CN105976351A - Central offset based three-dimensional image quality evaluation method - Google Patents

Abstract

The invention belongs to the field of image processing, and relates to a central offset based three-dimensional image quality evaluation method which aims to enable the consistency between a result of objective evaluation and subjective evaluation to be higher and provides a new idea for three-dimensional image quality evaluation. The central offset based three-dimensional image quality evaluation method comprises the steps of 1) calculating a comparison function which refers to a right image, the brightness, the contrast and the structure of the right image by adopting a structural similarity (SSIM) algorithm, acquiring an image quality weight matrix of the SSIM, and then amplifying the image quality weight matrix to the size which is identical to that of an original image through a nearest neighbor domain interpolation algorithm; 2) carrying out weighing calculation on the SSIM image quality matrix according to a central offset characteristic so as to acquire an image quality evaluation score of the right image; and then 3) repeating the above steps, calculating an image quality evaluation score of a left viewpoint, and carrying out weighted average on the image quality evaluation score of the right image and the image quality evaluation score of the left viewpoint so as to acquire a three-dimensional image quality objective evaluation value. The method provided by the invention is manly applied to image processing.

Description

Stereo image quality evaluation methodology based on central offset

Technical field

The invention belongs to image processing field, relate to image quality evaluating method and improve and optimizate, particularly relate to a kind of based on The objective evaluation method for quality of stereo images of central offset.

Background technology

From mankind's history of the twenties in last century, first stereoscopic motion picture " strength of love " was come out, to 3D recasting version in 2012 " Titanic " is shown, and 3D stereoscopic motion picture relies on its outstanding visual experience, viewing experience true to nature, has been obtained for more coming The favor of the most beholders.Stereo display technique refers to be carried out the stereoscopic vision characteristic of human eye by technological means such as optics Simulation, thus reproduce out by the steric information of space object, generate the display mode [1] with depth characteristic stereo-picture.

But stereo-picture is due to problems such as noise during shooting, image transmitting distortions, may cause left and right two width figures The quality of picture difference, and this species diversity may result in the discomfort of beholder.Therefore, the quality of stereo-picture is commented Valency has vital effect to the development of stereo-picture.Although the evaluation and test of stereo-picture subjectivity can be published picture in the most correct reflection The quality of picture, but subjective evaluation and test is wasted time and energy, and for substantial amounts of image and picture, is evaluated each image being difficult in fact Existing.Therefore, the method for objectively evaluating reasonably proposing a kind of stereo-picture has the biggest meaning.

The method for objectively evaluating of stereo-picture, according to the characteristic information of stereo-picture self, utilizes mathematical formulae or passes through structure Build mathematical model, use computer that stereo-picture is analyzed, thus calculate the mark that represent stereo image quality, use To describe mankind's subjective feeling for stereo-picture.

By the end of now, the method for objectively evaluating of stereo image quality is a lot, and they join complete in terms of different Examine stereo image quality to be evaluated, below existing objective evaluation method for quality of stereo images is analyzed.

Document [2] is on the basis of reference plane image quality evaluation, by engineering evaluation methodology Y-PSNR and knot Structure similarity combines, and uses two kinds of methods to evaluate the quality of stereo-picture left and right view respectively, then different by four kinds Method is calculated absolute difference information for evaluating third dimension, uses the method that local combines and the overall situation combines will the most respectively Picture quality and third dimension quality are integrated into unified stereo image quality index.Finally obatained score is averaged, as Evaluate the index of stereo image quality.This article also demonstrates simple plane picture method for objectively evaluating and cannot be suitable for simply Stereo image quality is evaluated, and needs to consider relief factor in stereo image quality evaluation procedure.Document [3] proposes one Plant binocular perceived quality model, be primarily based on binocular unsymmetry segmentation stereo-picture, then zones of different arranged different Perception weight, finally calculates stereo image quality, and the document demonstrates and combines binocular vision unsymmetry and can improve axonometric chart As objective evaluation accuracy.Document [4] is thought, human visual system is extremely sensitive to the marginal information of stereo-picture, therefore its Consider by marginal information, classical architecture index of similarity to be improved, it is proposed that a kind of structural similarity based on edge is commented Valency method, uses the method to evaluate stereo-picture left and right viewing quality.Then author is by based on adaptive weighting matching algorithm Calculate the disparity map of left and right view, by judging that distorted image disparity map calculates stereo-picture with the difference of reference picture disparity map Third dimension index.Finally left and right viewing quality is fitted with third dimension quality, obtains evaluating combining of stereo image quality Close index.Document [5] considers physiological property and the psychology characteristic of HVS, it is proposed that a SSIM algorithm improved.

The continuous intensification recognized HVS characteristic along with each field, incorporates more complicated and senior in objective evaluation model Human-eye visual characteristic becomes inevitable developing direction.Vision significance, as a kind of human visual system's advanced feature, refers to The attention intensity that image zones of different is distributed by human eye is different.Document [6] is by by the notable figure weighted image of original image Localized distortion Quality Map, improve the performance of evaluation algorithms.Under normal circumstances, image fault can affect marked feature and accurately carries Take.Document [6] assumes that original image is similar with the marked feature of distorted image, and it uses the notable figure of original image to figure picture element Amount carries out objective evaluation, test result indicate that^[6], when distorted image quality is higher, use the distortion map that original notable figure is carried out As quality evaluation is effective；But, along with the continuous reduction of distorted image quality, the factor impact that some can not be ignored is notable Property detection process, thus cause the marked feature of original image and distorted image to have obvious difference.Therefore, in experimentation Should consider that original and distorted image marked feature evaluates the quality of distorted image more accurately simultaneously.

Owing to human visual system is from bottom to top to the observation of image, it is impossible to the content of view picture figure is seen simultaneously simultaneously Observe, but only focus on the most attractive place in image, it is proposed that stereo image quality evaluation based on central offset Method has the strongest theoretic support.

Summary of the invention

For overcoming the deficiencies in the prior art, it is contemplated that combine central offset characteristic stereo image quality is carried out objective Evaluate.By central offset characteristic during eye-observation image, stereo image quality objective evaluation algorithm is optimized, makes visitor See the result evaluated higher with the concordance of subjective evaluation and test, propose a new thinking for stereo image quality evaluation.This Bright employed technical scheme comprise that, stereo image quality evaluation methodology based on central offset, step is as follows:

1) use structural similarity algorithm SSIM, calculate with reference to right image and the brightness of right image, contrast and structure Comparison function, by thus draw the picture quality weight matrix of SSIM, then by nearest-neighbor interpolation algorithm by picture quality weigh Value matrix is amplified to identical with original image size；

2) according to central offset characteristic, SSIM picture quality matrix is weighted, obtains the picture quality of right image Evaluate score；

3) then repeat above-mentioned steps, calculate the image quality evaluation score of left view point, both weighted averages are obtained Stereo image quality objective evaluation value.

Use structural similarity algorithm, for preventing blocking effect, the Gauss sliding window using M × M, standard deviation to be 1.5 Mouth obtains subimage block X, Y, meter to the right viewpoint of original three-dimensional image pair and the right viewpoint sampling of distortion stereo pairs respectively Calculate their brightness, structure and contrast similarity；

$l(X,Y)=\frac{2{\mathrm{\μ}}_X{\mathrm{\μ}}_Y+C_1}{{{\mathrm{\μ}}_X}^2+{{\mathrm{\μ}}_Y}^2+C_1}---\left(1\right)$

$s(X,Y)=\frac{2{\mathrm{\σ}}_{XY}+C_3}{{\mathrm{\σ}}_X{\mathrm{\σ}}_Y+C_3}---\left(2\right)$

$c(X,Y)=\frac{2{\mathrm{\σ}}_X{\mathrm{\σ}}_Y+C_2}{{{\mathrm{\σ}}_X}^2+{{\mathrm{\σ}}_Y}^2+C_2}---\left(3\right)$

Wherein:

${\mathrm{\σ}}_X={\left(\underset{i=1}{\overset{N}{\Σ}}{\mathrm{\ω}}_i{\left(x_i-{\mathrm{\μ}}_X\right)}^2\right)}^{1/2}---\left(5\right)$

${\mathrm{\σ}}_{XY}=\underset{i=1}{\overset{N}{\Σ}}{\mathrm{\ω}}_i(x_i-{\mathrm{\μ}}_X)(y_i-{\mathrm{\μ}}_Y)---\left(6\right)$

Wherein, C₁、C₂、C₃Representing the least normal amount, avoiding denominator is zero, and C₁=(k₁L)²、C₂=(k₂L)²、C₃ =(C₂/2)²；k₁、k₂It is the constant between [0,1] respectively；x_i,y_iIt is image block X respectively, the value of ith pixel point, μ in Y_X,μ_Y It is respectively image block X, the average of Y, σ_X,σ_YIt is respectively image block X, the variance of Y, σ_XYFor the covariance of image block X, Y, N is figure As the pixel quantity of block X or Y, l (X, Y), c (X, Y), s (X, Y) are respectively image block X, the brightness of Y, contrast and structural parameters Matrix；

The structural similarity of image block is defined as:

SSIM (x, y)=(l (X, Y))^τ(c(X,Y))^β(s(X,Y))^γ (7)

Wherein τ, beta, gamma is regulation parameter, takes τ=β=γ=1, and (x, y) is the pixel of image, formula (7) calculate sliding Structural similarity in dynamic window, SSIM (x, y) is the result that calculates behind the image upper left corner to the lower right corner of sliding window, Its size is ((W-10) × (H-10)), and wherein W and H represents horizontal pixel and the vertical pixel number of image；For so obtain The mass matrix Q of the right viewpoint of distortion stereo pairs_R(x, y), use nearest-neighbor interpolation algorithm by SSIM (x, y) be amplified to Original image size is identical.

Nearest-neighbor interpolation algorithm is that the gray value at target pixel points is, by this pixel of distance around this pixel The gray value of near pixel determines, and it is not affected by other all of pixel；

(i, j), (i, j+1), (i+1, j), (i+1, j+1) be by floating-point coordinate (i+a, j+b) (i, j before interpolation respectively The respectively integer part of denotation coordination, a, b then distinguish the fractional part of denotation coordination, and a ∈ [0,1), b ∈ [0,1)), four Individual neighborhood, f (i, j), f (i, j+1), f (i+1, j), f (i+1, j+1) be the gray value of corresponding pixel points respectively, A, B, C, D divide Not Biao Shi pixel f (i, j), f (i, j+1), f (i+1, j), f (i+1, j+1) constituted the upper left in region, upper right, lower-left, Lower right area, nearest-neighbor difference arithmetic just determines that the point that aforementioned four point mid-range objectives pixel (i+a, j+b) is nearest, Then the gray value of its closest approach is exactly the gray value of target pixel points, and nearest-neighbor algorithm below equation represents:

$f(i+a,j+b)=\left\{\begin{array}{c}f(i,j),(a,b)\⊆A\\ f(i,j+1),(a,b)\⊆C\\ f(i+1,j),(a,b)\⊆B\\ f(i+1,j+1),(a,b)\⊆D\end{array}\right.---\left(8\right)$

Finally, the right viewpoint of distortion stereo pairs that above-mentioned nearest-neighbor interpolation algorithm will be obtained is used by SSIM algorithm Mass matrix SSIM (x, y) is amplified to original image size, and the image after now amplifying is the mass matrix Q of right viewpoint_R(x, y)。

Weighted calculation SSIM mass matrix

Use the Quality Map Q of the SSIM right viewpoint of algorithm calculated distortion_R(x y), uses formula (10) to obtain the right viewpoint of distortion Objective Quality Assessment value Q_r, use same method to obtain Objective Quality Assessment value Q of distortion left view point_l, finally use formula (11), both weighted averages are obtained stereo image quality objective evaluation value,

$Q_r=\frac{\underset{x=1}{\overset{H}{\Σ}}\underset{y=1}{\overset{W}{\Σ}}{Q}_R(x,y)CB(x,y)}{\underset{x=1}{\overset{H}{\Σ}}\underset{y=1}{\overset{W}{\Σ}}CB(x,y)}---\left(10\right)$

There is the central offset (CB) that anisotropic gaussian kernel function [11] simulation attention is spread by mediad surrounding The factor:

$CB(x,y)=\exp \{-(\frac{{(x-x_0)}^2}{2{\mathrm{\σ}}_h^2}+\frac{{(y-y_0)}^2}{2{\mathrm{\σ}}_v^2})\}---\left(9\right)$

Wherein (x, (x, y) to central point (x y) to represent pixel for CB₀,y₀) offset information, (x₀,y₀) represent that distortion is right The center point coordinate of viewpoint, (x y) is pixel coordinate, σ_hAnd σ_vRepresent image level direction and the standard of vertical direction respectively Difference, takes σ_h=1/3W, σ_v=1/3H, wherein W and H represents horizontal pixel and the vertical pixel number of image.

Stereo image quality objective evaluation value: respectively obtain picture quality objective evaluation value Q of left view and right view_rWith Q_l, then can obtain stereo image quality objective evaluation value by weighted calculation:

Q=0.5 × Q_L+0.5×Q_R (11)。

The feature of the present invention and providing the benefit that:

By experimental result and data it can be seen that the PCC value of CB-SSIM algorithm is all more than 0.80, RMSE value all exists Less than 0.64.Compared with SSIM algorithm, the property indices of the CB-SSIM algorithm introducing the central offset factor all has different journey The raising of degree, illustrates that the central offset factor can improve the performance of stereo image quality objective evaluation；On the whole, for difference Type of distortion, PCC, KROCC and RMSE index of CB-SSIM algorithm is superior to the objective evaluation value of SSIM, CB-SSIM algorithm With can in certain degree improve stereo image quality objective evaluation accuracy.

Accompanying drawing illustrates:

Fig. 1 CB-SSIM theory diagram.

Fig. 2 nearest-neighbor interpolation algorithm principle.

8 width standard stereo images pair used by this algorithm of Fig. 3.In figure:

(a) source images Tree2 (b) source images " Family "

(c) source images " Girl " (d) source images " River "

(e) source images " Tree1 " (f) source images " Ox "

(g) source images " Tju " (h) source images " Woman ".

Detailed description of the invention

The invention provides a kind of objective evaluation method for quality of stereo images based on central offset characteristic, the present invention according to Structural similarity quality weight matrix and central offset characteristic, establish the axonometric chart of reflection subjective evaluation result accurately and effectively The objective evaluation model of picture element amount.

Below as a example by the right view of stereo-picture, basic step is as follows:

1., by structural similarity algorithm SSIM [7] using Zhou Wang to propose, calculate with reference to right image and right image The comparison function of brightness, contrast and structure, by thus draw the picture quality weight matrix of SSIM, then pass through nearest-neighbor Picture quality weight matrix is amplified to identical with original image size by interpolation algorithm.

2. according to central offset characteristic, SSIM picture quality matrix is weighted, obtains the picture quality of right image Evaluate score.

The most then repeat above-mentioned steps, calculate the image quality evaluation score of left view point, both weighted averages are obtained Stereo image quality objective evaluation value.

Each step will be carried out detailed analysis below:

1.1 structural similarity algorithms

Use the structural similarity algorithm [1] that Zhou Wang proposes.For preventing blocking effect, use M × M (M= 11), standard deviation is Gauss sliding window right viewpoint and the right side of distortion stereo pairs to original three-dimensional image pair respectively of 1.5 Viewpoint sampling obtains subimage block X and image block Y, calculates their brightness, structure and contrast similarity.

$l(X,Y)=\frac{2{\mathrm{\μ}}_X{\mathrm{\μ}}_Y+C_1}{{{\mathrm{\μ}}_X}^2+{{\mathrm{\μ}}_Y}^2+C_1}---\left(1\right)$

$s(X,Y)=\frac{2{\mathrm{\σ}}_{XY}+C_3}{{\mathrm{\σ}}_X{\mathrm{\σ}}_Y+C_3}---\left(2\right)$

$c(X,Y)=\frac{2{\mathrm{\σ}}_X{\mathrm{\σ}}_Y+C_2}{{{\mathrm{\σ}}_X}^2+{{\mathrm{\σ}}_Y}^2+C_2}---\left(3\right)$

Wherein:

${\mathrm{\σ}}_X={\left(\underset{i=1}{\overset{N}{\Σ}}{\mathrm{\ω}}_i{\left(x_i-{\mathrm{\μ}}_X\right)}^2\right)}^{1/2}---\left(5\right)$

${\mathrm{\σ}}_{XY}=\underset{i=1}{\overset{N}{\Σ}}{\mathrm{\ω}}_i(x_i-{\mathrm{\μ}}_X)(y_i-{\mathrm{\μ}}_Y)---\left(6\right)$

Wherein, C₁、C₂、C₃Representing the least normal amount, avoiding denominator is zero, and C₁=(k₁L)²、C₂=(k₂L)²、C₃ =(C₂/2)²；k₁、k₂It is the constant between [0,1] respectively；x_i,y_iIt is image block X respectively, the value of ith pixel point, μ in Y_X,μ_Y It is respectively image block X, the average of Y, σ_X,σ_YIt is respectively image block X, the variance of Y, σ_XYFor the covariance of image block X, Y, N is figure As the pixel quantity of block X or Y, l (X, Y), c (X, Y), s (X, Y) are respectively image block X, the brightness of Y, contrast and structural parameters Matrix.

The structural similarity of image block is defined as:

SSIM (x, y)=(l (X, Y))^τ(c(X,Y))^β(s(X,Y))^γ (7)

Wherein τ, beta, gamma is regulation parameter, takes τ=β=γ=1, and (x, y) is the pixel of image, formula (7) calculate sliding Structural similarity in dynamic window, SSIM (x, y) is the result that calculates behind the image upper left corner to the lower right corner of sliding window, Its size is ((W-10) × (H-10)), and wherein W and H represents horizontal pixel and the vertical pixel number of image.For so obtain The mass matrix Q of the right viewpoint of distortion stereo pairs_R(x, y), use nearest-neighbor interpolation algorithm by SSIM (x, y) be amplified to Original image size is identical.

1.2 nearest-neighbor interpolation algorithms

Nearest-neighbor interpolation algorithm [9], as a kind of simplest scaling algorithm, is suitably applied designed image scaling All spectra.Its principle is that the gray value at target pixel points is, the pixel nearest by this pixel of distance around this pixel The gray value of point determines, and it is not affected by other all of pixel.

In Fig. 2 (i, j), (i, j+1), (i+1, j), (i+1, j+1) be by floating-point coordinate (i+a, j+b) before interpolation respectively (integer part of i, j respectively denotation coordination, a, b denotation coordination the most respectively obtains fractional part, and a ∈ [0,1), b ∈ [0,1)), Four neighborhoods, f (i, j), f (i, j+1), f (i+1, j), f (i+1, j+1) be the gray value of corresponding pixel points respectively.A、B、C、 D represent respectively pixel f (i, j), f (i, j+1), f (i+1, j), f (i+1, j+1) constituted the upper left in region, upper right, a left side Under, lower right area.Nearest-neighbor difference arithmetic just determines that aforementioned four point mid-range objectives pixel (i+a, j+b) is nearest Point, then the gray value of its closest approach is exactly the gray value of target pixel points.Nearest-neighbor algorithm can represent by below equation:

$f(i+a,j+b)=\left\{\begin{array}{c}f(i,j),(a,b)\⊆A\\ f(i,j+1),(a,b)\⊆C\\ f(i+1,j),(a,b)\⊆B\\ f(i+1,j+1),(a,b)\⊆D\end{array}\right.---\left(8\right)$

Finally, the right viewpoint of distortion stereo pairs that above-mentioned nearest-neighbor interpolation algorithm will be obtained is used by SSIM algorithm Mass matrix SSIM (x, y) is amplified to original image size, and the image after now amplifying is the mass matrix of right viewpoint.

2.1 central offset characteristics

Central offset (Center Bias, CB) characteristic, refers to that human eye is invariably prone to the center from figure when watching image Beginning look for visual fixations point, then its attention is successively decreased [10] by mediad surrounding.It is to say, when the coordinate position of pixel More being in the centre position of image, this pixel is more easily subject to pay close attention to.This chapter uses has anisotropic gaussian kernel function [11] central offset (CB) factor that simulation attention is spread by mediad surrounding:

$CB(x,y)=\exp \{-(\frac{{(x-x_0)}^2}{2{\mathrm{\σ}}_h^2}+\frac{{(y-y_0)}^2}{2{\mathrm{\σ}}_v^2})\}---\left(9\right)$

Wherein (x, (x, y) to central point (x y) to represent pixel for CB₀,y₀) offset information.(x₀,y₀) represent that distortion is right The center point coordinate of viewpoint, (x y) is pixel coordinate, σ_hAnd σ_vRepresent image level direction and the standard of vertical direction respectively Difference, takes σ according to document [11]_h=1/3W, σ_v=1/3H, wherein W and H represents horizontal pixel and the vertical pixel number of image.

2.2 weighted calculation SSIM mass matrixes

Utilizing central offset model, the higher weight of region distribution that image center of adjusting the distance is nearer, in range image The weight that the region distribution farther out of heart point is relatively low.Use the Quality Map Q of the SSIM right viewpoint of algorithm calculated distortion_R(x y), uses public affairs Formula (10) obtains Objective Quality Assessment value Q of the right viewpoint of distortion_r.Same method is used to obtain the quality of distortion left view point objective Evaluation of estimate Q_l, finally use formula (11), both weighted averages obtained stereo image quality objective evaluation value.

$Q_r=\frac{\underset{x=1}{\overset{H}{\Σ}}\underset{y=1}{\overset{W}{\Σ}}{Q}_R(x,y)CB(x,y)}{\underset{x=1}{\overset{H}{\Σ}}\underset{y=1}{\overset{W}{\Σ}}CB(x,y)}---\left(10\right)$

3. stereo image quality objective evaluation value

Picture quality objective evaluation value Q of left view and right view can be respectively obtained by said method_rAnd Q_l, then Stereo image quality objective evaluation value can be obtained by weighted calculation:

Q=0.5 × Q_L+0.5×Q_R (11)

1 algorithm of table and the performance indications of SSIM algorithm

Stereo-picture selected by the design is all from broadband wireless communications and three-dimensional imaging institute image data base.From Stereo-picture storehouse choose containing personage, distant view, " Tree2 ", " Family ", " Girl " of close shot, " River ", " Tree1 ", " Ox ", " Tju ", " Woman " totally 8 undistorted standard stereo images, its resolution is 1280 × 1024.Owing to solid shows Show that equipment needs the right viewpoint of flip horizontal stereo pairs could embody third dimension, it is therefore desirable to mirror image places stereo pairs Right viewpoint figure.The subjectivity of stereo image quality is all commented by the stereo-picture in database according to International Telecommunication Union (ITU) Two standard: BT-500 and BT.1438-2000 of valency suggestion, are divided into 5 grades by all of stereo image quality: fabulous, Good, general, poor, excessively poor.

In order to true simulating stereo imaging system is to the distortion of stereo-picture and the universality of verifying this algorithm, it is right to test 8 width standard stereo images process carrying out JPEG compression distortion, Gaussian Blur distortion and Gauss white noise distortion, therefore there are 260 width distortion stereo pairs.

According to ITU-R BT.1438 standard, three-dimensional to all distortions on stereoscopic display device " 3D WINDOWS-19A0 " Image is to carrying out subjective testing, and viewing distance is 6 times of stereoscopic display device height.Test result according to all testers obtains To average suggestion value (Mean Opinion Score, MOS).Use Min-Max method for normalizing that MOS value is carried out normalizing herein Change processes, and expands to the scope value for [0,5]

$m_{i,j}=\frac{s_{i,j}-{\mathrm{Min}}_i}{{\mathrm{Max}}_i-{\mathrm{Min}}_i}---\left(12\right)$

Wherein, i represents the numbering with reference to stereo-picture, i ∈ [1,8] in the present invention.For a certain type distortion (such as JPEG distortion, Gaussian Blur distortion, white Gaussian noise distortion), s_i,jRepresent with reference to the distortion stereo-picture that stereo-picture i is corresponding The MOS value of jth kind distortion level, m_i,jRepresent m_i,jValue after Min-Max normalization.Min_iRepresent and exist with reference to stereo-picture In the case of certain type distortion, MOS value minimum in the MOS value of the stereo-picture of different strength of distortion.In like manner, according to above-mentioned former Reason normalization objective evaluation value.

The concordance of experimental result with subjective evaluation result in order to weigh the method for objectively evaluating that this chapter proposes, these selected works Take Pearson's correlation coefficient (Pearson Correlation Coefficient, PCC), Ken Deer rank order correlation coefficient (Kendall Rank Order Correlation Coefficient, KROCC) and mean square error (Root Mean Square Error, RMSE) three standards evaluate the concordance between evaluation result and the subjective evaluation result of objective algorithm, Monotonicity and accuracy.Kendall correlation coefficient is primarily used to weigh between objective algorithm evaluation and subjective evaluation result Monotonicity, this index is not to consider the relative distance between evaluation score, and weigh is the rank order between evaluation score； Pearson correlation coefficient balance is objective assessment score and MOS value dependency each other；RMSE value evaluation is objective Dispersion degree between evaluation score and subjective evaluation result i.e. accuracy.The absolute value of PCC and KROCC closer to 1, RMSE's Value, closer to 0, illustrates that objective evaluation result can effectively reflect subjective evaluation result.

Below in conjunction with technical scheme process in detail:

One, obtain evaluating data sample by subjective testing, choose training sample and test sample through repetition test.

Tested include that specialty is tested and amateur tested, be respectively provided with normal parallax third dimension, totally 20 tested, respectively In school postgraduate and undergraduate, male 11, women 9, it is engaged in tested totally 16 people of steric information treatment research, is engaged in other Tested totally 4 people of direction research.For the ease of intuitivism apprehension the design, it is provided that stereo image quality objective evaluation block diagram, as Shown in Fig. 1.

Two, by algorithm in this paper, distorted image and original image are carried out comparing calculation

1., by the structural similarity algorithm SSIM using Zhou Wang to propose, calculate with reference to the right figure of right image and the four diagnostic methods The comparison function of the brightness of picture, contrast and structure, by thus draw the picture quality weight matrix of SSIM, then pass through arest neighbors Picture quality weight matrix is amplified to identical with original image size by territory interpolation algorithm.

2. according to central offset characteristic, SSIM picture quality matrix is weighted, obtains the picture quality of right image Evaluate score.

The most then repeat above-mentioned steps, calculate the image quality evaluation score of left view point, both weighted averages are obtained Stereo image quality objective evaluation value.

By the data of table 1 it can be seen that the PCC value of CB-SSIM algorithm is all more than 0.80, RMSE value all 0.64 with Under.Compared with SSIM algorithm, introducing the property indices of CB-SSIM algorithm of the central offset factor all has carrying in various degree Height, illustrates that the central offset factor can improve the performance of stereo image quality objective evaluation；On the whole, for different distortions Type, PCC, KROCC and RMSE index of CB-SSIM algorithm be superior to the objective evaluation value of SSIM, CB-SSIM algorithm with can be Certain degree improves the accuracy of stereo image quality objective evaluation.

List of references

[1] Wang Qionghua .3d Display Technique and device [M]. Beijing: Science Press, 2011,6～8.

[2]You J,Xing L,Perkis A,et al.Perceptual quality assessment for stereoscopic images based on2D image quality metrics and disparity analysis [C].Proc.of International Workshop on Video Processing and Quality Metrics for Consumer Electronics,Scottsdale,AZ,USA.2010.

[3]Jung Y J,Kim H G,Ro Y M.Critical binocular asymmetry measure for perceptual quality assessment of synthesized stereo 3D images in view synthesis[J].Circuits and Systems for Video Technology,IEEE Transactions on, 2015,99 (3): 1～14

[4] Tian Haonan. stereo image quality evaluation [D] based on edge and local matching. Tianjin: University Of Tianjin, 2013.

[5]Shen L,Yang J,Zhang Z.Quality assessment of stereo images with stereo vision[C].Image and Signal Processing,2009.CISP'09.2nd International Congress on.IEEE,2009:1-4.

[6]Barland R,Saadane A.Blind quality metric using a perceptual importance map for jpeg-20000compressed images[C].Image Processing,2006IEEE International Conference on.IEEE,2006:2941-2944.

[7]Z.Wang,A.C.Bovik A,H.R.Sheikh,et al.Image Quality Assessment:From Error Visibility to Structural Similarity[J].IEEE Transactions on Image Processing,Vol.13,No.4,April 2004.

[8]Ciptadi A,Hermans T,Rehg J M.An in depth view of saliency[C].Eds: T.Burghardt,D.Damen,W.Mayol-Cuevas,M.Mirmehdi,In Proceedings of the British Machine Vision Conference(BMVC 2013).2013:9-13.

[9] Lin Yuan. Image Zooming Algorithm and FPGA realize [D]. Xiamen University, 2006.

[10]G.A.Ascoli,K.Svooboda,Y.Liu.Digital reconstruction of axonal and Dendritic morphology DIADEM challenge, 2010:http:.www.diademchallenge.org/

[11]Lin Y,Tang Y Y,Fang B,et al.A visual-attention model using earth mover's distance-based saliency measurement and nonlinear feature combination [J].Pattern Analysis and Machine Intelligence,IEEE Transactions on,2013,35 (2):314-328.

Claims (5)

1. a stereo image quality evaluation methodology based on central offset, is characterized in that, step is as follows:

1) use structural similarity algorithm SSIM, calculate with reference to right image and the comparison of the brightness of right image, contrast and structure Function, by thus draw the picture quality weight matrix of SSIM, then by nearest-neighbor interpolation algorithm by picture quality weights square Battle array is amplified to identical with original image size；

2) according to central offset characteristic, SSIM picture quality matrix is weighted, obtains the image quality evaluation of right image Score；

3) then repeat above-mentioned steps, calculate the image quality evaluation score of left view point, both weighted averages are obtained solid Picture quality objective evaluation value.

2. stereo image quality evaluation methodology based on central offset as claimed in claim 1, is characterized in that, uses structure phase Seemingly spending algorithm, for preventing blocking effect, using M × M, standard deviation is that the Gauss sliding window of 1.5 is respectively to original stereo figure As to right viewpoint and distortion stereo pairs right viewpoint sampling obtain subimage block X, Y, calculate their brightness, structure and Contrast similarity:

Wherein: $\begin{array}{c}l(X,Y)=\frac{2{\mathrm{\μ}}_X{\mathrm{\μ}}_Y+C_1}{{{\mathrm{\μ}}_X}^2+{{\mathrm{\μ}}_Y}^2+C_1}---\left(1\right)\\ s(X,Y)=\frac{2{\mathrm{\σ}}_{XY}+C_3}{{\mathrm{\σ}}_X{\mathrm{\σ}}_Y+C_3}---\left(2\right)\\ c(X,Y)=\frac{2{\mathrm{\σ}}_X{\mathrm{\σ}}_Y+C_2}{{{\mathrm{\σ}}_X}^2+{{\mathrm{\σ}}_Y}^2+C_2}---\left(3\right)\\ {\mathrm{\μ}}_X=\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\ω}}_i{x}_i---\left(4\right)\\ {\mathrm{\σ}}_X={\left(\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\ω}}_i{\left(x_i-{\mathrm{\μ}}_X\right)}^2\right)}^{1/2}---\left(5\right)\\ {\mathrm{\σ}}_{XY}=\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\ω}}_i(x_i-{\mathrm{\μ}}_X)(y_i-{\mathrm{\μ}}_Y)---\left(6\right)\end{array}$

Wherein, C₁、C₂、C₃Representing the least normal amount, avoiding denominator is zero, and C₁=(k₁L)²、C₂=(k₂L)²、C₃=(C₂/ 2)²；k₁、k₂It is the constant between [0,1] respectively；x_i,y_iIt is image block X respectively, the value of ith pixel point, μ in Y_X,μ_YRespectively For the average of image block X, Y, σ_X,σ_YIt is respectively image block X, the variance of Y, σ_XYFor the covariance of image block X, Y, N is image block X Or the pixel quantity of Y, l (X, Y), c (X, Y), s (X, Y) are respectively image block X, the brightness of Y, contrast and structural parameters matrix；

The structural similarity of image block is defined as:

SSIM (x, y)=(l (X, Y))^τ(c(X,Y))^β(s(X,Y))^γ (7)

Wherein τ, beta, gamma is regulation parameter, takes τ=β=γ=1, and (x, y) is the pixel of image, formula (7) calculate sliding window Structural similarity in Kou, (x, is y) result that calculates behind the image upper left corner to the lower right corner of sliding window to SSIM, and it is big Little is ((W-10) × (H-10)), and wherein W and H represents horizontal pixel and the vertical pixel number of image；For so obtain distortion The mass matrix Q of the right viewpoint of stereo pairs_R(x, y), (x y) is amplified to and artwork by SSIM to use nearest-neighbor interpolation algorithm As size is identical.

3. stereo image quality evaluation methodology based on central offset as claimed in claim 1, is characterized in that, nearest-neighbor is inserted Value-based algorithm is that the gray value at target pixel points is, by the gray value of the nearest pixel of this pixel of distance around this pixel Determine, and it is not affected by other all of pixel；

(i, j), (i, j+1), (i+1, j), (i+1, j+1) be by floating-point coordinate (i+a, j+b) before interpolation respectively (i, j be respectively The integer part of denotation coordination, a, b then distinguish denotation coordination fractional part, and a ∈ [0,1), b ∈ [0,1)), four neighbours Territory, f (i, j), f (i, j+1), f (i+1, j), f (i+1, j+1) be the gray value of corresponding pixel points respectively, A, B, C, D table respectively Show pixel f (i, j), f (i, j+1), f (i+1, j), f (i+1, j+1) constituted the upper left in region, upper right, lower-left, bottom right Region, nearest-neighbor difference arithmetic just determines that the point that aforementioned four point mid-range objectives pixel (i+a, j+b) is nearest, then its The gray value of closest approach is exactly the gray value of target pixel points, and nearest-neighbor algorithm below equation represents:

$f(i+a,j+b)=\left\{\begin{array}{c}f(i,j),(a,b)\⊆A\\ f(i,j+1),(a,b)\⊆C\\ f(i+1,j),(a,b)\⊆B\\ f(i+1,j+1),(a,b)\⊆D\end{array}\right.---\left(8\right)$

Finally, the matter of the right viewpoint of distortion stereo pairs that above-mentioned nearest-neighbor interpolation algorithm will be obtained is used by SSIM algorithm (x, y) is amplified to original image size to moment matrix SSIM, and the image after now amplifying is the mass matrix Q of right viewpoint_R(x,y)。

4. stereo image quality evaluation methodology based on central offset as claimed in claim 1, is characterized in that, uses SSIM to calculate The Quality Map Q of the right viewpoint of method calculated distortion_R(x y), uses formula (10) to obtain Objective Quality Assessment value Q of the right viewpoint of distortion_r, Same method is used to obtain Objective Quality Assessment value Q of distortion left view point_l, finally use formula (11), both weighted flat All obtain stereo image quality objective evaluation value,

$Q_r=\frac{\underset{x=1}{\overset{H}{\mathrm{\Σ}}}\underset{y=1}{\overset{W}{\mathrm{\Σ}}}{Q}_R(x,y)CB(x,y)}{\underset{x=1}{\overset{H}{\mathrm{\Σ}}}\underset{y=1}{\overset{W}{\mathrm{\Σ}}}CB(x,y)}---\left(10\right)$

There is anisotropic gaussian kernel function [11] simulation central offset (CB) factor that spread by mediad surrounding of attention:

$CB(x,y)=\exp \{-(\frac{{(x-x_0)}^2}{2{\mathrm{\σ}}_h^2}+\frac{{(y-y_0)}^2}{2{\mathrm{\σ}}_v^2})\}---\left(9\right)$

Wherein (x, (x, y) to central point (x y) to represent pixel for CB₀,y₀) offset information, (x₀,y₀) represent the right viewpoint of distortion Center point coordinate, (x y) is pixel coordinate, σ_hAnd σ_vRepresent image level direction and the standard deviation of vertical direction respectively, take σ_h =1/3W, σ_v=1/3H, wherein W and H represents horizontal pixel and the vertical pixel number of image.

5. stereo image quality evaluation methodology based on central offset as claimed in claim 1, is characterized in that, axonometric chart picture element Amount objective evaluation value: respectively obtain picture quality objective evaluation value Q of left view and right view_rAnd Q_l, then pass through weighted calculation Can obtain stereo image quality objective evaluation value:

Q=0.5 × Q_L+0.5×Q_R (11)。