CN103475897B - Adaptive image quality evaluation method based on distortion type judgment - Google Patents

Abstract

The invention discloses an adaptive image quality objective evaluation method based on the judgment of distortion type. The method firstly judges the distortion type of the image, and divides the distortion type into three types: Gaussian white noise distortion, JPEG distortion and quasi-fuzzy distortion. Class fuzzy distortion includes Gaussian blur distortion, JPEG2000 distortion and fast fading distortion; using the results of distortion discrimination, the image distorted by Gaussian white noise is evaluated by the structural similarity model based on the pixel domain, and the image distorted by JPEG is evaluated by the DCT domain Structural similarity model evaluation, using the structure similarity model evaluation based on wavelet domain for the image subject to fuzzy distortion. Implementation results show that the objective evaluation method proposed by the present invention combines the advantages of each structural similarity evaluation model well in the evaluation of distorted images of different distortion types through the distortion discrimination method, and the evaluation results are consistent with the subjective perception of human eyes. high.

Description

An Adaptive Image Quality Evaluation Method Based on Distortion Type Judgment

technical field

The invention relates to an image quality evaluation technology, in particular to an adaptive image objective quality evaluation method based on distortion type judgment.

Background technique

With the rapid development of modern communication technology, human beings have entered the information society. As an important information carrier, image quality will directly affect the accuracy and completeness of information acquisition by recipients. However, no matter in the process of image acquisition, processing, storage and transmission, it is inevitable that distortion or degradation will occur due to imperfect processing methods or irregular external equipment. Different degrees of distortion or degradation will cause image information different degrees of loss. In the case that image information technology is widely used, how to measure the loss of information contained in images caused by different degrees of distortion is particularly important. The emergence of the subject of image quality evaluation is just to solve this important practical problem. In terms of methods, image quality evaluation can be divided into subjective quality evaluation and objective quality evaluation. The former relies on the subjective perception of the experimenter to evaluate the image quality, so it is very cumbersome and time-consuming, and is not suitable for integration into practical applications; the latter is based on the model Quantitative indicators are given to simulate the human visual system (HVS, human visual system) perception mechanism to measure image quality. This method has the characteristics of simple operation, easy implementation, and real-time algorithm optimization, and has become a research focus in image quality evaluation. According to the different emphases of HVS description, image quality evaluation can be divided into two types based on error sensitivity and based on structural similarity. Based on the error sensitivity method, visual nonlinearity, multi-channel, contrast sensitivity bandpass, masking effect, interaction of different excitations between multi-channels and visual psychological characteristics are considered, but because humans have not yet fully grasped HVS, this type of method has obvious limitations obstacles to accurate modeling. Based on the structural similarity method, it is believed that natural images have a specific structure, and the human eye's perception of images is mainly extracted from these structural information to directly evaluate the structural similarity of image signals. The method has low complexity and strong applicability, but There are many types of structural similarity methods in the prior art, and these methods cannot guarantee that the quality evaluation results of images with different distortion types are relatively accurate.

Contents of the invention

The technical problem to be solved by the present invention is to provide an adaptive image quality objective evaluation method, which can effectively improve the accuracy of quality evaluation results of images with different distortion types.

The technical solution adopted by the present invention to solve the above-mentioned technical problems is: an adaptive image quality evaluation method based on distortion type judgment, and its processing process is:

First, determine the distortion type of the distorted image to be evaluated;

Secondly, perform corresponding processing in combination with the distortion type of the distorted image to be evaluated:

If the distorted image is distorted by Gaussian white noise, the original undistorted image and the distorted image to be evaluated are divided into multiple overlapping image blocks with a size of 8×8 in the pixel domain, and the original undistorted image and The brightness mean and standard deviation of all pixels in each image block in the distorted image to be evaluated in the pixel domain, and the brightness values of all the two image blocks with the same coordinate positions in the original undistorted image and the distorted image to be evaluated The covariance between the brightness values of all pixels obtains the pixel domain-based structural similarity between all two image blocks with the same coordinate positions in the original undistorted image and the distorted image to be evaluated;

If the distorted image is JPEG distortion, the original undistorted image and the distorted image to be evaluated are divided into multiple overlapping image blocks with a size of 8×8 in the DCT (Discrete Cosine Transform) domain, and the original undistorted image is calculated by calculating The mean and standard deviation of all DCT coefficients of the distorted image block and the distorted image block to be evaluated in the DCT domain, and all the two image blocks with the same coordinate position in the DCT domain of the original undistorted image and the distorted image to be evaluated The covariance between the DCT coefficients obtains the DCT domain-based structural similarity between all two image blocks with the same coordinate positions in the original undistorted image and the distorted image to be evaluated;

If the distorted image is similar to fuzzy distortion, the original undistorted image and the distorted image to be evaluated are divided into multiple overlapping image blocks with a size of 8×8 in the wavelet domain. By calculating the original undistorted image block and The mean and standard deviation of all wavelet coefficients of the distorted image block to be evaluated in the wavelet domain, and the difference between all wavelet coefficients of the original undistorted image and the distorted image to be evaluated in two image blocks with the same coordinate positions in the wavelet domain The covariance of the original undistorted image and the distorted image to be evaluated are obtained based on the wavelet domain structure similarity between two image blocks with the same coordinate position;

Finally, according to the structural similarity between the original undistorted image and the distorted image to be evaluated, the objective quality score of the distorted image to be evaluated is obtained.

A kind of adaptive image quality evaluation method based on distortion type judgment of the present invention, it specifically comprises the following steps:

①Let X denote the original undistorted image, let Y denote the distorted image to be evaluated, determine the distortion type of Y through the distortion type discrimination method, and the distortion type of Y is one of Gaussian white noise distortion, JPEG distortion, and quasi-fuzzy distortion Types, where the blur-like distortion includes Gaussian blur distortion, JPEG2000 distortion and fast fading distortion;

②If the distortion type of the distorted image Y is Gaussian white noise distortion, then use a sliding window with a size of 8×8 to move pixel by pixel in X, and divide X into M×N overlapping ones with a size of 8× 8 image blocks, the image block whose coordinate position in X is (i,j) is recorded as x _i,j ; similarly, a sliding window with a size of 8×8 is used to move pixel by pixel in Y, and Y is divided into M×N overlapping image blocks with a size of 8×8, the image block whose coordinate position in Y is (i,j) is recorded as y _i,j ; where, H represents the height of X and Y, W represents the width of X and Y, the symbol is the symbol of rounding down, 1≤i≤M, 1≤j≤N;

If the distortion type of the distorted image Y is JPEG distortion, use a sliding window with a size of 8×8 to move pixel by pixel in X, and divide X into M×N overlapping images with a size of 8×8 block, the image block whose coordinate position is (i, j) in X is denoted as x _{i, j} , perform two-dimensional DCT transformation on all image blocks x _{i, j} , and the corresponding transformed image block is Similarly, a sliding window with a size of 8×8 is used to move pixel by pixel in Y, and Y is divided into M×N overlapping image blocks with a size of 8×8, and the coordinate position in Y is (i ,j) is denoted as y _i,j , perform two-dimensional DCT transformation on all image blocks y _i,j, and obtain the corresponding transformed image block as in, H represents the height of X and Y, W represents the width of X and Y, 1≤i≤M, 1≤j≤N;

If the distortion type of the distorted image Y is fuzzy-like distortion, perform a first-level wavelet transform on X, extract the approximate component and record it as X _A , use a sliding window with a size of 8×8 to move point by point in X _A , and convert X _A is divided into M'×N' overlapping image blocks with a size of 8×8, and the image block whose coordinate position in X _A is (i',j') is recorded as Similarly, perform a first-level wavelet transform on Y, extract the approximate component and record it as Y _A , use a sliding window with a size of 8×8 to move point by point in Y _A , and divide Y _A into M'×N' overlapping and the size of the image block is 8×8, and the image block whose coordinate position in Y _A is (i', j') is recorded as in, H' means the height of X _A and Y _A , W' means the width of X _A and Y _A , 1≤i'≤M', 1≤j'≤N';

③If the distortion type of the distorted image Y is Gaussian white noise distortion, then calculate the brightness mean and standard deviation of all pixels in each image block in X, and calculate the brightness value and standard deviation of all pixels in each image block in Y The brightness mean and standard deviation, and then calculate the covariance between all pixels in the two image blocks with the same coordinate position in X and Y, and the image block x _{i with the coordinate position (i, j) in X,} The brightness mean and standard deviation of all pixels in _j are recorded as and The brightness mean and standard deviation of all the pixels in the image block y _{i, j} at the coordinate position (i, j) in Y are recorded as and The correlation between all the pixels in the image block x _i,j whose coordinate position is (i,j) in X and all the pixels in the image block y _i,j whose coordinate position is (i,j) in Y is Variance is recorded as ${\mathrm{\σ}}_{x_{i,j}}=\sqrt{\frac{1}{64}\underset{u=1}{\overset{8}{\mathrm{\Σ}}}\underset{v=1}{\overset{8}{\mathrm{\Σ}}}{(x_{i,j}(u,v)-{\mathrm{\μ}}_{x_{i,j}})}^2},{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i,j}}=\frac{1}{64}\underset{u=1}{\overset{8}{\mathrm{\Σ}}}\underset{v=1}{\overset{8}{\mathrm{\Σ}}}{\mathrm{the\; y}}_{i,j}(u,v),$ ${\mathrm{\σ}}_{{\mathrm{the\; y}}_{i,j}}=\sqrt{\frac{1}{64}\underset{u=1}{\overset{8}{\mathrm{\Σ}}}\underset{v=1}{\overset{8}{\mathrm{\Σ}}}{({\mathrm{the\; y}}_{i,j}(u,v)-{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i,j}})}^2},$ ${\mathrm{\σ}}_{x_{i,j}{\mathrm{the\; y}}_{i,j}}=\frac{1}{64}\underset{u=1}{\overset{8}{\mathrm{\Σ}}}\underset{v=1}{\overset{8}{\mathrm{\Σ}}}\left[(x_{i,j}(u,v)-{\mathrm{\μ}}_{x_{i,j}})({\mathrm{the\; y}}_{i,j}(u,v)-{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i,j}})\right],$ Among them, x _i,j (u,v) represents the brightness value of the pixel point whose coordinate position in x _i,j is (u,v), and y _i,j (u,v) represents the coordinate position in y _i,j as The brightness value of the pixel point of (u,v), 1≤u≤8, 1≤v≤8;

If the distortion type of the distorted image Y is JPEG distortion, calculate the brightness mean and standard deviation of all pixels in each image block in X, and calculate the brightness mean and standard deviation of all pixels in each image block in Y Standard deviation, then calculate the mean and standard deviation of the DCT AC coefficients of each image block in X, and calculate the mean and standard deviation of the DCT AC coefficients of each image block in Y, and finally calculate all coordinates in X and Y The covariance between all DCT AC coefficients of two image blocks with the same position, and the brightness mean and standard deviation of all pixels in the image block x _{i, j} whose coordinate position is (i, j) in X are recorded as and The brightness mean and standard deviation of all the pixels in the image block y _{i, j} at the coordinate position (i, j) in Y are recorded as and A new image block obtained by performing DCT transformation on the image block x _{i, j} whose coordinate position is (i, j) in X The mean and standard deviation of all the AC coefficients of are recorded as and A new image block obtained by performing DCT transformation on the image block y _{i, j} whose coordinate position is (i, j) in Y The mean and standard deviation of all the AC coefficients of are recorded as and The DCT domain image block whose coordinate position in X is (i, j) and the DCT domain image block with the coordinate position (i,j) in Y The covariance between all AC coefficients in is denoted as ${\mathrm{\σ}}_{x_{i,j}}=\sqrt{\frac{1}{64}\underset{u=1}{\overset{8}{\mathrm{\Σ}}}\underset{v=1}{\overset{8}{\mathrm{\Σ}}}{(x_{i,j}(u,v)-{\mathrm{\μ}}_{x_{i,j}})}^2},{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i,j}}=\frac{1}{64}\underset{u=1}{\overset{8}{\mathrm{\Σ}}}\underset{v=1}{\overset{8}{\mathrm{\Σ}}}{\mathrm{the\; y}}_{i,j}(u,v),$ ${\mathrm{\σ}}_{{\mathrm{the\; y}}_{i,j}}=\sqrt{\frac{1}{64}\underset{u=1}{\overset{8}{\mathrm{\Σ}}}\underset{v=1}{\overset{8}{\mathrm{\Σ}}}{({\mathrm{the\; y}}_{i,j}(u,v)-{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i,j}})}^2},{\mathrm{\μ}}_{x_{i,j}^{\mathrm{D.}}}=\frac{1}{64}\underset{u^{\mathrm{D.}}=1}{\overset{8}{\mathrm{\Σ}}}\underset{v^{\mathrm{D.}}=1}{\overset{8}{\mathrm{\Σ}}}{x}_{i,j}^{\mathrm{D.}}(u^{\mathrm{D.}},v^{\mathrm{D.}}),$ ${\mathrm{\σ}}_{x_{i,j}^{\mathrm{D.}}}=\sqrt{\frac{1}{64}\underset{u^{\mathrm{D.}}=1}{\overset{8}{\mathrm{\Σ}}}\underset{v^{\mathrm{D.}}=1}{\overset{8}{\mathrm{\Σ}}}{(x_{i,j}^{\mathrm{D.}}(u^{\mathrm{D.}},v^{\mathrm{D.}})-{\mathrm{\μ}}_{x_{i,j}^{\mathrm{D.}}})}^2},{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i,j}^{\mathrm{D.}}}=\frac{1}{64}\underset{u^{\mathrm{D.}}=1}{\overset{8}{\mathrm{\Σ}}}\underset{v^{\mathrm{D.}}=1}{\overset{8}{\mathrm{\Σ}}}{\mathrm{the\; y}}_{i,j}^{\mathrm{D.}}(u^{\mathrm{D.}},v^{\mathrm{D.}}),$ ${\mathrm{\σ}}_{{\mathrm{the\; y}}_{i,j}^{\mathrm{D.}}}=\sqrt{\frac{1}{64}\underset{u^{\mathrm{D.}}=1}{\overset{8}{\mathrm{\Σ}}}\underset{v^{\mathrm{D.}}=1}{\overset{8}{\mathrm{\Σ}}}{({\mathrm{the\; y}}_{i,j}^{\mathrm{D.}}(u^{\mathrm{D.}},v^{\mathrm{D.}})-{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i,j}^{\mathrm{D.}}})}^2},$ ${\mathrm{\σ}}_{x_{i,j}^{\mathrm{D.}},{\mathrm{the\; y}}_{i,j}^{\mathrm{D.}}}=\frac{1}{64}\underset{u^{\mathrm{D.}}=1}{\overset{8}{\mathrm{\Σ}}}\underset{v^{\mathrm{D.}}=1}{\overset{8}{\mathrm{\Σ}}}\left[(x_{i,j}^{\mathrm{D.}}(u^{\mathrm{D.}},v^{\mathrm{D.}})-{\mathrm{\μ}}_{x_{i,j}^{\mathrm{D.}}})({\mathrm{the\; y}}_{i,j}^{\mathrm{D.}}(u^{\mathrm{D.}},v^{\mathrm{D.}})-{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i,j}^{\mathrm{D.}}})\right],$ Among them, x _i,j (u,v) represents the brightness value of the pixel point whose coordinate position in x _i,j is (u,v), and y _i,j (u,v) represents the coordinate position in y _i,j as The brightness value of the pixel point of (u,v), 1≤u≤8, 1≤v≤8, express The DCT coefficient value at the coordinate position (u ^D , v ^D ), express The DCT coefficient value at the middle coordinate position (u ^D , v ^D ), 1≤u ^D ≤8, 1≤v ^D ≤8 and u ^D and v ^D are not 1 at the same time;

If the distortion type of the distorted image Y is fuzzy-like distortion, calculate the mean and standard deviation of all coefficient values in the approximate component X _A of X after the first-level wavelet transformation, and calculate the approximate component Y _A of Y after the first-level wavelet transformation The mean and standard deviation of all coefficient values in X _{A and Y A, and then calculate the covariance between all coefficients in the two image blocks with the same coordinate position in X A} and Y _A , and the coordinate position in X _A is (i ^W ,j ^W ) image block The means and standard deviations of all coefficients in and The image block whose coordinate position is (i ^W , j ^W ) in Y _A The mean and standard deviation of all coefficients in and The image block whose coordinate position is (i ^W , j ^W ) in X _A All the pixels in Y _A and the image block whose coordinate position is (i ^W , j ^W ) The covariance among all the coefficients in is denoted as ${\mathrm{\μ}}_{x_{i^W,j^W}^W}=\frac{1}{64}\underset{u^W=1}{\overset{8}{\mathrm{\Σ}}}\underset{v^W=1}{\overset{8}{\mathrm{\Σ}}}{x}_{i^W,j^W}^W(u^W,v^W),{\mathrm{\σ}}_{x_{i^W,j^W}^W}=\sqrt{\frac{1}{64}\underset{u^W=1}{\overset{8}{\mathrm{\Σ}}}\underset{v^W=1}{\overset{8}{\mathrm{\Σ}}}{(x_{i^W,j^W}^W(u^W,v^W)-{\mathrm{\μ}}_{x_{i^W,j^W}^W})}^2},$ ${\mathrm{\μ}}_{{\mathrm{the\; y}}_{i^W,j^W}^W}=\frac{1}{64}\underset{u^W=1}{\overset{8}{\mathrm{\Σ}}}\underset{v^W=1}{\overset{8}{\mathrm{\Σ}}}{\mathrm{the\; y}}_{i^W,j^W}^W(u^W,v^W),{\mathrm{\σ}}_{{\mathrm{the\; y}}_{i^W,j^W}^W}=\sqrt{\frac{1}{64}\underset{u^W=1}{\overset{8}{\mathrm{\Σ}}}\underset{v^W=1}{\overset{8}{\mathrm{\Σ}}}{({\mathrm{the\; y}}_{i^W,j^W}^W(u^W,v^W)-{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i^W,j^W}^W})}^2},$ ${\mathrm{\σ}}_{x_{i^W,j^W}^W,{\mathrm{the\; y}}_{i^W,j^W}^W}=\frac{1}{64}\underset{u^W=1}{\overset{8}{\mathrm{\Σ}}}\underset{v^W=1}{\overset{8}{\mathrm{\Σ}}}\left[(x_{i^W,j^W}^W(u^W,v^W)-{\mathrm{\μ}}_{x_{i^W,j^W}^W})({\mathrm{the\; y}}_{i^W,j^W}^W(u^W,v^W)-{\mathrm{\μ}}_{u_{i^W,j^W}^W})\right],$ in, express The coefficient value at the coordinate position (u ^W , v ^W ), express The coefficient value at the middle coordinate position (u ^W , v ^W ), 1≤u ^W ≤8, 1≤v ^W ≤8;

④ If the distortion type of the distorted image Y is Gaussian white noise distortion, then calculate the brightness function, contrast function and structure function between all the coordinate positions in X and Y between the same two image blocks, and set the coordinate position in X as ( The brightness function, contrast function and structure function between the image block x _i,j of i,j) and the image block y _i,j whose coordinate position is (i,j) in Y are respectively denoted as l(xi _{, j} ,y _i,j ), c(x _i,j ,y _i,j ) and s(x _i,j ,y _i,j ), $l(x_{i,j},{\mathrm{the\; y}}_{i,j})=\frac{2{\mathrm{\μ}}_{x_{i,j}}{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i,j}}+C_1}{{\mathrm{\μ}}_{x_{i,j}}^2+{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i,j}}^2+C_1},c(x_{i,j},{\mathrm{the\; y}}_{i,j})=\frac{2{\mathrm{\σ}}_{x_{i,j}}{\mathrm{\σ}}_{{\mathrm{the\; y}}_{i,j}}+C_2}{{\mathrm{\σ}}_{x_{i,j}}^2+{\mathrm{\σ}}_{{\mathrm{the\; y}}_{i,j}}^2+C_2},$ Among them, C ₁ , C ₂ , and C ₃ are small-value constants set to prevent the denominator from appearing zero;

If the distortion type of the distorted image Y is JPEG distortion, calculate the brightness function and contrast function between two image blocks with the same coordinate positions in X and Y, and calculate the two image blocks with the same coordinate positions in X and Y The structure function of the image block in the DCT domain, the brightness between the image block x i _{, j at the coordinate position (i, j) in X and the image block y i, j} at the coordinate position ₍ i, j) in Y function and contrast function are recorded as l(x _i,j ,y _i,j ) and c(x _i,j ,y _i,j ) respectively, and the image block x _i whose coordinate position in X is (i,j) _{, j} is the new image block after DCT transformation The new image block after DCT transformation with the image block y _{i, j} whose coordinate position is (i, j) in Y The structure degree function between is denoted as f(x _i,j ,y _i,j ), $c(x_{i,j},{\mathrm{the\; y}}_{i,j})=\frac{2{\mathrm{\σ}}_{x_{i,j}}{\mathrm{\σ}}_{{\mathrm{the\; y}}_{i,j}}+C_2}{{\mathrm{\σ}}_{x_{i,j}}^2+{\mathrm{\σ}}_{{\mathrm{the\; y}}_{i,j}}^2+C_2},f(x_{i,j},{\mathrm{the\; y}}_{i,j})=\frac{2{\mathrm{\σ}}_{x_{i,j}^{\mathrm{D.}}}{\mathrm{\σ}}_{{\mathrm{the\; y}}_{i,j}^{\mathrm{D.}}}+C_3}{{\mathrm{\σ}}_{x_{i,j}^{\mathrm{D.}}}^2+{\mathrm{\σ}}_{{\mathrm{the\; y}}_{i,j}^{\mathrm{D.}}}^2+C_3},$ Among them, C ₁ , C ₂ , and C ₃ are small-value constants set to prevent the denominator from appearing zero;

If the distortion type of the distorted image Y is fuzzy-like distortion, calculate the wavelet coefficient brightness function, wavelet coefficient contrast function and wavelet coefficient structure degree function between all the coordinate positions in X and Y between the two image blocks, and convert X The wavelet coefficient brightness function, the wavelet coefficient contrast function and the wavelet coefficient structure degree between the image block x _{i, j} at the coordinate position (i, j) and the image block y _{i, j at} the coordinate position (i, j) in Y The corresponding functions are denoted as l ^W (xi _,j ,y _i,j ), c ^W (xi _,j ,y _i,j ) and s ^W (xi _,j ,y _i,j ), $l^W(x_{i,j},{\mathrm{the\; y}}_{i,j})=\frac{2{\mathrm{\μ}}_{x_{i^W,j^W}^W}{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i^W,j^W}^W}+C_1}{{\mathrm{\μ}}_{x_{i^W,j^W}^W}^2+{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i^W,j^W}^W}^2+C_1},c^W(x_{i,j},{\mathrm{the\; y}}_{i,j})=\frac{2{\mathrm{\σ}}_{x_{i^W,j^W}^W}{\mathrm{\σ}}_{{\mathrm{the\; y}}_{i^W,j^W}^W}+C_2}{{\mathrm{\σ}}_{x_{i^W,j^W}^W}^2+{\mathrm{\σ}}_{{\mathrm{the\; y}}_{i^W,j^W}^W}^2+C_2},$ Among them, C ₁ , C ₂ , and C ₃ are small-value constants set to avoid zero in the denominator;

⑤If the distortion type of the distorted image Y is Gaussian white noise distortion, then calculate all the values in X and Y according to the brightness function, contrast function and structure function between two image blocks with the same coordinate positions in X and Y. The structural similarity between two image blocks with the same coordinate position, the image block x i _{, j with the coordinate position (i, j) in X and the image block y i, j} with the coordinate position (i, j) in Y _, The structural similarity between _j is recorded as SSIM(xi _,j ,y _i,j ), SSIM(xi _,j ,y _i,j )=[l(xi _,j ,y _i,j )] ^α [c(xi _,j ,y _i,j )] ^β [s(xi _,j ,y _i,j )] ^γ , where α, β and γ are adjustment factors;

If the distortion type of the distorted image Y is JPEG distortion, according to the brightness function and contrast function between two image blocks with the same coordinate positions in X and Y, and two images with the same coordinate positions in X and Y The structure degree function of the block in the DCT domain, calculate the structural similarity based on the DCT domain between two image blocks with the same coordinate position in X and Y, and take the image block x _i with the coordinate position (i, j) in X _,j and the image block y i, _j whose coordinate position is (i,j) in Y is denoted as FSSIM(xi _,j ,y _i,j ), FSSIM(xi _{,j j} ,y _i,j )=[l(x _i,j ,y _i,j )] ^α [c(x _i,j ,y _i,j )] ^β [f(x _{i,j ,} y _i,j )] ^γ , where α, β and γ are regulator factors;

If the distortion type of the distorted image Y is fuzzy-like distortion, then calculate X and The structural similarity based on the wavelet domain between all the two image blocks with the same coordinate position in Y, the image block x _i,j whose coordinate position is (i,j) in X and the coordinate position in Y are (i,j) The wavelet domain-based structural similarity between image blocks y _i,j of j) is denoted as WSSIM(xi _,j ,y _i,j ), WSSIM(xi _,j ,y _i,j )=[l ^W (xi _,j ,y _i,j )] ^α [c ^W (xi _,j ,y _i,j )] ^β [s ^W (xi _,j ,y _i,j )] ^γ , where α, β and γ are adjustment factors;

⑥ If the distortion type of the distorted image Y is Gaussian white noise distortion, then calculate the objective quality score of Y according to the structural similarity based on the pixel domain between all the two image blocks with the same coordinate positions in X and Y, and record is _Qwn , $Q_{\mathrm{wn}}=\frac{1}{m\×N}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}\mathrm{SSIM}(x_{i,j},{\mathrm{the\; y}}_{i,j});$

If the distortion type of the distorted image Y is JPEG distortion, then calculate the objective quality score of Y according to the structural similarity based on the DCT domain between two image blocks with the same coordinate positions in X and Y, and denote it as Q _jpeg , $Q_{\mathrm{jpeg}}=\frac{1}{m\×N}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}\mathrm{FSSIM}(x_{i,j},{\mathrm{the\; y}}_{i,j});$

If the distortion type of the distorted image Y is fuzzy-like distortion, then calculate the objective quality score of Y according to the structural similarity based on the wavelet domain between two image blocks with the same coordinate positions in X and Y, and denote it as Q _blur , $Q_{\mathrm{blurred}}=\frac{1}{m\×N}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}\mathrm{WSSIM}(x_{i,j},{\mathrm{the\; y}}_{i,j}).$

In the described step 1., the specific process of determining the distortion type of Y by the distortion type discrimination method is:

①-a. Carry out non-overlapping block segmentation with a size of 64×64 on X to obtain M’×N’ image blocks with a size of 64×64, and divide the image block whose coordinates are at (i’, j’) in X Denoted as x'_i',j' , perform a first-level wavelet decomposition on each image block x'_i',j', extract the diagonal components, and find the median value of the coefficient magnitude in the diagonal components of each image block, And calculate the noise standard deviation of each image block, and record the median value of the coefficient magnitude in the wavelet diagonal component of x'_i',j' as Its noise standard deviation is ${\mathrm{\σ}}_{{x^{\′}}_{i^{\′},j^{\′}}}=\frac{{\mathrm{MED}}_{{x^{\′}}_{i^{\′},j^{\′}}}}{0.6745},$ in, 1≤i'≤M', 1≤j'≤N';

Similarly, Y is divided into non-overlapping blocks with a size of 64×64 to obtain M'×N' image blocks with a size of 64×64, and the image block whose coordinates in Y are at (i', j') is recorded as y'_i',j' , perform a first-level wavelet decomposition on each image block y'_i',j' , extract the diagonal components, find out the median value of the coefficient magnitude in the diagonal components of each image block, and calculate The noise standard deviation of each image block, the median value of the magnitude of the coefficient in the wavelet diagonal component of y'_i',j' is recorded as Its noise standard deviation is ${\mathrm{\σ}}_{{{\mathrm{the\; y}}^{\′}}_{i^{\′},j^{\′}}}=\frac{{\mathrm{MED}}_{{{\mathrm{the\; y}}^{\′}}_{i^{\′},j^{\′}}}}{0.6745};$

①-b. Calculate the difference between the noise standard deviations between all image blocks with the same coordinate position in X and Y, and divide the image block x'i', j at the coordinate position (i', j _' ) in X and Y _' and y'_i',j', the noise standard deviation difference is recorded as △σ _{i', j'} , Then calculate the mean value of the noise standard deviation difference between all image blocks with the same coordinate position in X and Y, denoted as

①-c, judgment Whether it is established, if established, then determine that the distortion type of Y is Gaussian white noise distortion, and then end; otherwise, perform steps ①-d; wherein, Th _WN is the discrimination threshold of Gaussian white noise distortion;

①-d. Calculate the luminance difference map of X, which is denoted as X ^h , and the coefficient value of the point whose coordinate position is (i″, j″) in X ^h is denoted as X ^h (i″, j″), and X ^h ( i″,j″)=|X(i″,j″)-X(i″,j″+1)|, where, 1≤i″≤H, 1≤j″≤W-1, X(i ″, j″) indicates the brightness value of the pixel whose coordinate position in X is (i″, j″), and X(i″, j″+1) indicates that the coordinate position in X is (i″, j″+1) The brightness value of the pixel, the symbol "||" is the absolute value symbol;

Similarly, calculate the luminance difference map of Y, which is recorded as Y ^h , and the coefficient value of the point whose coordinate position is (i″, j″) in Y ^h is recorded as Y ^h (i″, j″), and Y ^h (i″ ,j″)=|Y(i″,j″)-Y(i″,j″+1)|, where, 1≤i″≤H, 1≤j″≤W-1, Y(i″, j″) represents the brightness value of the pixel whose coordinate position in Y is (i″, j″), and Y(i″, j″+1) represents the pixel whose coordinate position in Y is (i″, j″+1) The brightness value of the point;

①-e, perform 8×8 non-overlapped block segmentation on the luminance difference map X ^h of X, and obtain M″×N” non-overlapping size 8×8 image blocks, and set the coordinate position in X ^h at The image block at (i″’, j″’) is denoted as define image blocks The energy inside the block and the energy at the edge of the block are respectively and ${\mathrm{Ex}}_{i^{\′\′\′},j^{\′\′\′}}^{\mathrm{In}}=\frac{1}{56}\underset{p=1}{\overset{8}{\mathrm{\Σ}}}\underset{q=1}{\overset{7}{\mathrm{\Σ}}}{x}_{i^{\′\′\′},j^{\′\′\′}}^h(p,q),{\mathrm{Ex}}_{i^{\′\′\′},j^{\′\′\′}}^{\mathrm{Ed}}=\frac{1}{8}\underset{p=1}{\overset{8}{\mathrm{\Σ}}}{x}_{i^{\′\′\′},j^{\′\′\′}}^h(p,8),$ in, The coefficient value at the coordinate position of (p,q), The coefficient value at the coordinate position (p,8), 1≤i″'≤M″, 1≤j″'≤N″, 1≤p≤8, 1≤q≤7;

Similarly, carry out 8 × 8 non-overlapped block divisions to the luminance difference map Y ^h of Y to obtain M″ × N” non-overlapping size 8 × 8 image blocks, and set the coordinate position in Y ^h at (i ″', j″'), the image block at the position is denoted as define image blocks The energy inside the block and the energy at the edge of the block are respectively and ${\mathrm{Ey}}_{i^{\′\′\′},j^{\′\′\′}}^{\mathrm{In}}=\frac{1}{56}\underset{p=1}{\overset{8}{\mathrm{\Σ}}}\underset{q=1}{\overset{7}{\mathrm{\Σ}}}{\mathrm{the\; y}}_{i^{\′\′\′},j^{\′\′\′}}^h(p,q),{\mathrm{Ey}}_{i^{\′\′\′},j^{\′\′\′}}^{\mathrm{Ed}}=\frac{1}{8}\underset{p=1}{\overset{8}{\mathrm{\Σ}}}{\mathrm{the\; y}}_{i^{\′\′\′},j^{\′\′\′}}^h(p,8),$ in, The coefficient value at the coordinate position of (p,q), The coefficient value whose middle coordinate position is (p,8);

①-f, calculate the ratio between the edge energy of all image blocks in X ^h and the energy in the block, and the image block whose coordinate position in X ^h is at (i″’, j″’) The ratio of block edge energy to block energy is denoted as $R_{i^{\′\′\′},j^{\′\′\′}}^x=\frac{{\mathrm{Ex}}_{i^{\′\′\′},j^{\′\′\′}}^{\mathrm{Ed}}}{\mathrm{E.}{x}_{i^{\′\′\′},j^{\′\′\′}}^{\mathrm{In}}};$

Similarly, calculate the ratio between the edge energy of all image blocks in Y ^h and the energy in the block, and the image block whose coordinate position in Y ^h is at (i″’, j″’) The ratio of block edge energy to block energy is denoted as $R_{i^{\′\′\′},j^{\′\′\′}}^{\mathrm{the\; y}}=\frac{{\mathrm{Ey}}_{i^{\′\′\′},j^{\′\′\′}}^{\mathrm{Ed}}}{\mathrm{E.}{\mathrm{the\; y}}_{i^{\′\′\′},j^{\′\′\′}}^{\mathrm{In}}};$

Statistics satisfy the inequality The number of image blocks, and recorded as N ₀ ; define the discriminant index J, $J=\frac{N_0}{m^{\′\′}\×N^{\′\′}};$

①-g, judging whether J>Th _JPEG is established, if established, then determine that the distortion type of Y is JPEG distortion, and then end; otherwise, perform step ①-h; wherein, Th _JPEG is the JPEG distortion discrimination threshold;

①-h. Determine that the distortion type of Y is fuzzy-like distortion, that is, the distortion type of Y is Gaussian blur distortion, or JPEG2000 distortion, or fast fading distortion.

In step ④, C ₁ =0.01, C ₂ =0.02, C ₃ =0.01, α=β=γ=1.

Compared with the prior art, the present invention has the advantages of:

1) The method of the present invention combines the distortion type of the distorted image to be evaluated when obtaining the structural similarity between two image blocks with the same coordinate positions in the original undistorted image and the distorted image to be evaluated, so that the method of the present invention From the perspective of adaptive evaluation, it can adaptively choose to calculate the structural similarity based on the pixel domain, or the structural similarity based on the DCT domain, or the structural similarity based on the wavelet domain, thereby improving the quality of images with different distortion types The accuracy of the evaluation results.

2) The method of the present invention is in the process of discriminating the distortion type of distorted image by combining the distortion characteristics shown when the image is distorted by Gaussian white noise and the distortion characteristics shown when it is distorted by JPEG, in the case of the original reference image, realize The judgment of the distortion type of the distorted image is realized, and the discriminating process of the distortion type has high portability.

Description of drawings

Fig. 1 is the overall realization block diagram of the inventive method;

Fig. 2 is 12 pieces of original undistorted images in the training set involved in the method of the present invention;

Fig. 3 is the relationship figure between the threshold value and the judgment correct rate when determining that the image is Gaussian white noise distortion in the method of the present invention;

Fig. 4 is the relationship figure between the threshold size and the judgment correct rate when determining that the image is JPEG distortion in the inventive method;

Detailed ways

The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments.

The present invention proposes an adaptive image quality evaluation method based on distortion type judgment, and its processing process is as follows:

First, determine the distortion type of the distorted image to be evaluated;

Secondly, corresponding processing is performed in combination with the distortion type of the distorted image to be evaluated,

If the distorted image is distorted by Gaussian white noise, the original undistorted image and the distorted image to be evaluated are divided into multiple overlapping image blocks with a size of 8×8 in the pixel domain, and the original undistorted image and All pixels in each image block in the distorted image to be evaluated have the mean and standard deviation of brightness in the pixel domain, and all the two image blocks with the same coordinate positions in the original undistorted image and the distorted image to be evaluated The covariance between pixel brightness values obtains the pixel domain-based structural similarity between all two image blocks with the same coordinate positions in the original undistorted image and the distorted image to be evaluated;

If the distorted image is JPEG distortion, the original undistorted image and the distorted image to be evaluated are divided into multiple overlapping image blocks with a size of 8×8 in the DCT (Discrete Cosine Transform) transform domain, and the original undistorted image is calculated by calculating The mean and standard deviation of all coefficients of the undistorted image block and the distorted image block to be evaluated in the DCT domain, and the original undistorted image and the distorted image to be evaluated have the same coordinate positions in the two image blocks in the DCT domain The covariance between all coefficients of the original undistorted image and the distorted image to be evaluated are obtained based on the structure similarity of the DCT domain between two image blocks with the same coordinate positions;

If the distorted image is similar to fuzzy distortion, the original undistorted image and the distorted image to be evaluated are divided into multiple overlapping image blocks with a size of 8×8 in the wavelet domain. By calculating the original undistorted image block and The mean and standard deviation of all coefficients of the distorted image block to be evaluated in the wavelet domain, and the coefficients between all the coefficients of the original undistorted image and the distorted image to be evaluated in the wavelet domain where all coordinate positions are the same Covariance, to obtain the wavelet domain-based structural similarity between all two image blocks with the same coordinate positions in the original undistorted image and the distorted image to be evaluated;

Finally, according to the structural similarity between the original undistorted image and the distorted image to be evaluated, the objective quality score of the distorted image to be evaluated is obtained.

The self-adaptive image quality objective evaluation method of the present invention, its overall realization block diagram is as shown in Figure 1, and it specifically comprises the following steps:

①Let X denote the original undistorted image, let Y denote the distorted image to be evaluated, and then determine the distortion type of Y through the distortion type discrimination method, and the distortion type of Y is Gaussian white noise distortion, JPEG distortion, Gaussian blur distortion, JPEG2000-like One of the distortions.

At present, image distortion types generally include Gaussian white noise distortion (WN, white noise), JPEG distortion (JPEG) and class blur distortion, wherein, class blur distortion includes Gaussian blur distortion (Gblur, Gaussian blur), JPEG2000 distortion and There are three kinds of fast fading distortion (FF, fast fading). Here, the distortion type of Y is determined by the method of discriminating the distortion type.

The specific process of determining the distortion type of Y through the distortion type discrimination method in step ① is as follows:

①-a. Carry out non-overlapping block segmentation with a size of 64×64 on X to obtain M’×N’ image blocks with a size of 64×64, and divide the image block whose coordinates are at (i’, j’) in X denoted as x'_i',j' , perform a first-level wavelet decomposition (haar wavelet) on each image block x', extract the diagonal components, find out the median value of the coefficient magnitude in the diagonal components of each image block, and Calculate the noise standard deviation of each image block, and record the median value of the magnitude of the coefficient in the wavelet diagonal component of x'_i',j' as Its noise standard deviation is ${\mathrm{\σ}}_{{x^{\′}}_{i^{\′},j^{\′}}}=\frac{{\mathrm{MED}}_{{x^{\′}}_{i^{\′},j^{\′}}}}{0.6745},$ in, 1≤i'≤M', 1≤j'≤N', H and W represent the height and width of X respectively;

Similarly, Y is divided into non-overlapping blocks with a size of 64×64 to obtain M'×N' image blocks with a size of 64×64, and the image block whose coordinates are at (i', j') in Y is recorded as y'_i',j' , perform a first-level wavelet decomposition (haar wavelet) on each image block y', extract the diagonal components, find out the median value of the coefficient magnitude in the diagonal components of each image block, and calculate each The noise standard deviation of an image block, the median value of the coefficient amplitude in the wavelet diagonal component of y'_i',j' is recorded as Its noise standard deviation is ${\mathrm{\σ}}_{{{\mathrm{the\; y}}^{\′}}_{i^{\′},j^{\′}}}=\frac{{\mathrm{MED}}_{{{\mathrm{the\; y}}^{\′}}_{i^{\′},j^{\′}}}}{0.6745},$ in, 1≤i'≤M', 1≤j'≤N', H and W represent the height and width of Y respectively, that is, X and Y have the same size;

①-b. Calculate the difference between the noise standard deviations between all image blocks with the same coordinate position in X and Y, and divide the image block x'i', j at the coordinate position (i', j _' ) in X and Y _' and y'_i',j', the noise standard deviation difference is recorded as △σ _{i', j'} , Then calculate the mean value of the noise standard deviation difference between all image blocks with the same coordinate position in X and Y, denoted as

①-c, judgment Whether it is established, if established, then determine that the distortion type of Y is Gaussian white noise distortion, and then end; otherwise, perform steps ①-d; wherein, Th _WN is a Gaussian white noise distortion discrimination threshold, in this embodiment, Gaussian white noise The value of the discrimination threshold Th _WN is 0.8;

①-d. Calculate the luminance difference map of X, which is denoted as X ^h , and the coefficient value of the point whose coordinate position is (i″, j″) in X ^h is denoted as X ^h (i″, j″), and X ^h ( i″,j″)=|X(i″,j″)-X(i″,j″+1)|, where, 1≤i″≤H, 1≤j″≤W-1, X(i ″, j″) indicates the brightness value of the pixel whose coordinate position in X is (i″, j″), and X(i″, j″+1) indicates that the coordinate position in X is (i″, j″+1) The brightness value of the pixel, the symbol "||" is the absolute value symbol;

Similarly, calculate the luminance difference map of Y, which is recorded as Y ^h , and the coefficient value of the point whose coordinate position is (i″, j″) in Y ^h is recorded as Y ^h (i″, j″), and Y ^h (i″ ,j″)=|Y(i″,j″)-Y(i″,j″+1)|, where, 1≤i″≤H, 1≤j″≤W-1, Y(i″, j″) represents the brightness value of the pixel whose coordinate position is (i″, j″) in Y, and Y(i″, j″+1) represents the pixel whose coordinate position is (i″, j″+1) in Y The brightness value of the point;

①-e, perform 8×8 non-overlapped block segmentation on the luminance difference map X ^h of X, and obtain M″×N” non-overlapping size 8×8 image blocks, and set the coordinate position in X ^h at The image block at (i″’, j″’) is denoted as define image blocks The energy inside the block and the energy at the edge of the block are respectively and ${\mathrm{Ex}}_{i^{\′\′\′},j^{\′\′\′}}^{\mathrm{In}}=\frac{1}{56}\underset{p=1}{\overset{8}{\mathrm{\Σ}}}\underset{q=1}{\overset{7}{\mathrm{\Σ}}}{x}_{i^{\′\′\′},j^{\′\′\′}}^h(p,q),{\mathrm{Ex}}_{i^{\′\′\′},j^{\′\′\′}}^{\mathrm{Ed}}=\frac{1}{8}\underset{p=1}{\overset{8}{\mathrm{\Σ}}}{x}_{i^{\′\′\′},j^{\′\′\′}}^h(p,8),$ in, The coefficient value at the coordinate position of (p,q), The coefficient value at the coordinate position (p,8), 1≤i″'≤M″, 1≤j″'≤N″, 1≤p≤8, 1≤q≤7;

Similarly, carry out 8 × 8 non-overlapped block divisions to the luminance difference map Y ^h of Y to obtain M″ × N” non-overlapping size 8 × 8 image blocks, and set the coordinate position in Y ^h at (i ″', j″'), the image block at the position is denoted as define image blocks The energy inside the block and the energy at the edge of the block are respectively and ${\mathrm{Ey}}_{i^{\′\′\′},j^{\′\′\′}}^{\mathrm{In}}=\frac{1}{56}\underset{p=1}{\overset{8}{\mathrm{\Σ}}}\underset{q=1}{\overset{7}{\mathrm{\Σ}}}{\mathrm{the\; y}}_{i^{\′\′\′},j^{\′\′\′}}^h(p,q),{\mathrm{Ey}}_{i^{\′\′\′},j^{\′\′\′}}^{\mathrm{Ed}}=\frac{1}{8}\underset{p=1}{\overset{8}{\mathrm{\Σ}}}{\mathrm{the\; y}}_{i^{\′\′\′},j^{\′\′\′}}^h(p,8),$ in, for The coefficient value at the coordinate position of (p,q), for The value of the coefficient at the middle coordinate position (p,8) 1≤i″'≤M″, 1≤j″'≤N″, 1≤p≤8, 1≤q≤7;

①-f, calculate the ratio between the edge energy of all image blocks in X ^h and the energy in the block, and the image block whose coordinate position in X ^h is at (i″’, j″’) The ratio of block edge energy to block energy is denoted as $R_{i^{\′\′\′},j^{\′\′\′}}^x=\frac{{\mathrm{Ex}}_{i^{\′\′\′},j^{\′\′\′}}^{\mathrm{Ed}}}{\mathrm{E.}{x}_{i^{\′\′\′},j^{\′\′\′}}^{\mathrm{In}}};$

Similarly, calculate the ratio between the edge energy of all image blocks in Y ^h and the energy in the block, and the image block whose coordinate position in Y ^h is at (i″’, j″’) The ratio of block edge energy to block energy is denoted as $R_{i^{\′\′\′},j^{\′\′\′}}^{\mathrm{the\; y}}=\frac{{\mathrm{Ey}}_{i^{\′\′\′},j^{\′\′\′}}^{\mathrm{Ed}}}{\mathrm{E.}{\mathrm{the\; y}}_{i^{\′\′\′},j^{\′\′\′}}^{\mathrm{In}}};$

Statistics satisfy the inequality The number of image blocks, and recorded as N ₀ ; define the discriminant index J, $J=\frac{N_0}{m^{\′\′}\×N^{\′\′}};$

①-g, judging whether J>Th _JPEG is established, if established, then determine that the distortion type of Y is JPEG distortion, and then end; otherwise, perform step ①-h; wherein, Th _JPEG is the JPEG distortion discrimination threshold; in this embodiment , the value of the JPEG distortion discrimination threshold Th _JPEG is 0.57;

①-h. Determine that the distortion type of Y is fuzzy-like distortion, that is, the distortion type of Y is Gaussian blur distortion, or JPEG2000 distortion, or fast fading distortion.

In this embodiment, the image data used are 808 images provided by the Image Quality Estimation Database (LIVE) disclosed by the Image and Video Engineering Laboratory of the University of Texas in the United States, including 29 undistorted reference images and 779 distorted images , including 145 Gaussian white noise distorted images, 145 Gaussian blur distorted images, 175 JPEG distorted images, 169 JPEG2000 distorted images and 145 fast fading distorted images. In addition, Fig. 2 shows 12 undistorted images selected from 29 undistorted reference images with simple texture, complex texture and medium texture. These 12 undistorted images and their corresponding five distortion types are used as Training set images, including 60 Gaussian white noise distorted images, Gaussian blur distorted images and fast fading distorted images each, 70 JPEG distorted images and JPEG2000 distorted images each; the remaining 17 undistorted images and their corresponding 5 kinds of distortion Types of images are used as test set images, including 85 distorted images of Gaussian white noise, 85 distorted images of Gaussian blur and 85 distorted images of fast fading, 105 distorted images of JPEG, and 99 distorted images of JPEG2000.

The first step is to separate the distorted images distorted by Gaussian white noise from all the distorted images in the training database. When determining whether the distortion type of the distorted image Y is the Gaussian white noise distortion threshold Th _WN involved in the process of determining whether the distortion type of the distorted image Y is Gaussian white noise distortion, in the interval [-0.5,1.5], take a value every 0.05 as Th _WN detection value, for each detected value, carry out the data training of the discrimination steps ①-a, ①-b and ①-c, and obtain the judgment accuracy respectively. The relation curve between the Th _WN size and the discrimination accuracy is shown in Figure 3 , it can be seen from FIG. 3 that when Th _WN =0.8, the Gaussian white noise distorted image can be separated with a correct rate of 100%.

The second step separates JPEG distorted images from non-Gaussian white noise distorted images in the training set. Take a point every 0.01 in the interval [0.4,0.7] as the detection value of Th _JPEG , do the data training of steps ①-d to ①-g for each detection value, and calculate the accuracy of the judgment respectively, Th The relationship curve between the size of _JPEG and the correct rate of discrimination is shown in Figure 4. It can be seen from Figure 4 that when Th _JPEG = 0.57, the JPEG can be separated from the non-Gaussian white noise distorted image with a correct rate of 100%. Distorted image.

②If the distortion type of the distorted image Y is Gaussian white noise distortion, then use a sliding window with a size of 8×8 to move pixel by pixel in the order of first row and second column in X, and divide X into M×N overlapping And for an image block with a size of 8×8, the image block whose coordinate position in X is (i,j) is recorded as x _i,j ; similarly, a sliding window with a size of 8×8 is used to press first and then The order of columns moves pixel by pixel, and Y is divided into M×N overlapping image blocks with a size of 8×8, and the image block whose coordinate position in Y is (i,j) is recorded as y _i,j ;in, H represents the height of X and Y, W represents the width of X and Y, the symbol is the symbol of rounding down, 1≤i≤M, 1≤j≤N;

If the distortion type of the distorted image Y is JPEG distortion, use a sliding window with a size of 8×8 to move pixel by pixel in the order of the first row and the second column in X, and divide X into M×N overlapping and sized is an 8×8 image block, and the image block whose coordinate position is (i,j) in X is recorded as x _i,j , and two-dimensional DCT transformation is performed on all image blocks x _i,j to obtain the corresponding transformed image block for Similarly, a sliding window with a size of 8×8 is used to move you pixel by pixel in the order of first row and second column in Y, and Y is divided into M×N overlapping image blocks with a size of 8×8. The image block whose coordinate position is (i, j) in Y is denoted as y _{i, j} , and two-dimensional DCT transformation is performed on all image blocks y _{i, j} , and the corresponding transformed image block is in, H represents the height of X and Y, W represents the width of X and Y, 1≤i≤M, 1≤j≤N;

If the distortion type of the distorted image Y is similar to fuzzy distortion, perform a first-level wavelet transform on X, extract the approximate component and record it as X _A , and use a sliding window with a size of 8×8 in the order of first row and second column in X _A Move point by point, divide X _A into M'×N' overlapping image blocks with a size of 8×8, and record the image block with coordinate position (i',j') in X _A as Similarly, perform a first-level wavelet transform on Y, extract the approximate component and record it as Y _A , use a sliding window with a size of 8×8 to move point by point in Y _A in the order of first row and second column, and divide Y _A into M'×N' overlapping image blocks with a size of 8×8, the image block whose coordinate position is (i', j') in Y _A is recorded as in, H' means the height of X _A and Y _A , W' means the width of X _A and Y _A , 1≤i'≤M', 1≤j'≤N';

③If the distortion type of the distorted image Y is Gaussian white noise distortion, then calculate the brightness mean and standard deviation of all pixels in each image block in X, and calculate the brightness mean and standard deviation of all pixels in each image block in Y The brightness mean and standard deviation, and then calculate the covariance between all pixels in the two image blocks with the same coordinate positions in X and Y, and the image block x _{i with the coordinate position (i, j) in X,} The brightness mean and standard deviation of all pixels in _j are recorded as and The brightness mean and standard deviation of all the pixels in the image block y _{i, j} at the coordinate position (i, j) in Y are recorded as and The correlation between all the pixels in the image block x _i,j whose coordinate position is (i,j) in X and all the pixels in the image block y _i,j whose coordinate position is (i,j) in Y is Variance is recorded as ${\mathrm{\σ}}_{x_{i,j}}=\sqrt{\frac{1}{64}\underset{u=1}{\overset{8}{\mathrm{\Σ}}}\underset{v=1}{\overset{8}{\mathrm{\Σ}}}{(x_{i,j}(u,v)-{\mathrm{\μ}}_{x_{i,j}})}^2},{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i,j}}=\frac{1}{64}\underset{u=1}{\overset{8}{\mathrm{\Σ}}}\underset{v=1}{\overset{8}{\mathrm{\Σ}}}{\mathrm{the\; y}}_{i,j}(u,v),$ ${\mathrm{\σ}}_{{\mathrm{the\; y}}_{i,j}}=\sqrt{\frac{1}{64}\underset{u=1}{\overset{8}{\mathrm{\Σ}}}\underset{v=1}{\overset{8}{\mathrm{\Σ}}}{({\mathrm{the\; y}}_{i,j}(u,v)-{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i,j}})}^2},$ ${\mathrm{\σ}}_{x_{i,j}{\mathrm{the\; y}}_{i,j}}=\frac{1}{64}\underset{u=1}{\overset{8}{\mathrm{\Σ}}}\underset{v=1}{\overset{8}{\mathrm{\Σ}}}\left[(x_{i,j}(u,v)-{\mathrm{\μ}}_{x_{i,j}})({\mathrm{the\; y}}_{i,j}(u,v)-{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i,j}})\right],$ Among them, x _i,j (u,v) represents the brightness value of the pixel point whose coordinate position in x _i,j is (u,v), and y _i,j (u,v) represents the coordinate position in y _i,j as The brightness value of the pixel point of (u,v), 1≤u≤8, 1≤v≤8;

If the distortion type of the distorted image Y is JPEG distortion, calculate the brightness mean and standard deviation of all pixels in each image block in X, and calculate the brightness mean and standard deviation of all pixels in each image block in Y Standard deviation, then calculate the mean and standard deviation of the DCT AC coefficients of each image block in X, and calculate the mean and standard deviation of the DCT AC coefficients of each image block in Y, and finally calculate all coordinates in X and Y The covariance between all DCT AC coefficients of two image blocks with the same position, and the brightness mean and standard deviation of all pixels in the image block x _{i, j} whose coordinate position is (i, j) in X are recorded as and The brightness mean value and standard deviation of all pixels in the image block y _{i, j} at the coordinate position (i, j) in Y are recorded as and A new image block obtained by performing DCT transformation on the image block x _{i, j} whose coordinate position is (i, j) in X The mean and standard deviation of all the AC coefficients of are recorded as and A new image block obtained by performing DCT transformation on the image block y _{i, j} whose coordinate position is (i, j) in Y The mean and standard deviation of all the AC coefficients of are recorded as and The DCT domain image block whose coordinate position in X is (i,j) and the DCT domain image block with the coordinate position (i,j) in Y The covariance between all AC coefficients in is denoted as ${\mathrm{\σ}}_{x_{i,j}}=\sqrt{\frac{1}{64}\underset{u=1}{\overset{8}{\mathrm{\Σ}}}\underset{v=1}{\overset{8}{\mathrm{\Σ}}}{(x_{i,j}(u,v)-{\mathrm{\μ}}_{x_{i,j}})}^2},{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i,j}}=\frac{1}{64}\underset{u=1}{\overset{8}{\mathrm{\Σ}}}\underset{v=1}{\overset{8}{\mathrm{\Σ}}}{\mathrm{the\; y}}_{i,j}(u,v),$ ${\mathrm{\σ}}_{{\mathrm{the\; y}}_{i,j}}=\sqrt{\frac{1}{64}\underset{u=1}{\overset{8}{\mathrm{\Σ}}}\underset{v=1}{\overset{8}{\mathrm{\Σ}}}{({\mathrm{the\; y}}_{i,j}(u,v)-{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i,j}})}^2},{\mathrm{\μ}}_{x_{i,j}^{\mathrm{D.}}}=\frac{1}{64}\underset{u^{\mathrm{D.}}=1}{\overset{8}{\mathrm{\Σ}}}\underset{v^{\mathrm{D.}}=1}{\overset{8}{\mathrm{\Σ}}}{x}_{i,j}^{\mathrm{D.}}(u^{\mathrm{D.}},v^{\mathrm{D.}}),$ ${\mathrm{\σ}}_{x_{i,j}^{\mathrm{D.}}}=\sqrt{\frac{1}{64}\underset{u^{\mathrm{D.}}=1}{\overset{8}{\mathrm{\Σ}}}\underset{v^{\mathrm{D.}}=1}{\overset{8}{\mathrm{\Σ}}}{(x_{i,j}^{\mathrm{D.}}(u^{\mathrm{D.}},v^{\mathrm{D.}})-{\mathrm{\μ}}_{x_{i,j}^{\mathrm{D.}}})}^2},{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i,j}^{\mathrm{D.}}}=\frac{1}{64}\underset{u^{\mathrm{D.}}=1}{\overset{8}{\mathrm{\Σ}}}\underset{v^{\mathrm{D.}}=1}{\overset{8}{\mathrm{\Σ}}}{\mathrm{the\; y}}_{i,j}^{\mathrm{D.}}(u^{\mathrm{D.}},v^{\mathrm{D.}}),$ ${\mathrm{\σ}}_{{\mathrm{the\; y}}_{i,j}^{\mathrm{D.}}}=\sqrt{\frac{1}{64}\underset{u^{\mathrm{D.}}=1}{\overset{8}{\mathrm{\Σ}}}\underset{v^{\mathrm{D.}}=1}{\overset{8}{\mathrm{\Σ}}}{({\mathrm{the\; y}}_{i,j}^{\mathrm{D.}}(u^{\mathrm{D.}},v^{\mathrm{D.}})-{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i,j}^{\mathrm{D.}}})}^2},$ ${\mathrm{\σ}}_{x_{i,j}^{\mathrm{D.}},{\mathrm{the\; y}}_{i,j}^{\mathrm{D.}}}=\frac{1}{64}\underset{u^{\mathrm{D.}}=1}{\overset{8}{\mathrm{\Σ}}}\underset{v^{\mathrm{D.}}=1}{\overset{8}{\mathrm{\Σ}}}\left[(x_{i,j}^{\mathrm{D.}}(u^{\mathrm{D.}},v^{\mathrm{D.}})-{\mathrm{\μ}}_{x_{i,j}^{\mathrm{D.}}})({\mathrm{the\; y}}_{i,j}^{\mathrm{D.}}(u^{\mathrm{D.}},v^{\mathrm{D.}})-{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i,j}^{\mathrm{D.}}})\right],$ Among them, x _i,j (u,v) represents the brightness value of the pixel point whose coordinate position in x _i,j is (u,v), and y _i,j (u,v) represents the coordinate position in y _i,j as The brightness value of the pixel of (u,v), 1≤u≤8, 1≤v≤8, express The DCT coefficient value at the coordinate position (u ^D , v ^D ), express The DCT coefficient value at the middle coordinate position (u ^D , v ^D ), 1≤u ^D ≤8, 1≤v ^D ≤8 and u ^D and v ^D are not 1 at the same time;

If the distortion type of the distorted image Y is fuzzy-like distortion, calculate the mean and standard deviation of all coefficient values in the approximate component X _A of X after the first-level wavelet transformation, and calculate the approximate component Y _A of Y after the first-level wavelet transformation The mean and standard deviation of all coefficient values in X _{A and Y A, and then calculate the covariance between all coefficients in the two image blocks with the same coordinate position in X A} and Y _A , and the coordinate position in X _A is (i ^W ,j ^W ) image block The mean and standard deviation of all coefficients in and The image block whose coordinate position is (i ^W , j ^W ) in Y _A The mean and standard deviation of all coefficients in and The image block whose coordinate position is (i ^W , j ^W ) in X _A All the pixels in Y _A and the image block whose coordinate position is (i ^W , j ^W ) The covariance among all the coefficients in is denoted as ${\mathrm{\μ}}_{x_{i^W,j^W}^W}=\frac{1}{64}\underset{u^W=1}{\overset{8}{\mathrm{\Σ}}}\underset{v^W=1}{\overset{8}{\mathrm{\Σ}}}{x}_{i^W,j^W}^W(u^W,v^W),{\mathrm{\σ}}_{x_{i^W,j^W}^W}=\sqrt{\frac{1}{64}\underset{u^W=1}{\overset{8}{\mathrm{\Σ}}}\underset{v^W=1}{\overset{8}{\mathrm{\Σ}}}{(x_{i^W,j^W}^W(u^W,v^W)-{\mathrm{\μ}}_{x_{i^W,j^W}^W})}^2},$ ${\mathrm{\μ}}_{{\mathrm{the\; y}}_{i^W,j^W}^W}=\frac{1}{64}\underset{u^W=1}{\overset{8}{\mathrm{\Σ}}}\underset{v^W=1}{\overset{8}{\mathrm{\Σ}}}{\mathrm{the\; y}}_{i^W,j^W}^W(u^W,v^W),{\mathrm{\σ}}_{{\mathrm{the\; y}}_{i^W,j^W}^W}=\sqrt{\frac{1}{64}\underset{u^W=1}{\overset{8}{\mathrm{\Σ}}}\underset{v^W=1}{\overset{8}{\mathrm{\Σ}}}{({\mathrm{the\; y}}_{i^W,j^W}^W(u^W,v^W)-{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i^W,j^W}^W})}^2},$ ${\mathrm{\σ}}_{x_{i^W,j^W}^W,{\mathrm{the\; y}}_{i^W,j^W}^W}=\frac{1}{64}\underset{u^W=1}{\overset{8}{\mathrm{\Σ}}}\underset{v^W=1}{\overset{8}{\mathrm{\Σ}}}\left[(x_{i^W,j^W}^W(u^W,v^W)-{\mathrm{\μ}}_{x_{i^W,j^W}^W})({\mathrm{the\; y}}_{i^W,j^W}^W(u^W,v^W)-{\mathrm{\μ}}_{u_{i^W,j^W}^W})\right],$ in, express The coefficient value at the coordinate position (u ^W , v ^W ), express Coefficient value at (u ^W , v ^W ), 1≤u ^W ≤8, 1≤v ^W ≤8;

④ If the distortion type of the distorted image Y is Gaussian white noise distortion, then calculate the brightness function, contrast function and structure function between all the coordinate positions in X and Y between the same two image blocks, and set the coordinate position in X as ( The brightness function, contrast function and structure function between the image block x _i,j of i,j) and the image block y _i,j whose coordinate position is (i,j) in Y are respectively denoted as l(xi _{, j} ,y _i,j ), c(x _i,j ,y _i,j ) and s(x _i,j ,y _i,j ), $l(x_{i,j},{\mathrm{the\; y}}_{i,j})=\frac{2{\mathrm{\μ}}_{x_{i,j}}{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i,j}}+C_1}{{\mathrm{\μ}}_{x_{i,j}}^2+{\mathrm{\μ}}_{{\mathrm{the\; y}}_{i,j}}^2+C_1},c(x_{i,j},{\mathrm{the\; y}}_{i,j})=\frac{2{\mathrm{\σ}}_{x_{i,j}}{\mathrm{\σ}}_{{\mathrm{the\; y}}_{i,j}}+C_2}{{\mathrm{\σ}}_{x_{i,j}}^2+{\mathrm{\σ}}_{{\mathrm{the\; y}}_{i,j}}^2+C_2},$ Among them, C ₁ , C ₂ , and C ₃ are small-value constants set to prevent the denominator from appearing zero. In this embodiment, C ₁ =0.01, C ₂ =0.02, and C ₃ =0.01;

If the distortion type of the distorted image Y is JPEG distortion, calculate the brightness function and contrast function between two image blocks with the same coordinate positions in X and Y, and calculate the two image blocks with the same coordinate positions in X and Y The structure function of the image block in the DCT domain, the brightness between the image block x i _{, j at the coordinate position (i, j) in X and the image block y i, j} at the coordinate position ₍ i, j) in Y function, contrast function and structure function are recorded as l(xi _,j ,y _i,j ) and c(xi _,j ,y _i,j ) respectively, and the coordinate position in X is (i,j) The new image block of image block x _{i, j} after DCT transformation The new image block after DCT transformation with the image block y _{i, j} whose coordinate position is (i, j) in Y The structure degree function between is denoted as f(x _i,j ,y _i,j ), $c(x_{i,j},{\mathrm{the\; y}}_{i,j})=\frac{2{\mathrm{\σ}}_{x_{i,j}}{\mathrm{\σ}}_{{\mathrm{the\; y}}_{i,j}}+C_2}{{\mathrm{\σ}}_{x_{i,j}}^2+{\mathrm{\σ}}_{{\mathrm{the\; y}}_{i,j}}^2+C_2},f(x_{i,j},{\mathrm{the\; y}}_{i,j})=\frac{2{\mathrm{\σ}}_{x_{i,j}^{\mathrm{D.}}}{\mathrm{\σ}}_{{\mathrm{the\; y}}_{i,j}^{\mathrm{D.}}}+C_3}{{\mathrm{\σ}}_{x_{i,j}^{\mathrm{D.}}}^2+{\mathrm{\σ}}_{{\mathrm{the\; y}}_{i,j}^{\mathrm{D.}}}^2+C_3},$ Among them, C ₁ , C ₂ , and C ₃ are small-value constants set to prevent the denominator from appearing zero. In this embodiment, C ₁ =0.01, C ₂ =0.02, and C ₃ =0.01;

If the distortion type of the distorted image Y is fuzzy-like distortion, calculate the wavelet coefficient function, the wavelet coefficient contrast function and the wavelet coefficient structure function between two image blocks with the same coordinate positions in X and Y, and convert the coordinates in X to The wavelet coefficient function, the wavelet coefficient contrast function and the wavelet coefficient structure function between the image block x _{i, j} at the position (i, j) and the image block y _{i, j} at the coordinate position (i, j) in Y correspond to They are denoted as l ^W (xi _,j ,y _i,j ), c ^W (xi _,j ,y _i,j ) and s ^W (xi _,j ,y _i,j ) respectively, $c^W(x_{i,j},{\mathrm{the\; y}}_{i,j})=\frac{2{\mathrm{\σ}}_{x_{i^W,j^W}^W}{\mathrm{\σ}}_{{\mathrm{the\; y}}_{i^W,j^W}^W}+C_2}{{\mathrm{\σ}}_{x_{i^W,j^W}^W}^2+{\mathrm{\σ}}_{{\mathrm{the\; y}}_{i^W,j^W}^W}^2+C_2},{\mathrm{the\; s}}^W(x_{i,j},{\mathrm{the\; y}}_{i,j})=\frac{{\mathrm{\σ}}_{x_{i^W,j^W}^W,{\mathrm{the\; y}}_{i^W,j^W}^W}+C_3}{{\mathrm{\σ}}_{x_{i^W,j^W}^W}{\mathrm{\σ}}_{{\mathrm{the\; y}}_{i^W,j^W}^W}+C_3},$ Among them, C ₁ , C ₂ , and C ₃ are small-value constants set to prevent the denominator from appearing zero. In this embodiment, C ₁ =0.01, C ₂ =0.02, and C ₃ =0.01;

⑤If the distortion type of the distorted image Y is Gaussian white noise distortion, then calculate all the values in X and Y according to the brightness function, contrast function and structure function between two image blocks with the same coordinate positions in X and Y. The structural similarity between two image blocks with the same coordinate position, the image block x i _{, j with the coordinate position (i, j) in X and the image block y i, j} with the coordinate position (i, j) in Y _, The structural similarity between _j is recorded as SSIM(xi _,j ,y _i,j ), SSIM(xi _,j ,y _i,j )=[l(xi _,j ,y _i,j )] ^α [c(x _i,j ,y _i,j )] ^β [s(x _i,j ,y _i,j )] ^γ , where α, β and γ are adjustment factors, and in this embodiment, α=β =γ=1;

If the distortion type of the distorted image Y is JPEG distortion, according to the brightness function and contrast function between two image blocks with the same coordinate positions in X and Y, and two images with the same coordinate positions in X and Y The structure degree function of the block in the DCT domain, calculate the structural similarity based on the DCT domain between two image blocks with the same coordinate position in X and Y, and take the image block x _i with the coordinate position (i, j) in X _,j and the image block y i, _j whose coordinate position is (i,j) in Y is denoted as FSSIM(xi _,j ,y _i,j ), FSSIM(xi _{,j j} ,y _i,j )=[l(x _i,j ,y _i,j )] ^α [c(x _i,j ,y _i,j )] ^β [f(x _{i,j ,} y _i,j )] ^γ , wherein α, β and γ are adjustment factors, and in this embodiment, α=β=γ=1;

If the distortion type of the distorted image Y is fuzzy-like distortion, calculate X and Y according to the wavelet coefficient function, wavelet coefficient contrast function and wavelet coefficient structure function between two image blocks with the same coordinate positions in X and Y The structural similarity based on the wavelet domain between all the two image blocks with the same coordinate position in X, and the image block x _{i, j} with the coordinate position (i, j) in X and the coordinate position in Y (i, j ) between image blocks y _i,j based on the wavelet domain structure is denoted as WSSIM(xi _,j ,y _i,j ), WSSIM(xi _,j ,y _i,j )=[l ^W (xi _,j ,y _i,j )] ^α [c ^W (xi _,j ,y _i,j )] ^β [s ^W (xi _{,j ,} y _i,j )] ^γ , where α, β and γ are adjustment factors, and in this embodiment, α=β=γ=1;

⑥ If the distortion type of the distorted image Y is Gaussian white noise distortion, then calculate the objective quality score of Y according to the structural similarity based on the pixel domain between all the two image blocks with the same coordinate positions in X and Y, and record is _Qwn , $Q_{\mathrm{wn}}=\frac{1}{m\×N}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}\mathrm{SSIM}(x_{i,j},{\mathrm{the\; y}}_{i,j});$

If the distortion type of the distorted image Y is JPEG distortion, then calculate the objective quality score of Y according to the structural similarity based on the DCT domain between two image blocks with the same coordinate positions in X and Y, and denote it as Q _jpeg , $Q_{\mathrm{jpeg}}=\frac{1}{m\×N}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}\mathrm{FSSIM}(x_{i,j},{\mathrm{the\; y}}_{i,j});$

If the distortion type of the distorted image Y is similar to fuzzy distortion, then calculate the objective quality score of Y according to the structural similarity based on the wavelet domain between two image blocks with the same coordinate positions in X and Y, and denote it as Q _blur , $Q_{\mathrm{blurred}}=\frac{1}{m\×N}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}\mathrm{WSSIM}(x_{i,j},{\mathrm{the\; y}}_{i,j}).$

In this embodiment, using the 29 undistorted images and 779 single distorted images provided in LIVE and the DMOS (differential mean opinion scores) values corresponding to each distorted image, each distorted image is calculated according to steps ① to ⑥ The quality evaluation score Q of the 779 distorted images was used to perform nonlinear fitting of the four-parameter Logistic function on the objective quality evaluation scores Q and DMOS values of 779 distorted images; four commonly used objective parameters for evaluating image quality evaluation methods were used as evaluation indicators. The evaluation indicators are linear correlation coefficient CC (correlation coefficient), Spearman correlation coefficient SROCC (Spearman rank-order correlation coefficient), dispersion rate OR (out ratio) and mean square error coefficient RMSE (rooted mean squared error). The higher the CC value and SROCC value, the lower the OR value and RMSE value, the better the correlation between the image objective evaluation method and DMOS.

Table 1 lists the values of CC, SROCC, OR and RMSE coefficients for evaluating performance under various distortion types. From the data listed in Table 1, it can be seen that the objective quality score Q and subjective score DMOS of the distorted image obtained in this embodiment The correlation between is very high, and CC value all exceeds 0.94, and SROCC value all exceeds 0.91, and OR value is all lower than 0.41, and RMSE value is all lower than 5.4, and this shows that the objective evaluation result of the inventive method is consistent with the subjective perception of human eyes. The consistency of the results is very high, which fully demonstrates the effectiveness of the method of the present invention.

Table 1 Correlation between objective and subjective evaluation scores of distorted images obtained in this implementation

Claims (5)

1. A self-adaptive image quality evaluation method based on distortion type judgment is characterized in that the processing process is as follows:

firstly, determining the distortion type of a distorted image to be evaluated;

secondly, corresponding processing is carried out by combining the distortion type of the distorted image to be evaluated;

if the distorted image is Gaussian white noise distortion, dividing the original undistorted image and the distorted image to be evaluated into a plurality of overlapped image blocks with the size of 8 multiplied by 8 in a pixel domain, and acquiring the structural similarity based on the pixel domain between all two image blocks with the same coordinate position in the original undistorted image and the distorted image to be evaluated by calculating the brightness mean value and the standard deviation of all pixel points in each image block in the original undistorted image and the distorted image to be evaluated in the pixel domain and the covariance between all pixel point brightness values in all two image blocks with the same coordinate position in the original undistorted image and the distorted image to be evaluated;

if the distorted image is JPEG distorted, dividing the original undistorted image and the distorted image to be evaluated into a plurality of overlapped image blocks with the size of 8 multiplied by 8 in a DCT domain, and acquiring the DCT domain-based structural similarity between all two image blocks with the same coordinate position in the original undistorted image and the distorted image to be evaluated by calculating the mean value and the standard deviation of all DCT coefficients of the original undistorted image block and the distorted image to be evaluated in the DCT domain and the covariance between all DCT coefficients of all two image blocks with the same coordinate position in the DCT domain;

if the distorted image is similar to fuzzy distortion, dividing the original undistorted image and the distorted image to be evaluated into a plurality of overlapped image blocks with the size of 8 multiplied by 8 in a wavelet domain, and acquiring the structural similarity based on the wavelet domain between all two image blocks with the same coordinate position in the original undistorted image and the distorted image to be evaluated by calculating the mean value and the standard deviation of all wavelet coefficients of the original undistorted image block and the distorted image to be evaluated in the wavelet domain and the covariance between all wavelet coefficients of all two image blocks with the same coordinate position in the wavelet domain;

and finally, obtaining the objective quality score of the distorted image to be evaluated according to the structural similarity between the two image blocks with the same coordinate positions in the original undistorted image and the distorted image to be evaluated.

2. The adaptive image quality evaluation method based on distortion type judgment according to claim 1, characterized in that: the method specifically comprises the following steps:

firstly, enabling X to represent an original undistorted image, enabling Y to represent a distorted image to be evaluated, and determining the distortion type of Y through a distortion type distinguishing method, wherein the distortion type of Y is one of Gaussian white noise distortion, JPEG distortion and quasi-fuzzy distortion, and the quasi-fuzzy distortion comprises Gaussian fuzzy distortion, JPEG2000 distortion and fast fading distortion;

if the distortion type of the distorted image Y is Gaussian white noise distortion, a sliding window with the size of 8 multiplied by 8 is adopted to move in X pixel by pixel, the X is divided into M multiplied by N overlapped image blocks with the size of 8 multiplied by 8, and the image block with the coordinate position (i, j) in X is marked as X_i,j(ii) a Similarly, moving the sliding window with the size of 8 × 8 in Y pixel by pixel, dividing Y into M × N overlapped image blocks with the size of 8 × 8, and recording the image block with the coordinate position (i, j) in Y as Y_i,j(ii) a Wherein, h denotes the height of X and Y, W denotes the width of X and Y, symbolsI is more than or equal to 1 and less than or equal to M, and j is more than or equal to 1 and less than or equal to N;

if the distortion type of the distorted image Y is JPEG distortion, moving the distorted image Y pixel by pixel in X by adopting a sliding window with the size of 8 multiplied by 8, dividing X into M multiplied by N overlapped image blocks with the size of 8 multiplied by 8, and recording the image block with the coordinate position (i, j) in X as X_i,jFor all image blocks x_i,jPerforming two-dimensional DCT to obtain image blocks after corresponding transformationAlso, a slider having a size of 8X 8 was usedMoving the movable window in Y pixel by pixel, dividing Y into M × N overlapped image blocks with size of 8 × 8, and recording the image block with coordinate position (i, j) in Y as Y_i,jFor all image blocks y_i,jPerforming two-dimensional DCT to obtain image blocks after corresponding transformationWherein,h represents the height of X and Y, W represents the width of X and Y, i is more than or equal to 1 and less than or equal to M, and j is more than or equal to 1 and less than or equal to N;

if the distortion type of the distorted image Y is similar to fuzzy distortion, performing one-level wavelet transform on X, extracting approximate components and recording as X_AUsing a sliding window of size 8X 8 at X_AMoving X point by point_ADividing into M '× N' overlapped image blocks with size of 8 × 8, and dividing X into X blocks_AThe image block with the middle coordinate position (i ', j') is recorded asSimilarly, one-level wavelet transform is performed on Y, and approximate components are extracted and recorded as Y_AUsing a sliding window with a size of 8X 8 in Y_AMoving point by point, and moving Y_ADividing into M '× N' overlapped image blocks with size of 8 × 8, and dividing Y into_AThe image block with the middle coordinate position (i ', j') is recorded asWherein,h' represents X_AAnd Y_AW' represents X_AAnd Y_AThe width of the composite is that i 'is more than or equal to 1 and is more than or equal to M', j 'is more than or equal to 1 and is more than or equal to N';

thirdly, if the distortion type of the distorted image Y is Gaussian white noise distortion, calculating the brightness average sum of all pixel points in each image block in the XStandard deviation, calculating the brightness mean value and standard deviation of all pixel points in each image block in Y, then calculating the covariance between all pixel points in two image blocks with the same coordinate position in X and Y, and setting the coordinate position in X as (i, j) image block X_i,jThe corresponding brightness mean value and standard deviation of all the pixel points are recorded asAndthe coordinate position in Y is set as (i, j) image block Y_i,jThe corresponding brightness mean value and standard deviation of all the pixel points are recorded asAndthe coordinate position in X is set as (i, j) image block X_i,jAll the pixel points in (a) and the image block Y with coordinate position (i, j) in Y_i,jThe covariance of all the pixels in (1) is recorded as <math> <mrow> <msub> <mi>&mu;</mi> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <mn>64</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <mi>u</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <mi>v</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>u</mi> <mo>,</mo> <mi>v</mi> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </math> <math> <mrow> <msub> <mi>&sigma;</mi> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </msub> <mo>=</mo> <msqrt> <mfrac> <mn>1</mn> <mn>64</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <mi>u</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <mi>v</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <msup> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>u</mi> <mo>,</mo> <mi>v</mi> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>&mu;</mi> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> </msqrt> <mo>,</mo> <msub> <mi>&mu;</mi> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <mn>64</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <mi>u</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <mi>v</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>u</mi> <mo>,</mo> <mi>v</mi> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </math> <math> <mrow> <msub> <mi>&sigma;</mi> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </msub> <mo>=</mo> <msqrt> <mfrac> <mn>1</mn> <mn>64</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <mi>u</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <mi>v</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <msup> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>u</mi> <mo>,</mo> <mi>v</mi> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>&mu;</mi> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> </msqrt> <mo>,</mo> </mrow> </math> <math> <mrow> <msub> <mi>&sigma;</mi> <mrow> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <msub> <mi>v</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </mrow> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <mn>64</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <mi>u</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <mi>v</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <mo>[</mo> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>u</mi> <mo>,</mo> <mi>v</mi> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>&mu;</mi> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </msub> <mo>)</mo> </mrow> <mrow> <mo>(</mo> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>u</mi> <mo>,</mo> <mi>v</mi> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>&mu;</mi> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </msub> <mo>)</mo> </mrow> <mo>]</mo> <mo>,</mo> </mrow> </math> Wherein x is_i,j(u, v) denotes x_i,jThe brightness value y of the pixel point with the middle coordinate position (u, v)_i,j(u, v) represents y_i,jMiddle coordinate positionThe brightness value of the pixel point of (u, v) is that u is more than or equal to 1 and less than or equal to 8, and v is more than or equal to 1 and less than or equal to 8;

if the distortion type of the distorted image Y is JPEG distortion, calculating the brightness mean value and standard deviation of all pixel points in each image block in X, calculating the brightness mean value and standard deviation of all pixel points in each image block in Y, then calculating the mean value and standard deviation of DCT alternating current coefficients of each image block in X, calculating the mean value and standard deviation of DCT alternating current coefficients of each image block in Y, finally calculating the covariance between all DCT alternating current coefficients of two image blocks with the same coordinate position in X and Y, and setting the coordinate position in X as (i, j) image block X_i,jThe corresponding brightness mean value and standard deviation of all the pixel points are recorded asAndthe coordinate position in Y is set as (i, j) image block Y_i,jThe corresponding brightness mean value and standard deviation of all the pixel points are recorded asAndthe coordinate position in X is set as (i, j) image block X_i,jNew image block obtained after DCT transformationRespectively corresponding to the mean value and the standard deviation of all the AC coefficients are recorded asAndthe coordinate position in Y is set as (i, j) image block Y_i,jObtained after DCT transformationNew image blockRespectively corresponding to the mean value and the standard deviation of all the AC coefficients are recorded asAndDCT domain image block with coordinate position (i, j) in XAnd a DCT domain image block with coordinate position (i, j) in YIs recorded as the covariance between all the ac coefficients in <math> <mrow> <msub> <mi>&mu;</mi> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <mn>64</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <mi>u</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <mi>v</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>u</mi> <mo>,</mo> <mi>v</mi> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </math> <math> <mrow> <msub> <mi>&sigma;</mi> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </msub> <mo>=</mo> <msqrt> <mfrac> <mn>1</mn> <mn>64</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <mi>u</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <mi>v</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <msup> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>u</mi> <mo>,</mo> <mi>v</mi> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>&mu;</mi> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> </msqrt> <mo>,</mo> <msub> <mi>&mu;</mi> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <mn>64</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <mi>u</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <mi>v</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>u</mi> <mo>,</mo> <mi>v</mi> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </math> <math> <mrow> <msub> <mi>&sigma;</mi> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </msub> <mo>=</mo> <msqrt> <mfrac> <mn>1</mn> <mn>64</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <mi>u</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <mi>v</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <msup> <mrow> <mo>(</mo> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>u</mi> <mo>,</mo> <mi>v</mi> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>&mu;</mi> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> </msqrt> <mo>,</mo> <msub> <mi>&mu;</mi> <msubsup> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> <mi>D</mi> </msubsup> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <mn>64</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>u</mi> <mi>D</mi> </msup> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>v</mi> <mi>D</mi> </msup> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <msubsup> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> <mi>D</mi> </msubsup> <mrow> <mo>(</mo> <msup> <mi>u</mi> <mi>D</mi> </msup> <mo>,</mo> <msup> <mi>v</mi> <mi>D</mi> </msup> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </math> <math> <mrow> <msub> <mi>&sigma;</mi> <msubsup> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> <mi>D</mi> </msubsup> </msub> <mo>=</mo> <msqrt> <mfrac> <mn>1</mn> <mn>64</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>u</mi> <mi>D</mi> </msup> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>v</mi> <mi>D</mi> </msup> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <msup> <mrow> <mo>(</mo> <msubsup> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> <mi>D</mi> </msubsup> <mrow> <mo>(</mo> <msup> <mi>u</mi> <mi>D</mi> </msup> <mo>,</mo> <msup> <mi>v</mi> <mi>D</mi> </msup> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>&mu;</mi> <msubsup> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> <mi>D</mi> </msubsup> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> </msqrt> <mo>,</mo> <msub> <mi>&mu;</mi> <msubsup> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> <mi>D</mi> </msubsup> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <mn>64</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>u</mi> <mi>D</mi> </msup> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>v</mi> <mi>D</mi> </msup> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <msubsup> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> <mi>D</mi> </msubsup> <mrow> <mo>(</mo> <msup> <mi>u</mi> <mi>D</mi> </msup> <mo>,</mo> <msup> <mi>v</mi> <mi>D</mi> </msup> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </math> <math> <mrow> <msub> <mi>&sigma;</mi> <msubsup> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> <mi>D</mi> </msubsup> </msub> <mo>=</mo> <msqrt> <mfrac> <mn>1</mn> <mn>64</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>u</mi> <mi>D</mi> </msup> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>v</mi> <mi>D</mi> </msup> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <msup> <mrow> <mo>(</mo> <msubsup> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> <mi>D</mi> </msubsup> <mrow> <mo>(</mo> <msup> <mi>u</mi> <mi>D</mi> </msup> <mo>,</mo> <msup> <mi>v</mi> <mi>D</mi> </msup> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>&mu;</mi> <msubsup> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> <mi>D</mi> </msubsup> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> </msqrt> <mo>,</mo> </mrow> </math> <math> <mrow> <msub> <mi>&sigma;</mi> <mrow> <msubsup> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> <mi>D</mi> </msubsup> <mo>,</mo> <msubsup> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> <mi>D</mi> </msubsup> </mrow> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <mn>64</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>u</mi> <mi>D</mi> </msup> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>v</mi> <mi>D</mi> </msup> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <mo>[</mo> <mrow> <mo>(</mo> <msubsup> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> <mi>D</mi> </msubsup> <mrow> <mo>(</mo> <msup> <mi>u</mi> <mi>D</mi> </msup> <mo>,</mo> <msup> <mi>v</mi> <mi>D</mi> </msup> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>&mu;</mi> <msubsup> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> <mi>D</mi> </msubsup> </msub> <mo>)</mo> </mrow> <mrow> <mo>(</mo> <msubsup> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> <mi>D</mi> </msubsup> <mrow> <mo>(</mo> <msup> <mi>u</mi> <mi>D</mi> </msup> <mo>,</mo> <msup> <mi>v</mi> <mi>D</mi> </msup> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>&mu;</mi> <msubsup> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> <mi>D</mi> </msubsup> </msub> <mo>)</mo> </mrow> <mo>]</mo> <mo>,</mo> </mrow> </math> Wherein x is_i,j(u, v) denotes x_i,jThe brightness value y of the pixel point with the middle coordinate position (u, v)_i,j(u, v) represents y_i,jThe brightness value of the pixel point with the middle coordinate position (u, v) is that u is more than or equal to 1 and less than or equal to 8, v is more than or equal to 1 and less than or equal to 8,to representThe middle coordinate position is (u)^D,v^D) The value of the DCT coefficient of (a),to representThe middle coordinate position is (u)^D,v^D) Value of DCT coefficient of 1. ltoreq. u^D≤8,1≤v^DU is less than or equal to 8^DAnd v^DIs not 1 at the same time;

if the distortion type of the distorted image Y is similar to fuzzy distortion, calculating an approximate component X of the distorted image Y after the first-level wavelet transform_AThe mean value and standard deviation of all the coefficient values in the above-mentioned formula, and calculating the approximate component Y of Y after one-stage wavelet transform_AThe mean and standard deviation of all the coefficient values in (A) and then calculating X_AAnd Y_AAll co-variances between all coefficients in two image blocks with the same coordinate position, X_AThe middle coordinate position is (i)^W,j^W) Image block ofThe mean and standard deviation of all coefficients are respectively recorded asAndwill Y_AThe middle coordinate position is (i)^W,j^W) Image block ofThe mean and standard deviation of all coefficients in (A) are respectively recorded asAndmixing X_AThe middle coordinate position is (i)^W,j^W) Image block ofAll pixel points in (2) and Y_AThe middle coordinate position is (i)^W,j^W) Image block ofIs recorded as the covariance between all coefficients in <math> <mrow> <msub> <mi>&mu;</mi> <msubsup> <mi>x</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <mn>64</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>u</mi> <mi>W</mi> </msup> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>v</mi> <mi>W</mi> </msup> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <msubsup> <mi>x</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> <mrow> <mo>(</mo> <msup> <mi>u</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>v</mi> <mi>W</mi> </msup> <mo>)</mo> </mrow> <mo>,</mo> <msub> <mi>&sigma;</mi> <msubsup> <mi>x</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> </msub> <mo>=</mo> <msqrt> <mfrac> <mn>1</mn> <mn>64</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>u</mi> <mi>W</mi> </msup> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>v</mi> <mi>W</mi> </msup> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <msup> <mrow> <mo>(</mo> <msubsup> <mi>x</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> <mrow> <mo>(</mo> <msup> <mi>u</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>v</mi> <mi>W</mi> </msup> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>&mu;</mi> <msubsup> <mi>x</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> </msqrt> <mo>,</mo> </mrow> </math> <math> <mrow> <msub> <mi>&mu;</mi> <msubsup> <mi>y</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <mn>64</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>u</mi> <mi>W</mi> </msup> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>v</mi> <mi>W</mi> </msup> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <msubsup> <mi>y</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> <mrow> <mo>(</mo> <msup> <mi>u</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>v</mi> <mi>W</mi> </msup> <mo>)</mo> </mrow> <mo>,</mo> <msub> <mi>&sigma;</mi> <msubsup> <mi>y</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> </msub> <mo>=</mo> <msqrt> <mfrac> <mn>1</mn> <mn>64</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>u</mi> <mi>W</mi> </msup> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>v</mi> <mi>W</mi> </msup> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <msup> <mrow> <mo>(</mo> <msubsup> <mi>y</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> <mrow> <mo>(</mo> <msup> <mi>u</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>v</mi> <mi>W</mi> </msup> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>&mu;</mi> <msubsup> <mi>y</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> </msqrt> <mo>,</mo> </mrow> </math> <math> <mrow> <msub> <mi>&sigma;</mi> <mrow> <msubsup> <mi>x</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> <mo>,</mo> <msubsup> <mi>y</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> </mrow> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <mn>64</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>u</mi> <mi>W</mi> </msup> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>v</mi> <mi>W</mi> </msup> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <mo>[</mo> <mrow> <mo>(</mo> <msubsup> <mi>x</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> <mrow> <mo>(</mo> <msup> <mi>u</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>v</mi> <mi>W</mi> </msup> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>&mu;</mi> <msubsup> <mi>x</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>i</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> </msub> <mo>)</mo> </mrow> <mrow> <mo>(</mo> <msubsup> <mi>y</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> <mrow> <mo>(</mo> <msup> <mi>u</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>v</mi> <mi>W</mi> </msup> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>&mu;</mi> <msubsup> <mi>y</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> </msub> <mo>)</mo> </mrow> <mo>]</mo> <mo>,</mo> </mrow> </math> Wherein,to representThe middle coordinate position is (u)^W,v^W) The value of the coefficient of (a) is,to representThe middle coordinate position is (u)^W,v^W) Coefficient value of 1. ltoreq. u^W≤8,1≤v^W≤8；

If the distortion type of the distorted image Y is Gaussian white noise distortion, calculating a brightness function, a contrast function and a structure function between two image blocks with the same coordinate positions in X and YLet X be an image block X with coordinate position (i, j)_i,jAnd image block Y with coordinate position (i, j) in Y_i,jThe correspondence of the brightness function, the contrast function and the structure function between the two is respectively marked as l (x)_i,j,y_i,j)、c(x_i,j,y_i,j) And s (x)_i,j,y_i,j)， <math> <mrow> <mi>l</mi> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <mn>2</mn> <msub> <mi>&mu;</mi> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </msub> <msub> <mi>&mu;</mi> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </msub> <mo>+</mo> <msub> <mi>C</mi> <mn>1</mn> </msub> </mrow> <mrow> <msubsup> <mi>&mu;</mi> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mn>2</mn> </msubsup> <mo>+</mo> <msubsup> <mi>&mu;</mi> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mn>2</mn> </msubsup> <mo>+</mo> <msub> <mi>C</mi> <mn>1</mn> </msub> </mrow> </mfrac> <mo>,</mo> <mi>c</mi> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>,</mo> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <mn>2</mn> <msub> <mi>&sigma;</mi> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </msub> <msub> <mi>&sigma;</mi> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </msub> <mo>+</mo> <msub> <mi>C</mi> <mn>2</mn> </msub> </mrow> <mrow> <msubsup> <mi>&sigma;</mi> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mn>2</mn> </msubsup> <mo>+</mo> <msubsup> <mi>&sigma;</mi> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mn>2</mn> </msubsup> <mo>+</mo> <msub> <mi>C</mi> <mn>2</mn> </msub> </mrow> </mfrac> <mo>,</mo> </mrow> </math> Wherein, C₁、C₂、C₃A small value constant set to avoid zero denominator;

if the distortion type of the distorted image Y is JPEG distortion, all coordinate positions in X and Y are calculatedSetting a brightness function and a contrast function between two identical image blocks, calculating the structure degree function of all two image blocks with the same coordinate position in X and Y in a DCT domain, and setting the coordinate position in X as the image block X of (i, j)_i,jAnd image block Y with coordinate position (i, j) in Y_i,jThe correspondence between the brightness function and the contrast function is denoted as l (x), respectively_i,j,y_i,j) And c (x)_i,j,y_i,j) Let X be an image block X with coordinate position (i, j)_i,jNew image blocks after DCT transformationAnd image block Y with coordinate position (i, j) in Y_i,jNew image blocks after DCT transformationThe function of structural degree between is recorded as f (x)_i,j,y_i,j)， <math> <mrow> <mi>c</mi> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>,</mo> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <mn>2</mn> <msub> <mi>&sigma;</mi> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </msub> <msub> <mi>&sigma;</mi> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> </msub> <mo>+</mo> <msub> <mi>C</mi> <mn>2</mn> </msub> </mrow> <mrow> <msubsup> <mi>&sigma;</mi> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mn>2</mn> </msubsup> <mo>+</mo> <msubsup> <mi>&sigma;</mi> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mn>2</mn> </msubsup> <mo>+</mo> <msub> <mi>C</mi> <mn>2</mn> </msub> </mrow> </mfrac> <mo>,</mo> <mi>f</mi> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>,</mo> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <msub> <mi>&sigma;</mi> <mrow> <msubsup> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> <mi>D</mi> </msubsup> <mo>,</mo> <msubsup> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> <mi>D</mi> </msubsup> </mrow> </msub> <mo>+</mo> <msub> <mi>C</mi> <mn>3</mn> </msub> </mrow> <mrow> <msub> <mi>&sigma;</mi> <msubsup> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> <mi>D</mi> </msubsup> </msub> <msub> <mi>&sigma;</mi> <msubsup> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> <mi>D</mi> </msubsup> </msub> <mo>+</mo> <msub> <mi>C</mi> <mn>3</mn> </msub> </mrow> </mfrac> <mo>,</mo> </mrow> </math> Wherein, C₁、C₂、C₃A small value constant set to avoid zero denominator;

if the distortion type of the distorted image Y is similar to fuzzy distortion, calculating a wavelet coefficient brightness function, a wavelet coefficient contrast function and a wavelet coefficient structure degree function between two image blocks with the same coordinate positions in X and Y, and setting the coordinate position in X as an image block X of (i, j)_i,jAnd image block Y with coordinate position (i, j) in Y_i,jThe correspondence of the wavelet coefficient brightness function, the wavelet coefficient contrast function and the wavelet coefficient structure degree function is respectively marked as l^W(x_i,j,y_i,j)、c^W(x_i,j,y_i,j) And s^W(x_i,j,y_i,j)， <math> <mrow> <msup> <mi>l</mi> <mi>W</mi> </msup> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>,</mo> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <mn>2</mn> <msub> <mi>&mu;</mi> <msubsup> <mi>x</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>i</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> </msub> <msub> <mi>&mu;</mi> <msubsup> <mi>y</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> </msub> <mo>+</mo> <msub> <mi>C</mi> <mn>1</mn> </msub> </mrow> <mrow> <msubsup> <mi>&mu;</mi> <msubsup> <mi>x</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mn>2</mn> </msubsup> <mn>2</mn> </msubsup> <mo>+</mo> <msubsup> <mi>&mu;</mi> <msubsup> <mi>y</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> <mn>2</mn> </msubsup> <mo>+</mo> <msub> <mi>C</mi> <mn>1</mn> </msub> </mrow> </mfrac> <mo>,</mo> <msup> <mi>c</mi> <mi>W</mi> </msup> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <mn>2</mn> <msub> <mi>&sigma;</mi> <msubsup> <mi>x</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> </msub> <msub> <mi>&sigma;</mi> <msubsup> <mi>y</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> </msub> <mo>+</mo> <msub> <mi>C</mi> <mn>2</mn> </msub> </mrow> <mrow> <msubsup> <mi>&sigma;</mi> <msubsup> <mi>x</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> <mn>2</mn> </msubsup> <mo>+</mo> <msubsup> <mi>&sigma;</mi> <msubsup> <mi>y</mi> <mrow> <msup> <mi>i</mi> <mi>W</mi> </msup> <mo>,</mo> <msup> <mi>j</mi> <mi>W</mi> </msup> </mrow> <mi>W</mi> </msubsup> <mn>2</mn> </msubsup> <mo>+</mo> <msub> <mi>C</mi> <mn>2</mn> </msub> </mrow> </mfrac> <mo>,</mo> </mrow> </math> Wherein, C₁、C₂、C₃A small value constant set to avoid zero denominator;

fifthly, if the distortion type of the distorted image Y is Gaussian white noise distortion, calculating the structural similarity between the two image blocks with the same coordinate positions in X and Y according to the brightness function, the contrast function and the structural degree function between the two image blocks with the same coordinate positions in X and Y, and setting the coordinate position in X as the image block X of (i, j)_i,jAnd image block Y with coordinate position (i, j) in Y_i,jThe structural similarity between them is denoted as SSIM (x)_i,j,y_i,j)，SSIM(x_i,j,y_i,j)＝[l(x_i,j,y_i,j)]^α[c(x_i,j,y_i,j)]^β[s(x_i,j,y_i,j)]^γWherein α, β and γ are regulatory factors;

if the distortion type of the distorted image Y is JPEG distortion, calculating the structural similarity based on the DCT domain between two image blocks with the same coordinate positions in X and Y according to a brightness function and a contrast function between the two image blocks with the same coordinate positions in X and Y and a structural function of the two image blocks with the same coordinate positions in X and Y in the DCT domain, and carrying out image block X with the coordinate position (i, j) in X_i,jAnd image block Y with coordinate position (i, j) in Y_i,jThe structural similarity based on DCT domain between them is marked as FSSIM (x)_i,j,y_i,j)，FSSIM(x_i,j,y_i,j)＝[l(x_i,j,y_i,j)]^α[c(x_i,j,y_i,j)]^β[f(x_i,j,y_i,j)]^γWherein α, β and γ are regulatory factors;

if the distortion type of the distorted image Y is similar to fuzzy distortion, calculating a wavelet domain-based image block between all two image blocks with the same coordinate position in X and Y according to a wavelet coefficient brightness function, a wavelet coefficient contrast function and a wavelet coefficient structure degree function between all two image blocks with the same coordinate position in X and YStructural similarity, namely, the image block X with coordinate position (i, j) in X_i,jAnd image block Y with coordinate position (i, j) in Y_i,jThe structural similarity based on the wavelet domain between the two is marked as WSSIM (x)_i,j,y_i,j)， <math> <mrow> <mi>WSSIM</mi> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>,</mo> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>=</mo> <msup> <mrow> <mo>[</mo> <msup> <mi>l</mi> <mi>W</mi> </msup> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>,</mo> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>]</mo> </mrow> <mi>&alpha;</mi> </msup> <msup> <mrow> <mo>[</mo> <msup> <mi>c</mi> <mi>W</mi> </msup> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>,</mo> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>]</mo> </mrow> <mi>&beta;</mi> </msup> <msup> <mrow> <mo>[</mo> <msup> <mi>s</mi> <mi>W</mi> </msup> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>,</mo> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>]</mo> </mrow> <mi>&gamma;</mi> </msup> <mo>,</mo> </mrow> </math> Wherein α, β and γ are regulatory factors;

if the distortion type of the distorted image Y is Gaussian white noise distortion, calculating the objective quality score of the Y according to the pixel domain-based structural similarity between the two image blocks with the same coordinate positions in the X and the Y, and marking the objective quality score as Q_wn， <math> <mrow> <msub> <mi>Q</mi> <mi>wn</mi> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <mrow> <mi>M</mi> <mo>&times;</mo> <mi>N</mi> </mrow> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>M</mi> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>N</mi> </munderover> <mi>SSIM</mi> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>,</mo> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>;</mo> </mrow> </math>

If the distortion type of the distorted image Y is JPEG distortion, calculating the objective quality score of the Y according to the structural similarity based on the DCT domain between all two image blocks with the same coordinate position in the X and the Y, and recording the objective quality score as the JPEG distortionQ_jpeg， <math> <mrow> <msub> <mi>Q</mi> <mi>jpeg</mi> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <mrow> <mi>M</mi> <mo>&times;</mo> <mi>N</mi> </mrow> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>M</mi> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>N</mi> </munderover> <mi>FSSIM</mi> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>,</mo> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>;</mo> </mrow> </math>

If the distortion type of the distorted image Y is similar to fuzzy distortion, calculating the objective quality score of the Y according to the structural similarity based on the wavelet domain between the two image blocks with the same coordinate positions in the X and the Y, and marking the objective quality score as Q_blur， <math> <mrow> <msub> <mi>Q</mi> <mi>blur</mi> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <mrow> <mi>M</mi> <mo>&times;</mo> <mi>N</mi> </mrow> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>M</mi> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>N</mi> </munderover> <mi>WSSIM</mi> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>,</mo> <msub> <mi>y</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>.</mo> </mrow> </math>

3. The adaptive image quality evaluation method based on distortion type determination according to claim 2, characterized in that: the specific process of determining the distortion type of Y by the distortion type discrimination method in the step I is as follows:

(r-a) dividing X into 64 × 64 non-overlapped blocks to obtain M ' × N ' 64 × 64 image blocks, and marking the image block with X coordinate at (i ', j ') as X '_i',j'For each image block x'_i',j'Performing one-level wavelet decomposition, extracting diagonal components, finding out the median of coefficient amplitude in the diagonal components of each image block, calculating the noise standard deviation of each image block, and dividing x'_i',j'The median value of the coefficient amplitude in the diagonal component of the wavelet is recorded asHaving a noise standard deviation of Wherein, 1≤i'≤M',1≤j'≤N'；

similarly, Y is divided into 64 × 64 non-overlapping blocks to obtain M ' × N ' 64 × 64 image blocks, and the image block with Y coordinates (i ', j ') is denoted as Y '_i',j'For each image block y'_i',j'Performing a first-level waveletDecomposing, extracting diagonal components, finding out the median of coefficient amplitude in the diagonal components of each image block, calculating the noise standard deviation of each image block, and calculating y'_i',j'The median value of the coefficient amplitude in the diagonal component of the wavelet is recorded asHaving a noise standard deviation of <math> <mrow> <msub> <mi>&sigma;</mi> <msub> <msup> <mi>y</mi> <mo>&prime;</mo> </msup> <mrow> <msup> <mi>i</mi> <mo>&prime;</mo> </msup> <mo>,</mo> <msup> <mi>j</mi> <mo>&prime;</mo> </msup> </mrow> </msub> </msub> <mo>=</mo> <mfrac> <msub> <mi>MED</mi> <msub> <msup> <mi>y</mi> <mo>&prime;</mo> </msup> <mrow> <msup> <mi>i</mi> <mo>&prime;</mo> </msup> <mo>,</mo> <msup> <mi>j</mi> <mo>&prime;</mo> </msup> </mrow> </msub> </msub> <mn>0.6745</mn> </mfrac> <mo>;</mo> </mrow> </math>

Calculating the difference of noise standard deviations between all image blocks with the same coordinate position in X and Y, and combining the image blocks X 'at the coordinate positions (i', j ') in X and Y'_i',j'And y'_i',j'Is recorded as the difference of the standard deviation of the noiseThen calculating the mean value of the differences of the noise standard deviations among all the image blocks with the same coordinate position in the X and the Y, and recording the mean value as the difference value <math> <mrow> <mover> <mi>&Delta;&sigma;</mi> <mo>&OverBar;</mo> </mover> <mo>=</mo> <mfrac> <mn>1</mn> <mrow> <msup> <mi>M</mi> <mo>&prime;</mo> </msup> <mo>&times;</mo> <msup> <mi>N</mi> <mo>&prime;</mo> </msup> </mrow> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>i</mi> <mo>&prime;</mo> </msup> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>M</mi> <mo>&prime;</mo> </msup> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <msup> <mi>j</mi> <mo>&prime;</mo> </msup> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mo>&prime;</mo> </msup> </munderover> <mi>&Delta;</mi> <msub> <mi>&sigma;</mi> <mrow> <msup> <mi>i</mi> <mo>&prime;</mo> </msup> <mo>,</mo> <msup> <mi>j</mi> <mo>&prime;</mo> </msup> </mrow> </msub> <mo>;</mo> </mrow> </math>

(ii) judgmentIf yes, determining that the distortion type of Y is Gaussian white noise distortion, and then ending; otherwise, executing the steps (i) and (d); wherein Th_WNA threshold is judged for Gaussian white noise distortion;

computing the brightness difference graph of X, and recording as X^hIs mixing X^hThe coefficient value with the (i ', j') point as the middle coordinate position is recorded as X^h(i″,j″)，X^h(i ", j") ═ X (i ", j") -X (i ", j" +1) |, where,1 is more than or equal to i 'and less than or equal to H,1 is more than or equal to j' -1, X (i ', j') represents the brightness value of the pixel point with the coordinate position (i ', j'), X (i ', j' +1) represents the brightness value of the pixel point with the coordinate position (i ', j' +1) in X, and the symbol "|" is an absolute value symbol;

similarly, a luminance difference map of Y is calculated, denoted as Y^hIs a reaction of Y^hThe coefficient value with the middle coordinate position as the (i ', j') point is recorded as Y^h(i″,j″)，Y^h(i ", j") - | Y (i ", j") -Y (i ", j" +1) |, wherein 1 ≦ i "≦ H,1 ≦ j" ≦ W-1, Y (i ", j") represents the luminance value of the pixel point with the coordinate position (i ", j") in Y, and Y (i ", j" +1) represents the luminance value of the pixel point with the coordinate position (i ", j" +1) in Y;

(ii) luminance difference map X of-e versus X^hDividing the image into 8 × 8 non-overlapping blocks to obtain M '× N' non-overlapping 8 × 8 image blocks, and dividing X^hAn image block with a middle coordinate position at (i ', j') is marked asDefining image blocksRespectively the block energy and the block edge energy ofAnd <math> <mrow> <msubsup> <mi>Ex</mi> <mrow> <msup> <mi>i</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> <mo>,</mo> <msup> <mi>j</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> </mrow> <mi>In</mi> </msubsup> <mo>=</mo> <mfrac> <mn>1</mn> <mn>56</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <mi>p</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <mi>q</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>7</mn> </munderover> <msubsup> <mi>x</mi> <mrow> <msup> <mi>i</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> <mo>,</mo> <msup> <mi>j</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> </mrow> <mi>h</mi> </msubsup> <mrow> <mo>(</mo> <mi>p</mi> <mo>,</mo> <mi>q</mi> <mo>)</mo> </mrow> <mo>,</mo> <msubsup> <mi>Ex</mi> <mrow> <msup> <mi>i</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> <mo>,</mo> <msup> <mi>j</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> </mrow> <mi>Ed</mi> </msubsup> <mo>=</mo> <mfrac> <mn>1</mn> <mn>8</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <mi>p</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <msubsup> <mi>x</mi> <mrow> <msup> <mi>i</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> <mo>,</mo> <msup> <mi>j</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> </mrow> <mi>h</mi> </msubsup> <mrow> <mo>(</mo> <mi>p</mi> <mo>,</mo> <mn>8</mn> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </math> wherein,is composed ofThe middle coordinate position is the coefficient value of (p, q),is composed ofThe middle coordinate position is a coefficient value of (p,8),1≤i″'≤M″,1≤j″'≤N″，1≤p≤8,1≤q≤7；

similarly, a luminance difference map Y for Y^hDividing the image into 8 × 8 non-overlapping blocks to obtain M '× N' non-overlapping 8 × 8 image blocks, and dividing Y into Y blocks^hAn image block with a middle coordinate position at (i ', j') is marked asDefining image blocksRespectively the block energy and the block edge energy ofAnd <math> <mrow> <msubsup> <mi>Ey</mi> <mrow> <msup> <mi>i</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> <mo>,</mo> <msup> <mi>j</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> </mrow> <mi>In</mi> </msubsup> <mo>=</mo> <mfrac> <mn>1</mn> <mn>56</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <mi>p</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <mi>q</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>7</mn> </munderover> <msubsup> <mi>y</mi> <mrow> <msup> <mi>i</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> <mo>,</mo> <msup> <mi>j</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> </mrow> <mi>h</mi> </msubsup> <mrow> <mo>(</mo> <mi>p</mi> <mo>,</mo> <mi>q</mi> <mo>)</mo> </mrow> <mo>,</mo> <msubsup> <mi>Ey</mi> <mrow> <msup> <mi>i</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> <mo>,</mo> <msup> <mi>j</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> </mrow> <mi>Ed</mi> </msubsup> <mo>=</mo> <mfrac> <mn>1</mn> <mn>8</mn> </mfrac> <munderover> <mi>&Sigma;</mi> <mrow> <mi>p</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>8</mn> </munderover> <msubsup> <mi>y</mi> <mrow> <msup> <mi>i</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> <mo>,</mo> <msup> <mi>j</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> </mrow> <mi>h</mi> </msubsup> <mrow> <mo>(</mo> <mi>p</mi> <mo>,</mo> <mn>8</mn> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </math> wherein,is composed ofThe middle coordinate position is the coefficient value of (p, q),is composed ofThe middle coordinate position is the coefficient value of (p, 8);

(ii) f, calculating X^hThe ratio between the edge energy and the intra-block energy of all image blocks in (A) and (B) is defined as X^hImage block with middle coordinate position at (i ', j')The ratio of the block edge energy to the block energy is recorded as <math> <mrow> <msubsup> <mi>R</mi> <mrow> <msup> <mi>i</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> <mo>,</mo> <msup> <mi>j</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> </mrow> <mi>x</mi> </msubsup> <mo>=</mo> <mfrac> <msubsup> <mi>Ex</mi> <mrow> <msup> <mi>i</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> <msup> <mi>j</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> </mrow> <mi>Ed</mi> </msubsup> <msubsup> <mi>Ex</mi> <mrow> <msup> <mi>i</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> <mo>,</mo> <msup> <mi>j</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> </mrow> <mi>In</mi> </msubsup> </mfrac> <mo>;</mo> </mrow> </math>

Likewise, calculate Y^hThe ratio between the edge energy and the intra-block energy of all image blocks in the image block, Y^hImage block with middle coordinate position at (i ', j')Block edge ofThe ratio of the edge energy to the block energy is recorded as <math> <mrow> <msubsup> <mi>R</mi> <mrow> <msup> <mi>i</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> <mo>,</mo> <msup> <mi>j</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> </mrow> <mi>y</mi> </msubsup> <mo>=</mo> <mfrac> <msubsup> <mi>Ey</mi> <mrow> <msup> <mi>i</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> <msup> <mi>j</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> </mrow> <mi>Ed</mi> </msubsup> <msubsup> <mi>Ey</mi> <mrow> <msup> <mi>i</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> <mo>,</mo> <msup> <mi>j</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> </mrow> <mi>In</mi> </msubsup> </mfrac> <mo>;</mo> </mrow> </math>

Statistics satisfy inequalityImage block ofAnd is marked as N₀(ii) a The determination index J is defined and the determination index J, <math> <mrow> <mi>J</mi> <mo>=</mo> <mfrac> <msub> <mi>N</mi> <mn>0</mn> </msub> <mrow> <msup> <mi>M</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> <mo>&times;</mo> <msup> <mi>N</mi> <mrow> <mo>&prime;</mo> <mo>&prime;</mo> </mrow> </msup> </mrow> </mfrac> <mo>;</mo> </mrow> </math>

(ii) g, judgment J>Th_JPEGIf yes, determining that the distortion type of Y is JPEG distortion, and then ending; otherwise, executing the steps of (i) - (h); wherein Th_JPEGA threshold is determined for JPEG distortion;

and (h) determining the distortion type of Y as a fuzzy-like distortion, namely determining the distortion type of Y as a Gaussian fuzzy distortion, or JPEG2000 distortion, or fast fading distortion.

4. The adaptive image quality evaluation method based on distortion type judgment according to claim 3, wherein: in the step I-c, a threshold Th for distinguishing Gaussian white noise distortion_WNIs 0.8; in the steps of (1) and (g), a JPEG distortion discrimination threshold Th_JPEGIs 0.57.

5. The adaptive image quality evaluation method based on distortion type determination according to claim 2, characterized in that: in the fourth step, C is taken₁＝0.01、C₂＝0.02、C₃0.01, 1, γ.