CN107105223A - A kind of tone mapping method for objectively evaluating image quality based on global characteristics - Google Patents
A kind of tone mapping method for objectively evaluating image quality based on global characteristics Download PDFInfo
- Publication number
- CN107105223A CN107105223A CN201710164242.8A CN201710164242A CN107105223A CN 107105223 A CN107105223 A CN 107105223A CN 201710164242 A CN201710164242 A CN 201710164242A CN 107105223 A CN107105223 A CN 107105223A
- Authority
- CN
- China
- Prior art keywords
- pixel value
- designated
- represent
- fitting
- tone mapping
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N17/00—Diagnosis, testing or measuring for television systems or their details
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N17/00—Diagnosis, testing or measuring for television systems or their details
- H04N17/02—Diagnosis, testing or measuring for television systems or their details for colour television signals
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
Method for objectively evaluating image quality is mapped the invention discloses a kind of tone based on global characteristics, it includes two processes of training stage and test phase；That takes into account the influence that natural scene statistical nature and Color Statistical feature are mapped tone, extract the global characteristics vector of tone mapping graph picture, then the global characteristics vector for all tone mapping graph pictures concentrated using support vector regression to training image is trained, and constructs quality prediction model；In test phase, by the global characteristics vector for calculating the tone mapping graph picture as test, and the quality prediction model constructed according to the training stage, prediction obtains the Objective Quality Assessment predicted value of the tone mapping graph picture, because the global characteristics Vector Message of acquisition has stronger stability, and can preferably reflect the quality change situation of tone mapping graph picture, therefore the correlation being effectively improved between objective evaluation result and subjective perception.
Description
Technical field
The present invention relates to a kind of image quality evaluating method, more particularly, to a kind of tone mapping graph based on global characteristics
As assessment method for encoding quality.
Background technology
With the fast development of Display Technique, high dynamic range images (HDR) increasingly attract attention.HDR
The level of image enriches, and can reach the effect of shadow that reality is more approached more than normal image.However, traditional display device is only
The display output of low-dynamic range can be supported.In order to solve real scene and the dynamic range of traditional display device is unmatched
Contradiction, is currently suggested tone mapping (Tone Mapping) algorithm of many high dynamic range images.High dynamic range images
The target of tone-mapping algorithm be by the luminance compression of high dynamic range images to traditional display device acceptable model
Enclose, while retaining the detailed information of artwork as far as possible, and avoid causing image flaw.Therefore, how accurately, objectively to evaluate not
With the performance of tone mapping method, guidance content is made and post-processing has a very important role.
And for tone mapping image quality evaluation, if directly existing image quality evaluating method is applied to
Tone mapping graph picture, then because tone mapping graph picture only has high dynamic range images as reference, therefore can lead to not accurate
Prediction obtains objective evaluation value.Therefore, how visual signature is efficiently extracted out in evaluation procedure so that objective evaluation result
More feel to meet human visual system, be that tone mapping graph picture is being carried out to need to research and solve during evaluating objective quality
The problem of.
The content of the invention
It is objective that the technical problems to be solved by the invention are to provide a kind of tone mapping picture quality based on global characteristics
Evaluation method, it can effectively improve the correlation between objective evaluation result and subjective perception.
The present invention solve the technical scheme that is used of above-mentioned technical problem for：A kind of tone mapping graph based on global characteristics
As assessment method for encoding quality, it is characterised in that including two processes of training stage and test phase；
Described training stage process is concretely comprised the following steps：
1. N width tone mapping graphs _ 1, are chosen as composing training image set, are designated asWherein, N>1,1≤
K≤N,RepresentIn kth width tone mapping graph picture,In every width tone mapping
The width of image is W, and height is H；
1. _ 2, calculateIn every width tone mapping graph picture natural scene statistical nature vector, willNatural scene statistical nature vector be designated asWherein,Dimension be 5 × 1；
1. _ 3, calculateIn every width tone mapping graph picture Color Statistical characteristic vector, will's
Color Statistical characteristic vector is designated asWherein,Dimension be 18 × 1；
1. _ 4, willIn every width tone mapping graph picture natural scene statistical nature vector color
Statistical nature vector is constitutedIn every width tone mapping graph picture global characteristics vector, willThe overall situation
Characteristic vector is designated as Fk,Wherein, FkDimension be 23 × 1, symbol " [] " be vector representation symbol,Representing willWithConnect to form a global characteristics vector；
1. _ 5, willIn all tone mapping graphs as respective global characteristics vector mean subjective
Scored difference composing training sample data sets, and the N number of average master of N number of global characteristics vector is included in training sample data set
See scoring difference；Then using method of the support vector regression as machine learning, to all in training sample data set
Global characteristics vector is trained so that the error between regression function value and mean subjective the scoring difference obtained by training
Minimum, fitting obtains optimal weight vector woptWith optimal bias term bopt；Followed by optimal weight vector woptMost
Excellent bias term bopt, quality prediction model is constructed, f (F) is designated as,Wherein, f () is function table
Show form, F is used for the global characteristics vector for representing tone mapping graph picture, and is used as the input vector of quality prediction model, (wopt
)TFor woptTransposition,For F linear function；
Described test phase process is concretely comprised the following steps：
2. it is used as the tone mapping graph of test as I for any one widthtest, according to step 1. _ 2 to step 1. _ 4 identical
Operation, obtains ItestGlobal characteristics vector, be designated as Ftest；Then the quality prediction model constructed according to the training stage is to Ftest
Tested, prediction obtains FtestCorresponding predicted value, regard the predicted value as ItestObjective Quality Assessment predicted value, be designated as
Qtest,Wherein, ItestWidth be W', and height be H', FtestDimension be 23 × 1,Represent FtestLinear function.
Described step 1. _ 2 inAcquisition process be：
1. _ 2a, calculatingIn every width tone mapping graph picture in all pixels point pixel value it is equal
Value, willIn the average of pixel value of all pixels point be designated as ρ,Then calculateIn every width tone mapping graph picture in all pixels point pixel value standard deviation, willIn institute
The standard deviation for having the pixel value of pixel is designated as δ,Then calculate
In every width tone mapping graph picture in all pixels point pixel value the degree of bias, willIn all pixels point pixel value
The degree of bias be designated as θ,And calculateIn every width tone mapping graph picture in
All pixels point pixel value kurtosis, willIn the kurtosis of pixel value of all pixels point be designated as κ,Calculate againIn every width tone mapping graph picture in all pixels
The entropy of the pixel value of point, willIn the entropy of pixel value of all pixels point be designated as η,Wherein, 1≤
X≤W, 1≤y≤H,RepresentMiddle coordinate position is the pixel value of the pixel of (x, y), 0≤g≤255, pgTable
ShowIn all pixels point pixel value in belong to the probability density function values of g-th of density value,
1. _ 2b, by rightIn all tone mapping graph pictures each in all pixels point pixel
The average of value carries out Gauss Distribution Fitting, and fitting is obtainedIn all tone mapping graph pictures average Gauss
Fitting of distribution curve, the average Gauss Distribution Fitting curve then obtained according to fitting obtains ρ match value, is designated as fρ,Equally, by rightIn all tone mapping graph pictures each in
The standard deviation of pixel value of all pixels point carry out Gauss Distribution Fitting, fitting obtainsIn institute it is coloured
The standard deviation Gauss Distribution Fitting curve of mapping graph picture is adjusted, the standard deviation Gauss Distribution Fitting curve then obtained according to fitting is obtained
δ match value is obtained, f is designated asδ,By rightIn all tones
Mapping graph picture each in the degree of bias of pixel value of all pixels point carry out Gauss Distribution Fitting, fitting obtainsIn all tone mapping graph pictures degree of bias Gauss Distribution Fitting curve, then according to fitting obtain it is inclined
The match value that Gauss Distribution Fitting curve obtains θ is spent, f is designated asθ,By rightIn all tone mapping graph pictures each in all pixels point pixel value kurtosis carry out Gaussian Profile
Fitting, fitting is obtainedIn all tone mapping graph pictures kurtosis Gauss Distribution Fitting curve, Ran Hougen
The kurtosis Gauss Distribution Fitting curve obtained according to fitting obtains κ match value, is designated as fκ,By rightIn all tone mapping graphs as in respective
The entropy of the pixel value of all pixels point carries out Gauss Distribution Fitting, and fitting is obtainedIn all tones mapping
The entropy Gauss Distribution Fitting curve of image, the entropy Gauss Distribution Fitting curve then obtained according to fitting obtains η match value, note
For fη,Wherein, μρAnd σρRepresent the parameter of average Gauss Distribution Fitting curve
Value, exp () is represented using natural radix e as the exponential function at bottom, μδAnd σδRepresent the parameter of standard deviation Gauss Distribution Fitting curve
Value, μθAnd σθRepresent the parameter value of degree of bias Gauss Distribution Fitting curve, μκAnd λκRepresent the parameter of kurtosis Gauss Distribution Fitting curve
Value, μηAnd σηRepresent the parameter value of entropy Gauss Distribution Fitting curve；
1. _ 2c, by fρ、fδ、fθ、fκAnd fηArranged in sequence, is obtained Wherein, symbol
" [] " is vector representation symbol.
Described step 1. _ 3 inAcquisition process be：
1. _ 3a, general{ R is designated as respectively in three components of RGB colork(x,y)}、{Gk(x, y) } and { Bk(x,
Y) }, wherein, 1≤x≤W, 1≤y≤H, Rk(x, y) represents { Rk(x, y) } in coordinate position for (x, y) pixel pixel
Value, Gk(x, y) represents { Gk(x, y) } in coordinate position for (x, y) pixel pixel value, Bk(x, y) represents { Bk(x,y)}
Middle coordinate position is the pixel value of the pixel of (x, y)；
1. _ 3b, to { Rk(x, y) } operation is normalized, by { Rk(x, y) } obtained after normalization operation image note
ForWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) And to { Gk(x, y) } operation is normalized, by { Gk(x, y) } obtained after normalization operation
Image be designated asWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) To { Bk(x, y) } operation is normalized, by { Bk(x, y) } obtained after normalization operation
Image is designated asWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) Wherein,Represent { Rk(x, y) } in all pixels point pixel value average, Represent { Rk(x, y) } in all pixels point pixel value standard deviation, Represent { Gk(x, y) } in all pixels point pixel value average, Represent { Gk(x, y) } in all pixels point pixel value standard deviation, Represent { Bk(x, y) } in all pixels point pixel value average, Represent { Bk(x, y) } in all pixels point pixel value standard deviation,
1. _ 3c, using Generalized Gaussian Distribution Model pairDistribution of color be fitted, fitting is obtainedMatched curve, be designated as gR(h),And use Generalized Gaussian
Distributed model pairDistribution of color be fitted, fitting is obtainedMatched curve, be designated as gG(h),Using Generalized Gaussian Distribution Model pairDistribution of color carry out
Fitting, fitting is obtainedMatched curve, be designated as gB(h),Its
In, 0≤h≤255, αRRepresent matched curve gR(h) scale parameter, βRRepresent matched curve gR(h) form parameter, exp ()
Exponential function of the expression using natural radix e the bottom of as, symbol " | | " it is the symbol that takes absolute value,T is
Integration variable, αGRepresent matched curve gG(h) scale parameter, βGRepresent matched curve gG(h) form parameter,αBRepresent matched curve gB(h) scale parameter, βBRepresent matched curve gB(h) shape ginseng
Number,
1. _ 3d, general{ L is designated as respectively in three components of CIELAB color spacesk(x,y)}、{ak(x, y) } and { bk
(x, y) }, wherein, Lk(x, y) represents { Lk(x, y) } in coordinate position for (x, y) pixel pixel value, ak(x, y) is represented
{ak(x, y) } in coordinate position for (x, y) pixel pixel value, bk(x, y) represents { bk(x, y) } in coordinate position for (x,
Y) pixel value of pixel；
1. _ 3e, to { Lk(x, y) } operation is normalized, by { Lk(x, y) } obtained after normalization operation image note
ForWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) And to { ak(x, y) } operation is normalized, by { ak(x, y) } obtained after normalization operation
Image be designated asWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) To { bk(x, y) } operation is normalized, by { bk(x, y) } obtained after normalization operation
Image is designated asWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) Wherein,Represent { Lk(x, y) } in all pixels point pixel value average, Represent { Lk(x, y) } in all pixels point pixel value standard deviation, Represent { ak(x, y) } in all pixels point pixel value average, Represent { ak(x, y) } in all pixels point pixel value standard deviation, Represent { bk(x, y) } in all pixels point pixel value average, Represent { bk(x, y) } in all pixels point pixel value standard deviation,
1. _ 3f, using Generalized Gaussian Distribution Model pairDistribution of color be fitted, fitting is obtainedMatched curve, be designated as gL(h),And using Generalized Gaussian point
Cloth model pairDistribution of color be fitted, fitting is obtainedMatched curve, be designated as ga(h),Using Generalized Gaussian Distribution Model pairDistribution of color carry out
Fitting, fitting is obtainedMatched curve, be designated as gb(h),Its
In, αLRepresent matched curve gL(h) scale parameter, βLRepresent matched curve gL(h) form parameter,αaRepresent matched curve ga(h) scale parameter, βaRepresent matched curve ga(h) shape ginseng
Number,αbRepresent matched curve gb(h) scale parameter, βbRepresent matched curve gb(h) shape
Parameter,
1. _ 3g, general{ Y is designated as respectively in three components of YCbCr color spacesk(x,y)}、{Uk(x, y) } and { Vk
(x, y) }, wherein, Yk(x, y) represents { Yk(x, y) } in coordinate position for (x, y) pixel pixel value, Uk(x, y) is represented
{Uk(x, y) } in coordinate position for (x, y) pixel pixel value, Vk(x, y) represents { Vk(x, y) } in coordinate position for (x,
Y) pixel value of pixel；
1. _ 3h, to { Yk(x, y) } operation is normalized, by { Yk(x, y) } obtained after normalization operation image note
ForWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) And to { Uk(x, y) } operation is normalized, by { Uk(x, y) } obtained after normalization operation
Image be designated asWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) To { Vk(x, y) } operation is normalized, by { Vk(x, y) } obtained after normalization operation
Image be designated asWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) Wherein,Represent { Yk(x, y) } in all pixels point pixel value average, Represent { Yk(x, y) } in all pixels point pixel value standard deviation, Represent { Uk(x, y) } in all pixels point pixel value average, Represent { Uk(x, y) } in all pixels point pixel value standard deviation, Represent { Vk(x, y) } in all pixels point pixel value average, Represent { Vk(x, y) } in all pixels point pixel value standard deviation,
1. _ 3i, using Generalized Gaussian Distribution Model pairDistribution of color be fitted, fitting is obtainedMatched curve, be designated as gY(h),And using Generalized Gaussian point
Cloth model pairDistribution of color be fitted, fitting is obtainedMatched curve, be designated as gU(h),Using Generalized Gaussian Distribution Model pairDistribution of color carry out
Fitting, fitting is obtainedMatched curve, be designated as gV(h),Its
In, αYRepresent matched curve gY(h) scale parameter, βYRepresent matched curve gY(h) form parameter,αURepresent matched curve gU(h) scale parameter, βURepresent matched curve gU(h) shape ginseng
Number,αVRepresent matched curve gV(h) scale parameter, βVRepresent matched curve gV(h) shape
Parameter,
1. _ 3j, by αR、βR、αG、βG、αB、βB、αL、βL、αa、βa、αb、βb、αY、βY、αU、βU、αVAnd βVArranged in sequence, is obtained Wherein, symbol " [] "
For vector representation symbol.
Compared with prior art, the advantage of the invention is that：
The inventive method considers the influence that natural scene statistical nature and Color Statistical feature are mapped tone, extracts
The global characteristics vector of tone mapping graph picture, all tone mapping graphs then concentrated using support vector regression to training image
The global characteristics vector of picture is trained, and constructs quality prediction model；In test phase, reflected by calculating the tone as test
The global characteristics vector of image, and the quality prediction model constructed according to the training stage are penetrated, prediction obtains the tone mapping graph picture
Objective Quality Assessment predicted value, and can be preferably because the global characteristics Vector Message of acquisition has stronger stability
Reflect the quality change situation of tone mapping graph picture, therefore the phase being effectively improved between objective evaluation result and subjective perception
Guan Xing.
Brief description of the drawings
Fig. 1 realizes block diagram for the totality of the inventive method.
Embodiment
The present invention is described in further detail below in conjunction with accompanying drawing embodiment.
A kind of tone mapping method for objectively evaluating image quality based on global characteristics proposed by the present invention, it is totally realized
Block diagram is as shown in figure 1, it includes two processes of training stage and test phase.
Described training stage process is concretely comprised the following steps：
1. N width tone mapping graphs _ 1, are chosen as composing training image set, are designated asWherein, N>1,
120 width tone mapping graph pictures in TMID databases, the 1811 width colors chosen in ESPL-LIVE databases are chosen in the present embodiment
Tune mapping graph picture, 1≤k≤N,RepresentIn kth width tone mapping graph picture,In
Every width tone mapping graph picture width be W, and height be H.
1. _ 2, calculateIn every width tone mapping graph picture natural scene statistics
(SceneNaturalness) characteristic vector, willNatural scene statistical nature vector be designated asWherein,Dimension
For 5 × 1.
In the present embodiment, step 1. _ 2 inAcquisition process be：
1. _ 2a, calculatingIn every width tone mapping graph picture in all pixels point pixel value it is equal
Value, willIn the average of pixel value of all pixels point be designated as ρ,Then calculateIn every width tone mapping graph picture in all pixels point pixel value standard deviation, willIn institute
The standard deviation for having the pixel value of pixel is designated as δ,Then calculate
In every width tone mapping graph picture in all pixels point pixel value the degree of bias, willIn all pixels point pixel value
The degree of bias be designated as θ,And calculateIn every width tone mapping graph picture in
All pixels point pixel value kurtosis, willIn the kurtosis of pixel value of all pixels point be designated as κ,Calculate againIn every width tone mapping graph picture in all pixels
The entropy of the pixel value of point, willIn the entropy of pixel value of all pixels point be designated as η,Wherein, 1≤
X≤W, 1≤y≤H,RepresentMiddle coordinate position is the pixel value of the pixel of (x, y), 0≤g≤255, pgTable
ShowIn all pixels point pixel value in belong to the probability density function values of g-th of density value,
1. _ 2b, by rightIn all tone mapping graph pictures each in all pixels point pixel
The average of value carries out Gauss Distribution Fitting, and fitting is obtainedIn all tone mapping graph pictures average Gauss
Fitting of distribution curve, the average Gauss Distribution Fitting curve then obtained according to fitting obtains ρ match value, is designated as fρ,Equally, by rightIn all tone mapping graph pictures each in
The standard deviation of pixel value of all pixels point carry out Gauss Distribution Fitting, fitting obtainsIn institute it is coloured
The standard deviation Gauss Distribution Fitting curve of mapping graph picture is adjusted, the standard deviation Gauss Distribution Fitting curve then obtained according to fitting is obtained
δ match value is obtained, f is designated asδ,By rightIn all tones
Mapping graph picture each in the degree of bias of pixel value of all pixels point carry out Gauss Distribution Fitting, fitting obtainsIn all tone mapping graph pictures degree of bias Gauss Distribution Fitting curve, then according to fitting obtain it is inclined
The match value that Gauss Distribution Fitting curve obtains θ is spent, f is designated asθ,By rightIn all tone mapping graph pictures each in all pixels point pixel value kurtosis carry out Gaussian Profile
Fitting, fitting is obtainedIn all tone mapping graph pictures kurtosis Gauss Distribution Fitting curve, then basis
The match value that obtained kurtosis Gauss Distribution Fitting curve obtains κ is fitted, f is designated asκ,By rightIn all tone mapping graphs as in respective
The entropy of the pixel value of all pixels point carries out Gauss Distribution Fitting, and fitting is obtainedIn all tones mapping
The entropy Gauss Distribution Fitting curve of image, the entropy Gauss Distribution Fitting curve then obtained according to fitting obtains η match value, note
For fη,Wherein, μρAnd σρRepresent the parameter of average Gauss Distribution Fitting curve
Value, takes μ in the present embodimentρ=121.70, σρ=36.11, exp () are represented using natural radix e as the exponential function at bottom, μδWith
σδThe parameter value of standard deviation Gauss Distribution Fitting curve is represented, μ is taken in the present embodimentδ=56.47, σδ=18.43, μθAnd σθTable
Show the parameter value of degree of bias Gauss Distribution Fitting curve, μ is taken in the present embodimentθ=0.15, σθ=0.89, μκAnd λκRepresent kurtosis
The parameter value of Gauss Distribution Fitting curve, takes μ in the present embodimentκ=2.82, λκ=18.86, μηAnd σηRepresent entropy Gaussian Profile
The parameter value of matched curve, takes μ in the present embodimentη=7.56, ση=0.27.
1. _ 2c, by fρ、fδ、fθ、fκAnd fηArranged in sequence, is obtained Wherein, symbol
" [] " is vector representation symbol.
1. _ 3, calculateIn every width tone mapping graph picture Color Statistical (Chromatic
Information) characteristic vector, willColor Statistical characteristic vector be designated asWherein,Dimension be 18 × 1.
In the present embodiment, step 1. _ 3 inAcquisition process be：
1. _ 3a, general{ R is designated as respectively in three components of RGB colork(x,y)}、{Gk(x, y) } and { Bk(x,
Y) }, wherein, 1≤x≤W, 1≤y≤H, Rk(x, y) represents { Rk(x, y) } in coordinate position for (x, y) pixel pixel
Value, Gk(x, y) represents { Gk(x, y) } in coordinate position for (x, y) pixel pixel value, Bk(x, y) represents { Bk(x,y)}
Middle coordinate position is the pixel value of the pixel of (x, y).
1. _ 3b, to { Rk(x, y) } operation is normalized, by { Rk(x, y) } obtained after normalization operation image note
ForWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) And to { Gk(x, y) } operation is normalized, by { Gk(x, y) } obtained after normalization operation
Image be designated asWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) To { Bk(x, y) } operation is normalized, by { Bk(x, y) } obtained after normalization operation
Image is designated asWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) Wherein,Represent { Rk(x, y) } in all pixels point pixel value average, Represent { Rk(x, y) } in all pixels point pixel value standard deviation, Represent { Gk(x, y) } in all pixels point pixel value average, Represent { Gk(x, y) } in all pixels point pixel value standard deviation, Represent { Bk(x, y) } in all pixels point pixel value average, Represent { Bk(x, y) } in all pixels point pixel value standard deviation,
1. _ 3c, using existing generalized Gaussian distribution (GGD) model pairDistribution of color be fitted, intend
Conjunction is obtainedMatched curve, be designated as gR(h),And using existing
Some generalized Gaussian distribution (GGD) models pairDistribution of color be fitted, fitting is obtainedFitting
Curve, is designated as gG(h),Using existing generalized Gaussian distribution (GGD) mould
Type pairDistribution of color be fitted, fitting is obtainedMatched curve, be designated as gB(h),Wherein, 0≤h≤255, αRRepresent matched curve gR(h) scale parameter,
βRRepresent matched curve gR(h) form parameter, exponential function of the exp () expressions using natural radix e the bottom of as, symbol " | | " it is to take
Absolute value sign,T is integration variable, αGRepresent matched curve gG(h) scale parameter, βGTable
Show matched curve gG(h) form parameter,αBRepresent matched curve gB(h) scale parameter,
βBRepresent matched curve gB(h) form parameter,
1. _ 3d, general{ L is designated as respectively in three components of CIELAB color spacesk(x,y)}、{ak(x, y) } and { bk
(x, y) }, wherein, Lk(x, y) represents { Lk(x, y) } in coordinate position for (x, y) pixel pixel value, ak(x, y) is represented
{ak(x, y) } in coordinate position for (x, y) pixel pixel value, bk(x, y) represents { bk(x, y) } in coordinate position for (x,
Y) pixel value of pixel.
1. _ 3e, to { Lk(x, y) } operation is normalized, by { Lk(x, y) } obtained after normalization operation image note
ForWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) And to { ak(x, y) } operation is normalized, by { ak(x, y) } obtained after normalization operation
Image be designated asWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) To { bk(x, y) } operation is normalized, by { bk(x, y) } obtained after normalization operation
Image is designated asWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) Wherein,Represent { Lk(x, y) } in all pixels point pixel value average, Represent { Lk(x, y) } in all pixels point pixel value standard deviation, Represent { ak(x, y) } in all pixels point pixel value average, Represent { ak(x, y) } in all pixels point pixel value standard deviation, Represent { bk(x, y) } in all pixels point pixel value average, Represent { bk(x, y) } in all pixels point pixel value standard deviation,
1. _ 3f, using existing generalized Gaussian distribution (GGD) model pairDistribution of color be fitted, intend
Conjunction is obtainedMatched curve, be designated as gL(h),And using existing
Some generalized Gaussian distribution (GGD) models pairDistribution of color be fitted, fitting is obtainedFitting
Curve, is designated as ga(h),Using existing generalized Gaussian distribution (GGD) mould
Type pairDistribution of color be fitted, fitting is obtainedMatched curve, be designated as gb(h),Wherein, αLRepresent matched curve gL(h) scale parameter, βLRepresent fitting
Curve gL(h) form parameter,αaRepresent matched curve ga(h) scale parameter, βaRepresent to intend
Close curve ga(h) form parameter,αbRepresent matched curve gb(h) scale parameter, βbRepresent
Matched curve gb(h) form parameter,
1. _ 3g, general{ Y is designated as respectively in three components of YCbCr color spacesk(x,y)}、{Uk(x, y) } and { Vk
(x, y) }, wherein, Yk(x, y) represents { Yk(x, y) } in coordinate position for (x, y) pixel pixel value, Uk(x, y) is represented
{Uk(x, y) } in coordinate position for (x, y) pixel pixel value, Vk(x, y) represents { Vk(x, y) } in coordinate position for (x,
Y) pixel value of pixel.
1. _ 3h, to { Yk(x, y) } operation is normalized, by { Yk(x, y) } obtained after normalization operation image note
ForWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) And to { Uk(x, y) } operation is normalized, by { Uk(x, y) } obtained after normalization operation
Image be designated asWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) To { Vk(x, y) } operation is normalized, by { Vk(x, y) } obtained after normalization operation
Image is designated asWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) Wherein,Represent { Yk(x, y) } in all pixels point pixel value average, Represent { Yk(x, y) } in all pixels point pixel value standard deviation, Represent { Uk(x, y) } in all pixels point pixel value average, Represent { Uk(x, y) } in all pixels point pixel value standard deviation, Represent { Vk(x, y) } in all pixels point pixel value average, Represent { Vk(x, y) } in all pixels point pixel value standard deviation,
1. _ 3i, using existing generalized Gaussian distribution (GGD) model pairDistribution of color be fitted, intend
Conjunction is obtainedMatched curve, be designated as gY(h),And using existing
Generalized Gaussian distribution (GGD) model pairDistribution of color be fitted, fitting is obtainedFitting it is bent
Line, is designated as gU(h),Using existing generalized Gaussian distribution (GGD) model
It is rightDistribution of color be fitted, fitting is obtainedMatched curve, be designated as gV(h),Wherein, αYRepresent matched curve gY(h) scale parameter, βYRepresent fitting
Curve gY(h) form parameter,αURepresent matched curve gU(h) scale parameter, βURepresent to intend
Close curve gU(h) form parameter,αVRepresent matched curve gV(h) scale parameter, βVRepresent
Matched curve gV(h) form parameter,
1. _ 3j, by αR、βR、αG、βG、αB、βB、αL、βL、αa、βa、αb、βb、αY、βY、αU、βU、αVAnd βVArranged in sequence, is obtained Wherein, symbol
" [] " is vector representation symbol.
1. _ 4, willIn every width tone mapping graph picture natural scene statistical nature vector color
Statistical nature vector is constitutedIn every width tone mapping graph picture global characteristics vector, willThe overall situation
Characteristic vector is designated as Fk,Wherein, FkDimension be 23 × 1, symbol " [] " be vector representation symbol,Representing willWithConnect to form a global characteristics vector.
1. _ 5, willIn all tone mapping graphs as respective global characteristics vector mean subjective
Scored difference composing training sample data sets, and the N number of average master of N number of global characteristics vector is included in training sample data set
See scoring difference；Then using method of the support vector regression as machine learning, to all in training sample data set
Global characteristics vector is trained so that the error between regression function value and mean subjective the scoring difference obtained by training
Minimum, fitting obtains optimal weight vector woptWith optimal bias term bopt；Followed by optimal weight vector woptMost
Excellent bias term bopt, quality prediction model is constructed, f (F) is designated as,Wherein, f () is function table
Show form, F is used for the global characteristics vector for representing tone mapping graph picture, and is used as the input vector of quality prediction model, (wopt
)TFor woptTransposition,For F linear function.
Described test phase process is concretely comprised the following steps：
2. it is used as the tone mapping graph of test as I for any one widthtest, according to step 1. _ 2 to step 1. _ 4 identical
Operation, obtains ItestGlobal characteristics vector, be designated as Ftest；Then the quality prediction model constructed according to the training stage is to Ftest
Tested, prediction obtains FtestCorresponding predicted value, regard the predicted value as ItestObjective Quality Assessment predicted value, be designated as
Qtest,Wherein, ItestWidth be W', and height is H', and W' can be identical with W or differs,
H' can be identical with H or differed, FtestDimension be 23 × 1,Represent FtestLinear function.
In the present embodiment, the TMID databases set up using Canadian University of Waterloo (CA) Waterloo, Ontario, N2L3GI Canada and Texas ,Usa university are difficult to understand
The ESPL-LIVE databases that STING branch school is set up are as tone mapped image data storehouse, and TMID databases reflect including 120 width tones
Image is penetrated, ESPL-LIVE databases include 1811 width tone mapping graph pictures.It is normal using 2 that assess image quality evaluating method
Pearson linearly dependent coefficients (the Pearson linear under the conditions of evaluation index, i.e. nonlinear regression are used as with objective parameter
Correlation coefficient, PLCC) and Spearman orders coefficient of rank correlation (Spearman rank order
Correlation coefficient, SROCC).The evaluation result of the higher explanation the inventive method of PLCC and SROCC is with being averaged
The correlation of subjective scoring difference is better.Table 1 gives Objective Quality Assessment predicted value and the average master that the inventive method is obtained
The correlation between scoring difference is seen, from table 1 it follows that the quality of the tone mapping graph picture obtained using the inventive method
Correlation between objective evaluation predicted value and mean subjective scoring difference is very high, shows objective evaluation result and human eye master
The result that perception is known is more consistent, it is sufficient to illustrate the validity of the inventive method.
It is related between Objective Quality Assessment predicted value and mean subjective scoring difference that table 1 is obtained using the inventive method
Property
| Database | PLCC | SROCC |
| TMID | 0.744 | 0.698 |
| ESPL-LIVE | 0.639 | 0.629 |
Claims (3)
1. a kind of tone mapping method for objectively evaluating image quality based on global characteristics, it is characterised in that including the training stage and
Two processes of test phase；
Described training stage process is concretely comprised the following steps：
1. N width tone mapping graphs _ 1, are chosen as composing training image set, are designated asWherein, N>1,1≤k≤
N,RepresentIn kth width tone mapping graph picture,In every width tone mapping graph
The width of picture is W, and height is H；
1. _ 2, calculateIn every width tone mapping graph picture natural scene statistical nature vector, will's
Natural scene statistical nature vector is designated asWherein,Dimension be 5 × 1；
1. _ 3, calculateIn every width tone mapping graph picture Color Statistical characteristic vector, willColor
Statistical nature vector is designated asWherein,Dimension be 18 × 1；
1. _ 4, willIn every width tone mapping graph picture natural scene statistical nature vector Color Statistical
Characteristic vector is constitutedIn every width tone mapping graph picture global characteristics vector, willGlobal characteristics
Vector is designated as Fk,Wherein, FkDimension be 23 × 1, symbol " [] " be vector representation symbol,
Representing willWithConnect to form a global characteristics vector；
1. _ 5, willIn all tone mapping graphs score as respective global characteristics vector mean subjective
Difference composing training sample data sets, are commented in training sample data set comprising N number of N number of mean subjective of global characteristics vector
Divide difference；Then using method of the support vector regression as machine learning, to all overall situations in training sample data set
Characteristic vector is trained so that the error between regression function value and mean subjective the scoring difference obtained by training is most
Small, fitting obtains optimal weight vector woptWith optimal bias term bopt；Followed by optimal weight vector woptWith it is optimal
Bias term bopt, quality prediction model is constructed, f (F) is designated as,Wherein, f () is function representation
Form, F is used for the global characteristics vector for representing tone mapping graph picture, and is used as the input vector of quality prediction model, (wopt)T
For woptTransposition,For F linear function；
Described test phase process is concretely comprised the following steps：
2. it is used as the tone mapping graph of test as I for any one widthtest, according to step, 1. _ 2 to step, 1. _ 4 identical is operated,
Obtain ItestGlobal characteristics vector, be designated as Ftest；Then the quality prediction model constructed according to the training stage is to FtestSurveyed
Examination, prediction obtains FtestCorresponding predicted value, regard the predicted value as ItestObjective Quality Assessment predicted value, be designated as Qtest,Wherein, ItestWidth be W', and height be H', FtestDimension be 23 × 1,Table
Show FtestLinear function.
2. a kind of tone mapping method for objectively evaluating image quality based on global characteristics according to claim 1, it is special
Levy in being described step 1. _ 2Acquisition process be：
1. _ 2a, calculatingIn every width tone mapping graph picture in all pixels point pixel value average,
WillIn the average of pixel value of all pixels point be designated as ρ,Then calculateIn every width tone mapping graph picture in all pixels point pixel value standard deviation, willIn institute
The standard deviation for having the pixel value of pixel is designated as δ,Then calculate
In every width tone mapping graph picture in all pixels point pixel value the degree of bias, willIn all pixels point pixel value
The degree of bias be designated as,And calculateIn every width tone mapping graph picture in
All pixels point pixel value kurtosis, willIn the kurtosis of pixel value of all pixels point be designated as κ,Calculate againIn every width tone mapping graph picture in all pixels
The entropy of the pixel value of point, willIn the entropy of pixel value of all pixels point be designated as η,Wherein, 1≤
X≤W, 1≤y≤H,RepresentMiddle coordinate position is the pixel value of the pixel of (x, y), 0≤g≤255, pgTable
ShowIn all pixels point pixel value in belong to the probability density function values of g-th of density value,
1. _ 2b, by rightIn all tone mapping graph pictures each in all pixels point pixel value
Average carries out Gauss Distribution Fitting, and fitting is obtainedIn all tone mapping graph pictures average Gaussian Profile
Matched curve, the average Gauss Distribution Fitting curve then obtained according to fitting obtains ρ match value, is designated as fρ,Equally, by rightIn all tone mapping graph pictures each in
The standard deviation of pixel value of all pixels point carry out Gauss Distribution Fitting, fitting obtainsIn institute it is coloured
The standard deviation Gauss Distribution Fitting curve of mapping graph picture is adjusted, the standard deviation Gauss Distribution Fitting curve then obtained according to fitting is obtained
δ match value is obtained, f is designated asδ,By rightIn all tones
Mapping graph picture each in the degree of bias of pixel value of all pixels point carry out Gauss Distribution Fitting, fitting obtainsIn all tone mapping graph pictures degree of bias Gauss Distribution Fitting curve, then according to fitting obtain it is inclined
Gauss Distribution Fitting curve is spent to obtainMatch value, be designated as,By rightIn all tone mapping graph pictures each in all pixels point pixel value kurtosis carry out Gaussian Profile
Fitting, fitting is obtainedIn all tone mapping graph pictures kurtosis Gauss Distribution Fitting curve, Ran Hougen
The kurtosis Gauss Distribution Fitting curve obtained according to fitting obtains κ match value, is designated as fκ,By rightIn all tone mapping graphs as in respective
The entropy of the pixel value of all pixels point carries out Gauss Distribution Fitting, and fitting is obtainedIn all tones mapping
The entropy Gauss Distribution Fitting curve of image, the entropy Gauss Distribution Fitting curve then obtained according to fitting obtains η match value, note
For fη,Wherein, μρAnd σρRepresent the parameter of average Gauss Distribution Fitting curve
Value, exp () is represented using natural radix e as the exponential function at bottom, μδAnd σδRepresent the parameter of standard deviation Gauss Distribution Fitting curve
Value,WithRepresent the parameter value of degree of bias Gauss Distribution Fitting curve, μκAnd λκRepresent the ginseng of kurtosis Gauss Distribution Fitting curve
Numerical value, μηAnd σηRepresent the parameter value of entropy Gauss Distribution Fitting curve；
1. _ 2c, by fρ、fδ、、fκAnd fηArranged in sequence, is obtained Wherein, symbol " [] "
For vector representation symbol.
3. a kind of tone mapping method for objectively evaluating image quality based on global characteristics according to claim 1 or 2, its
In being characterised by described step 1. _ 3Acquisition process be：
1. _ 3a, general{ R is designated as respectively in three components of RGB colork(x,y)}、{Gk(x, y) } and { Bk(x, y) },
Wherein, 1≤x≤W, 1≤y≤H, Rk(x, y) represents { Rk(x, y) } in coordinate position for (x, y) pixel pixel value, Gk
(x, y) represents { Gk(x, y) } in coordinate position for (x, y) pixel pixel value, Bk(x, y) represents { Bk(x, y) } in coordinate
Position is the pixel value of the pixel of (x, y)；
1. _ 3b, to { Rk (x, y) operation } is normalized, by { Rk (x, y) it is (x that the image } obtained after normalization operation, which is designated as middle coordinate position, y) pixel value of pixel is designated as and to { Gk (x, y) operation } is normalized, { Gk (x, y) } is obtained after normalization operation
Image be designated as the pixel value of pixel of (x, y) being designated as that operation is normalized to { Bk (x, y) } by middle coordinate position, { Bk (x, y) } is obtained after normalization operation
It is (x that image, which is designated as middle coordinate position, y) pixel value of pixel is designated as wherein, represent { Rk (x, y) average of the pixel value of all pixels point in }, 3 represent { Rk (x, y) standard deviation of the pixel value of all pixels point in }, represent { Gk (x, y) average of the pixel value of all pixels point in }, represent { Gk (x, y) standard deviation of the pixel value of all pixels point in }, the average of the pixel value of all pixels point in expression { Bk (x, y) }, represents { Bk (x, y) standard deviation of the pixel value of all pixels point in }
<mrow>
<msubsup>
<mi>&sigma;</mi>
<mi>k</mi>
<mi>B</mi>
</msubsup>
<mo>=</mo>
<msqrt>
<mfrac>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>y</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>H</mi>
</munderover>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>x</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>W</mi>
</munderover>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>B</mi>
<mi>k</mi>
</msub>
<mo>(</mo>
<mrow>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
</mrow>
<mo>)</mo>
<mo>-</mo>
<msubsup>
<mi>u</mi>
<mi>k</mi>
<mi>B</mi>
</msubsup>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
<mrow>
<mi>W</mi>
<mo>&times;</mo>
<mi>H</mi>
</mrow>
</mfrac>
</msqrt>
<mo>;</mo>
</mrow>
1. _ 3c, using Generalized Gaussian Distribution Model pairDistribution of color be fitted, fitting is obtained's
Matched curve, is designated as gR(h),And use Generalized Gaussian Distribution Model pairDistribution of color be fitted, fitting is obtainedMatched curve, be designated as gG(h),Using Generalized Gaussian Distribution Model pairDistribution of color carry out
Fitting, fitting is obtainedMatched curve, be designated as gB(h),Its
In, 0≤h≤255, αRRepresent matched curve gR(h) scale parameter, βRRepresent matched curve gR(h) form parameter, exp ()
Exponential function of the expression using natural radix e the bottom of as, symbol " | | " it is the symbol that takes absolute value,t
For integration variable, αGRepresent matched curve gG(h) scale parameter, βGRepresent matched curve gG(h) form parameter,αBRepresent matched curve gB(h) scale parameter, βBRepresent matched curve gB(h) shape ginseng
Number,
1. _ 3d, general{ L is designated as respectively in three components of CIELAB color spacesk(x,y)}、{ak(x, y) } and { bk(x,
Y) }, wherein, Lk(x, y) represents { Lk(x, y) } in coordinate position for (x, y) pixel pixel value, ak(x, y) represents { ak
(x, y) } in coordinate position for (x, y) pixel pixel value, bk(x, y) represents { bk(x, y) } in coordinate position be (x, y)
Pixel pixel value；
1. _ 3e, to { Lk (x, y) operation } is normalized, by { Lk (x, y) it is (x that the image } obtained after normalization operation, which is designated as middle coordinate position, y) pixel value of pixel is designated as and to { ak (x, y) operation } is normalized, { ak (x, y) } is obtained after normalization operation
Image be designated as the pixel value of pixel of (x, y) being designated as that operation is normalized to { bk (x, y) } by middle coordinate position, { bk (x, y) } is obtained after normalization operation
It is (x that image, which is designated as middle coordinate position, y) pixel value of pixel is designated as wherein, represent { Lk (x, y) average of the pixel value of all pixels point in }, represent { Lk (x, y) standard deviation of the pixel value of all pixels point in }, the average of the pixel value of all pixels point in expression { ak (x, y) }, represents { ak (x, y) standard deviation of the pixel value of all pixels point in }, the average of the pixel value of all pixels point in expression { bk (x, y) }, represents { bk (x, y) standard deviation of the pixel value of all pixels point in }
<mrow>
<msubsup>
<mi>&sigma;</mi>
<mi>k</mi>
<mi>b</mi>
</msubsup>
<mo>=</mo>
<msqrt>
<mfrac>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>y</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>H</mi>
</munderover>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>x</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>W</mi>
</munderover>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>b</mi>
<mi>k</mi>
</msub>
<mo>(</mo>
<mrow>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
</mrow>
<mo>)</mo>
<mo>-</mo>
<msubsup>
<mi>u</mi>
<mi>k</mi>
<mi>b</mi>
</msubsup>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
<mrow>
<mi>W</mi>
<mo>&times;</mo>
<mi>H</mi>
</mrow>
</mfrac>
</msqrt>
<mo>;</mo>
</mrow>
1. _ 3f, using Generalized Gaussian Distribution Model pairDistribution of color be fitted, fitting is obtained's
Matched curve, is designated as gL(h),And use Generalized Gaussian Distribution Model pairDistribution of color be fitted, fitting is obtainedMatched curve, be designated as ga(h),Using Generalized Gaussian Distribution Model pairDistribution of color carry out
Fitting, fitting is obtainedMatched curve, be designated as gb(h),Its
In, αLRepresent matched curve gL(h) scale parameter, βLRepresent matched curve gL(h) form parameter,αaRepresent matched curve ga(h) scale parameter, βaRepresent matched curve ga(h) shape ginseng
Number,αbRepresent matched curve gb(h) scale parameter, βbRepresent matched curve gb(h) shape
Parameter,
1. _ 3g, general{ Y is designated as respectively in three components of YCbCr color spacesk(x,y)}、{Uk(x, y) } and { Vk(x,
Y) }, wherein, Yk(x, y) represents { Yk(x, y) } in coordinate position for (x, y) pixel pixel value, Uk(x, y) represents { Uk
(x, y) } in coordinate position for (x, y) pixel pixel value, Vk(x, y) represents { Vk(x, y) } in coordinate position be (x, y)
Pixel pixel value；
1. _ 3h, to { Yk(x, y) } operation is normalized, by { Yk(x, y) } image that is obtained after normalization operation is designated asWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) And to { Uk(x, y) } operation is normalized, by { Uk(x, y) } obtained after normalization operation
Image be designated asWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) To { Vk(x, y) } operation is normalized, by { Vk(x, y) } obtained after normalization operation
Image is designated asWillMiddle coordinate position is designated as the pixel value of the pixel of (x, y) Wherein,Represent { Yk(x, y) } in all pixels point pixel value average, Represent { Yk(x, y) } in all pixels point pixel value standard deviation, Represent { Uk(x, y) } in all pixels point pixel value average, Represent { Uk(x, y) } in all pixels point pixel value standard deviation, Represent { Vk(x, y) } in all pixels point pixel value average, Represent { Vk(x, y) } in all pixels point pixel value standard deviation,
1. _ 3i, using Generalized Gaussian Distribution Model pairDistribution of color be fitted, fitting is obtained's
Matched curve, is designated as gY(h),And use Generalized Gaussian Distribution Model pairDistribution of color be fitted, fitting is obtainedMatched curve, be designated as gU(h),Using Generalized Gaussian Distribution Model pairDistribution of color carry out
Fitting, fitting is obtainedMatched curve, be designated as gV(h),Its
In, αYRepresent matched curve gY(h) scale parameter, βYRepresent matched curve gY(h) form parameter,αURepresent matched curve gU(h) scale parameter, βURepresent matched curve gU(h) shape ginseng
Number,αVRepresent matched curve gV(h) scale parameter, βVRepresent matched curve gV(h) shape
Parameter,
1. _ 3j, by αR、βR、αG、βG、αB、βB、αL、βL、αa、βa、αb、βb、αY、βY、αU、βU、αVAnd βVArranged in sequence, is obtained Wherein, symbol " [] " is arrow
Amount represents symbol.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710164242.8A CN107105223B (en) | 2017-03-20 | 2017-03-20 | A kind of tone mapping method for objectively evaluating image quality based on global characteristics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710164242.8A CN107105223B (en) | 2017-03-20 | 2017-03-20 | A kind of tone mapping method for objectively evaluating image quality based on global characteristics |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107105223A true CN107105223A (en) | 2017-08-29 |
| CN107105223B CN107105223B (en) | 2018-12-07 |
Family
ID=59675339
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710164242.8A Active CN107105223B (en) | 2017-03-20 | 2017-03-20 | A kind of tone mapping method for objectively evaluating image quality based on global characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107105223B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108322733A (en) * | 2018-01-17 | 2018-07-24 | 宁波大学 | It is a kind of without refer to high dynamic range images method for evaluating objective quality |
| CN109218716A (en) * | 2018-10-22 | 2019-01-15 | 天津大学 | Based on color statistics and comentropy without reference tone mapping graph image quality evaluation method |
| CN109788275A (en) * | 2018-12-28 | 2019-05-21 | 天津大学 | Naturality, structure and binocular asymmetry are without reference stereo image quality evaluation method |
| CN109919959A (en) * | 2019-01-24 | 2019-06-21 | 天津大学 | Tone mapping image quality evaluating method based on color, naturality and structure |
| CN110717892A (en) * | 2019-09-18 | 2020-01-21 | 宁波大学 | Tone mapping image quality evaluation method |
| CN110910346A (en) * | 2019-10-17 | 2020-03-24 | 浙江工商职业技术学院 | Tone mapping image quality evaluation method based on dense scale invariant feature transformation |
| WO2021030506A1 (en) * | 2019-08-15 | 2021-02-18 | Dolby Laboratories Licensing Corporation | Efficient user-defined sdr-to-hdr conversion with model templates |
| CN113099215A (en) * | 2021-03-19 | 2021-07-09 | 宁波大学 | Cartoon image quality evaluation method |
| CN117041531A (en) * | 2023-09-04 | 2023-11-10 | 无锡维凯科技有限公司 | Mobile phone camera focusing detection method and system based on image quality evaluation |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103338380A (en) * | 2013-06-06 | 2013-10-02 | 宁波大学 | Adaptive image quality objective evaluation method |
| CN105741328A (en) * | 2016-01-22 | 2016-07-06 | 西安电子科技大学 | Shot image quality evaluation method based on visual perception |
| CN105825500A (en) * | 2016-03-10 | 2016-08-03 | 江苏商贸职业学院 | Camera image quality evaluation method and device |
-
2017
- 2017-03-20 CN CN201710164242.8A patent/CN107105223B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103338380A (en) * | 2013-06-06 | 2013-10-02 | 宁波大学 | Adaptive image quality objective evaluation method |
| CN105741328A (en) * | 2016-01-22 | 2016-07-06 | 西安电子科技大学 | Shot image quality evaluation method based on visual perception |
| CN105825500A (en) * | 2016-03-10 | 2016-08-03 | 江苏商贸职业学院 | Camera image quality evaluation method and device |
Non-Patent Citations (3)
| Title |
|---|
| SHANSHAN WANG: "SUPPORTING BINOCULAR VISUAL QUALITY PREDICTION USING MACHINE LEARNING", 《MULTIMEDIA AND EXPO WORKSHOPS (ICMEW), 2014 IEEE INTERNATIONAL CONFERENCE ON》 * |
| 李柯蒙: "基于双目特征联合的无参考立体图像质量评价", 《光电子.激光》 * |
| 管非凡: "高动态范围图像客观质量评价方法", 《计算机应用》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108322733A (en) * | 2018-01-17 | 2018-07-24 | 宁波大学 | It is a kind of without refer to high dynamic range images method for evaluating objective quality |
| CN109218716A (en) * | 2018-10-22 | 2019-01-15 | 天津大学 | Based on color statistics and comentropy without reference tone mapping graph image quality evaluation method |
| CN109788275A (en) * | 2018-12-28 | 2019-05-21 | 天津大学 | Naturality, structure and binocular asymmetry are without reference stereo image quality evaluation method |
| CN109919959A (en) * | 2019-01-24 | 2019-06-21 | 天津大学 | Tone mapping image quality evaluating method based on color, naturality and structure |
| CN109919959B (en) * | 2019-01-24 | 2023-01-20 | 天津大学 | Tone mapping image quality evaluation method based on color, naturalness and structure |
| WO2021030506A1 (en) * | 2019-08-15 | 2021-02-18 | Dolby Laboratories Licensing Corporation | Efficient user-defined sdr-to-hdr conversion with model templates |
| CN110717892B (en) * | 2019-09-18 | 2022-06-28 | 宁波大学 | Tone mapping image quality evaluation method |
| CN110717892A (en) * | 2019-09-18 | 2020-01-21 | 宁波大学 | Tone mapping image quality evaluation method |
| CN110910346A (en) * | 2019-10-17 | 2020-03-24 | 浙江工商职业技术学院 | Tone mapping image quality evaluation method based on dense scale invariant feature transformation |
| CN113099215A (en) * | 2021-03-19 | 2021-07-09 | 宁波大学 | Cartoon image quality evaluation method |
| CN113099215B (en) * | 2021-03-19 | 2022-06-21 | 宁波大学 | Cartoon image quality evaluation method |
| CN117041531A (en) * | 2023-09-04 | 2023-11-10 | 无锡维凯科技有限公司 | Mobile phone camera focusing detection method and system based on image quality evaluation |
| CN117041531B (en) * | 2023-09-04 | 2024-03-15 | 无锡维凯科技有限公司 | Mobile phone camera focusing detection method and system based on image quality evaluation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107105223B (en) | 2018-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107105223A (en) | A kind of tone mapping method for objectively evaluating image quality based on global characteristics | |
| CN110046673B (en) | No-reference tone mapping image quality evaluation method based on multi-feature fusion | |
| US10210433B2 (en) | Method for evaluating quality of tone-mapping image based on exposure analysis | |
| Qureshi et al. | Towards the design of a consistent image contrast enhancement evaluation measure | |
| CN103996192B (en) | Non-reference image quality evaluation method based on high-quality natural image statistical magnitude model | |
| CN109389591B (en) | Color descriptor-based color image quality evaluation method | |
| CN108090902A (en) | A kind of non-reference picture assessment method for encoding quality based on multiple dimensioned generation confrontation network | |
| CN105261013B (en) | A kind of scan image quality overall evaluation method and evaluation system | |
| CN108010024B (en) | Blind reference tone mapping image quality evaluation method | |
| CN108428227A (en) | Non-reference picture quality appraisement method based on full convolutional neural networks | |
| CN109218716B (en) | No-reference tone mapping image quality evaluation method based on color statistics and information entropy | |
| CN105574901B (en) | A kind of general non-reference picture quality appraisement method based on local contrast pattern | |
| CN110717892B (en) | Tone mapping image quality evaluation method | |
| CN109191460B (en) | Quality evaluation method for tone mapping image | |
| CN107040775B (en) | A kind of tone mapping method for objectively evaluating image quality based on local feature | |
| CN104376565B (en) | Based on discrete cosine transform and the non-reference picture quality appraisement method of rarefaction representation | |
| CN107767363A (en) | It is a kind of based on natural scene without refer to high-dynamics image quality evaluation algorithm | |
| WO2017049703A1 (en) | Image contrast enhancement method | |
| CN105654465B (en) | A kind of stereo image quality evaluation method filtered between the viewpoint using parallax compensation | |
| CN109859166A (en) | It is a kind of based on multiple row convolutional neural networks without ginseng 3D rendering method for evaluating quality | |
| Singh et al. | Image contrast enhancement with brightness preservation using an optimal gamma and logarithmic approach | |
| CN107318014A (en) | The video quality evaluation method of view-based access control model marking area and space-time characterisation | |
| CN106651829A (en) | Non-reference image objective quality evaluation method based on energy and texture analysis | |
| CN102722888A (en) | Stereoscopic image objective quality evaluation method based on physiological and psychological stereoscopic vision | |
| CN106127234A (en) | The non-reference picture quality appraisement method of feature based dictionary |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20201202 Address after: Room 1,020, Nanxun Science and Technology Pioneering Park, No. 666 Chaoyang Road, Nanxun District, Huzhou City, Zhejiang Province, 313000 Patentee after: Huzhou You Yan Intellectual Property Service Co.,Ltd. Address before: 315211 Zhejiang Province, Ningbo Jiangbei District Fenghua Road No. 818 Patentee before: Ningbo University |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20221221 Address after: Floor 4, Unit 1, Building J, No. 6, Zhongtai Road, Loucun Community, Xinhu Street, Guangming District, Shenzhen, Guangdong 518107 Patentee after: Shenzhen Weishi Photoelectric Technology Co.,Ltd. Address before: 313000 room 1020, science and Technology Pioneer Park, 666 Chaoyang Road, Nanxun Town, Nanxun District, Huzhou, Zhejiang. Patentee before: Huzhou You Yan Intellectual Property Service Co.,Ltd. |
|
| TR01 | Transfer of patent right |