula is:

${\mathrm{SS}}_k=10\·{\log }_{10}\frac{255\·255}{\frac{1}{N_{\mathrm{MB}}}\underset{i,j}{\mathrm{\Σ}}{\mathrm{MA}}_k(i,j)\·{\mathrm{Diff}}_k(i,j)}$

SS in the formula _kRepresent a frame video image definition, N _MBThe number of representing image block in the two field picture, MA _k(i, j) (i, j) the motion severe degree of individual image block in the expression k frame video image.

${\mathrm{Diff}}_k(i,j)=\underset{(m,n)\∈\mathrm{block}(i,j)}{\mathrm{\Σ}}{(L_{k,\mathrm{Input}}(m,n)-L_{k,\mathrm{Output}}(m,n))}^2,$

L wherein _{K, Input}(m, n) and L _{K, Output}(m, n) be respectively the video of acquired original and handle the back reconstruction video (m n) locates grey scale pixel value; Calculate the definition of one section video then, its computing formula is:

$\mathrm{PS}=\frac{1}{N}\underset{k=1}{\overset{N}{\mathrm{\Σ}}}(\frac{1}{{\mathrm{MA}}_k}\·{\mathrm{FT}}_k\·{\mathrm{SS}}_k)$

PS represents the definition of one section video in the formula, and PS is big more, and the explanation video image picture is clear more, MA _kThe motion severe degree of representing the k frame video image.

4, a kind of movement character combined method for evaluating video quality according to claim 1 when it is characterized in that calculating the fluency of one section video, calculates the video fluency according to this visual characteristic of human eye, and its computing formula is:

$\mathrm{PT}=\frac{1}{N}\underset{k=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{MA}}_k\·{\mathrm{FT}}_k)$

PT represents the fluency of one section (N frame) video in the formula, and PT novel photopic vision more is frequently smooth more, MA _kThe motion severe degree of representing the k frame video image, FT _kFrame-skipping number during for coding between k frame and the k-1 frame.

5, a kind of movement character combined method for evaluating video quality according to claim 1, when it is characterized in that calculating the comprehensive evaluation index value of one section video image, its computing formula is:

$P=W_S\·\mathrm{PS}+W_T\·\frac{1}{\mathrm{PT}}$

P represents the quality overall evaluation value of one section video image in the formula, and PS is a definition, and PT is a fluency, W _SAnd W _TThe weight in comprehensive evaluation value for definition and fluency.

Images