CN104902268A - Non-reference three-dimensional image objective quality evaluation method based on local ternary pattern - Google Patents
Non-reference three-dimensional image objective quality evaluation method based on local ternary pattern Download PDFInfo
- Publication number
- CN104902268A CN104902268A CN201510310558.4A CN201510310558A CN104902268A CN 104902268 A CN104902268 A CN 104902268A CN 201510310558 A CN201510310558 A CN 201510310558A CN 104902268 A CN104902268 A CN 104902268A
- Authority
- CN
- China
- Prior art keywords
- dis
- image
- stereo
- distortion
- picture
- 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
- 238000000034 method Methods 0.000 title claims abstract description 56
- 238000013441 quality evaluation Methods 0.000 title claims abstract description 21
- 239000013598 vector Substances 0.000 claims abstract description 68
- 238000011156 evaluation Methods 0.000 claims abstract description 16
- 238000001914 filtration Methods 0.000 claims abstract description 13
- 230000004927 fusion Effects 0.000 claims abstract description 13
- 238000012545 processing Methods 0.000 claims abstract description 5
- 230000000007 visual effect Effects 0.000 claims description 33
- 238000012549 training Methods 0.000 claims description 15
- 238000001303 quality assessment method Methods 0.000 claims description 4
- 238000012360 testing method Methods 0.000 claims description 4
- 230000003044 adaptive effect Effects 0.000 claims description 3
- 238000012886 linear function Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000008447 perception Effects 0.000 abstract description 6
- 238000007619 statistical method Methods 0.000 abstract 1
- 230000004438 eyesight Effects 0.000 description 13
- 230000006870 function Effects 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 4
- 238000013210 evaluation model Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000009412 basement excavation Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 230000000873 masking effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 230000016776 visual perception Effects 0.000 description 1
- 238000005303 weighing Methods 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a non-reference three-dimensional image objective quality evaluation method based on a local ternary pattern. The method comprises the following steps: performing Gaussian gradient filtering on left and right viewpoint images of a distorted three-dimensional image to be evaluated to obtain respective amplitude images and phase images, and calculating a parallax image between the left and right viewpoint images; calculating left and right viewpoint feature fusion images according to the amplitude images, the phase images and the parallel image; processing the left and right viewpoint feature fusion images with local ternary pattern operation to obtain upper and lower mode images of the local ternary pattern; performing statistic operation on the upper and lower mode images with a histogram statistical method to correspondingly obtain an upper mode image histogram statistical feature vector and a lower mode image histogram statistical feature vector; and performing support vector regression prediction according to the histogram statistical feature vectors to obtain an objective quality evaluation predicted value. The non-reference three-dimensional image objective quality evaluation method has the advantage that the relevance between an objective evaluation result and subjective perception can be effectively enhanced.
Description
Technical field
The present invention relates to a kind of stereo image quality evaluation method, especially relate to a kind of nothing based on local tertiary mode with reference to three-dimensional image objective quality evaluation method.
Background technology
Since entering 21st century, along with reaching its maturity of stereoscopic image/video system treatment technology, and the fast development of computer network and the communication technology, cause the tight demand of people's stereoscopic image/video system.Compare traditional one-view image/video system, stereoscopic image/video system is owing to can provide depth information to strengthen the sense of reality of vision, to user with brand-new visual experience on the spot in person more and more welcomed by the people, be considered to the developing direction that Next-Generation Media is main, cause the extensive concern of academia, industrial circle.But people, in order to obtain better three-dimensional telepresenc and visual experience, have higher requirement to stereoscopic vision subjective perceptual quality.Stereoscopic vision subjective perceptual quality is the important indicator weighing stereoscopic image/video systematic function quality.In stereoscopic image/video system, the processing links such as collection, coding, transmission, decoding and display all can introduce certain distortion, the impact that these distortions will produce stereoscopic vision subjective perceptual quality in various degree, therefore how effectively carrying out reference-free quality evaluation is the difficulties needing solution badly.To sum up, evaluate stereo image quality, and the foundation objective evaluation model consistent with subjective quality assessment seems particularly important.At present, researcher proposes much for the nothing reference evaluation method of single viewpoint vision quality, but owing to lacking Systems Theory further investigation stereoscopic vision perception characteristic, therefore also not effectively without reference stereo image quality evaluation method.Comparing single viewpoint vision quality without reference evaluation model, considering that the three-dimensional masking effect of different type of distortion and associated binocular competition/third dimension master factor such as suppression and binocular fusion are on the impact of visual quality without needing with reference to stereo image quality evaluation model.Therefore, can not simply existing single viewpoint vision quality directly be expanded to without in reference stereo image quality evaluation method without reference evaluation model.Existingly mainly carry out prediction and evaluation model by machine learning without reference mass method for objectively evaluating, but for stereo-picture, existing stereo-picture evaluation method or the simple extension of plane picture evaluation method, do not consider binocular vision characteristic, therefore, how characteristic information extraction effectively in evaluation procedure, binocular vision characteristic combination is carried out in evaluation procedure, making objective evaluation result more meet human visual perception system, is the problem that stereo-picture carries out needing in evaluating objective quality process to research and solve.
Summary of the invention
Technical problem to be solved by this invention is to provide a kind of nothing based on local tertiary mode with reference to three-dimensional image objective quality evaluation method, it can fully take into account stereoscopic vision characteristic, thus effectively can improve the correlation between objective evaluation result and subjective perception.
The present invention solves the problems of the technologies described above adopted technical scheme: a kind of nothing based on local tertiary mode is with reference to three-dimensional image objective quality evaluation method, it is characterized in that its processing procedure is: first, Gauss's gradient filtering is implemented respectively to the left visual point image of the stereo-picture of distortion to be evaluated and right visual point image, obtain respective magnitude image and phase image, and calculate the anaglyph between the left visual point image of the stereo-picture of distortion to be evaluated and right visual point image; Secondly, according to the respective magnitude image of the left visual point image of the stereo-picture of distortion to be evaluated and right visual point image and phase image, and the anaglyph between left visual point image and right visual point image, calculate the left and right viewpoint Fusion Features image of the stereo-picture of distortion to be evaluated; Then, adopt local tertiary mode operation to process the left and right viewpoint Fusion Features image of the stereo-picture of distortion to be evaluated, obtain the upper mode image of its local tertiary mode and lower mode image; Afterwards, adopt statistics with histogram method to carry out statistical operation to upper mode image and lower mode image respectively, under correspondence obtains the upper pattern image histogram statistical nature vector sum of the stereo-picture of distortion to be evaluated, pattern image histogram statistical nature is vectorial; Finally, according to pattern image histogram statistical nature vector under the upper pattern image histogram statistical nature vector sum of the stereo-picture of distortion to be evaluated, support vector regression prediction is adopted to obtain the evaluating objective quality predicted value of the stereo-picture of distortion to be evaluated.
This nothing comprises the following steps with reference to three-dimensional image objective quality evaluation method:
1. S is made
disrepresent the stereo-picture of distortion to be evaluated, by S
disleft visual point image be designated as { L
dis(x, y) }, by S
disright visual point image be designated as { R
dis(x, y) }, wherein, 1≤x≤W, 1≤y≤H, W represents S
diswidth, H represents S
disheight, L
dis(x, y) represents { L
dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y), R
dis(x, y) represents { R
dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y);
2. to { L
dis(x, y) } implement Gauss's gradient filtering, obtain { L
dis(x, y) } magnitude image and phase image, correspondence is designated as { G
l_dis(x, y) } and { P
l_dis(x, y) }; Equally, to { R
dis(x, y) } implement Gauss's gradient filtering, obtain { R
dis(x, y) } magnitude image and phase image, correspondence is designated as { G
r_dis(x, y) } and { P
r_dis(x, y) }; Wherein, G
l_dis(x, y) represents { G
l_dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y), P
l_dis(x, y) represents { P
l_dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y), G
r_dis(x, y) represents { G
r_dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y), P
r_dis(x, y) represents { P
r_dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y);
3. block matching method is adopted to calculate { L
dis(x, y) } and { R
dis(x, y) } between anaglyph, be designated as { d
dis(x, y) }, wherein, d
dis(x, y) represents { d
dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y);
4. according to { G
l_dis(x, y) } and { P
l_dis(x, y) }, { G
r_dis(x, y) } and { P
r_dis(x, y) }, { d
dis(x, y) }, calculate S
disleft and right viewpoint Fusion Features image, be designated as { F
dis(x, y) }, by { F
dis(x, y) } in coordinate position be that the pixel value of the pixel of (x, y) is designated as F
dis(x, y),
, wherein, G
r_dis(x+d
dis(x, y), y) represents { G
r_dis(x, y) } in coordinate position be (x+d
dis(x, y), the pixel value of pixel y),
p
r_dis(x+d
dis(x, y), y) represents { P
r_dis(x, y) } in coordinate position be (x+d
dis(x, y), the pixel value of pixel y), cos () is for getting cosine function;
5. adopt local tertiary mode operation to { F
dis(x, y) } process, obtain { F
dis(x, y) } the upper mode image of local tertiary mode and lower mode image, correspondence is designated as { LTP
u(x, y) } and { LTP
d(x, y) }, wherein, LTP
u(x, y) represents { LTP
u(x, y) } in coordinate position be the pixel value of the pixel of (x, y), LTP
d(x, y) represents { LTP
d(x, y) } in coordinate position be the pixel value of the pixel of (x, y);
6. adopt statistics with histogram method to { LTP
u(x, y) } carry out statistical operation, obtain S
disupper pattern image histogram statistical nature vector, be designated as { H
u(m) }; Equally, adopt statistics with histogram method to { LTP
d(x, y) } carry out statistical operation, obtain S
dislower pattern image histogram statistical nature vector, be designated as { H
d(m) }; Wherein, { H
u(m) } dimension be 1 × m' dimension, H
um () represents { H
u(m) } in m element, { H
d(m) } dimension be 1 × m' dimension, H
dm () represents { H
d(m) } in m element, 1≤m≤m', m '=P+2, P represent local tertiary mode operation in field parameter;
7. n is adopted " an original undistorted stereo-picture, set up its distortion stereo-picture set under the different distortion level of different type of distortion, this distortion stereo-picture set comprises several distortion stereo-pictures; Then utilize subjective quality assessment method evaluation to go out the subjective scoring of the every width distortion stereo-picture in this distortion stereo-picture set, the subjective scoring of the jth width distortion stereo-picture in this distortion stereo-picture set is designated as DMOS
j; Again according to step 1. to step operation 6., under obtaining the upper pattern image histogram statistical nature vector sum of the every width distortion stereo-picture in this distortion stereo-picture set in an identical manner, pattern image histogram statistical nature is vectorial, and pattern image histogram statistical nature vector correspondence under the upper pattern image histogram statistical nature vector sum of the jth width distortion stereo-picture in this distortion stereo-picture set is designated as { H
u,j(m) } and { H
d,j(m) }; Wherein, n " initial value of >1, j is 1,1≤j≤N', N' represent total width number of the distortion stereo-picture comprised in this distortion stereo-picture set, 0≤DMOS
j≤ 100, { H
u,j(m) } and { H
d,j(m) } dimension be 1 × m' dimension, H
u,jm () represents { H
u,j(m) } in m element, H
d,jm () represents { H
d,j(m) } in m element, 1≤m≤m', m '=P+2, P represent local tertiary mode operation in field parameter;
8. using this distorted image set as training set; Then support vector regression is utilized to train pattern image histogram statistical nature vector under the subjective scoring of all distortion stereo-pictures in training set and upper pattern image histogram statistical nature vector sum, make through training the error between regression function value and subjective scoring obtained minimum, matching obtains optimum weighted vector W
optwith the bias term b of optimum
opt; Then W is utilized
optand b
optstructure obtains support vector regression training pattern; Again according to support vector regression training pattern, to S
disupper pattern image histogram statistical nature vector { H
u(m) } and lower pattern image histogram statistical nature vector { H
d(m) } to test, prediction obtains S
disevaluating objective quality predicted value, be designated as Q, Q=f (x),
wherein, Q is the function of x, and f () is function representation form, and x is input, and x represents S
disupper pattern image histogram statistical nature vector { H
u(m) } and lower pattern image histogram statistical nature vector { H
d(m) }, (W
opt)
tfor W
opttransposed vector,
for the linear function of x.
Described step 2. in Gauss's gradient filtering in scale parameter σ value be 0.5.
Described step 5. in the operation of local tertiary mode in field parameter P value be 8, local radius parameter R value is 1, adaptive thresholding value matrix { T (x, y) (x is designated as under }, value T (x, the y) value of element y) is α × F
dis(x, y), wherein, α is intensity factor, gets α=0.05.
Compared with prior art, the invention has the advantages that: by deep excavation stereoscopic vision perception characteristic, local tertiary mode operation is carried out to the left and right viewpoint Fusion Features image of the stereo-picture of distortion to be evaluated, obtain the upper mode image of its local tertiary mode and lower mode image, statistics with histogram method is adopted to carry out statistical operation to upper mode image and lower mode image respectively again, under correspondence obtains the upper pattern image histogram statistical nature vector sum of the stereo-picture of distortion to be evaluated, pattern image histogram statistical nature is vectorial, characteristic vector pickup method is simple, computation complexity is low, eigenvector information due to the stereo-picture of the distortion to be evaluated of acquisition can reflect the mass change situation of the stereo-picture of distortion to be evaluated preferably, that is: the evaluating objective quality predicted value of the stereo-picture of the distortion to be evaluated obtained can be enable to reflect human eye vision subjective perceptual quality exactly, effectively can improve the correlation of objective evaluation result and subjective perception.
Accompanying drawing explanation
Fig. 1 be the inventive method totally realize block diagram.
Embodiment
Below in conjunction with accompanying drawing embodiment, the present invention is described in further detail.
A kind of nothing based on local tertiary mode that the present invention proposes is with reference to three-dimensional image objective quality evaluation method, it totally realizes block diagram as shown in Figure 1, its processing procedure is: first, Gauss's gradient filtering is implemented respectively to the left visual point image of the stereo-picture of distortion to be evaluated and right visual point image, obtain respective magnitude image and phase image, and calculate the anaglyph between the left visual point image of the stereo-picture of distortion to be evaluated and right visual point image; Secondly, according to the respective magnitude image of the left visual point image of the stereo-picture of distortion to be evaluated and right visual point image and phase image, and the anaglyph between left visual point image and right visual point image, calculate the left and right viewpoint Fusion Features image of the stereo-picture of distortion to be evaluated; Then, adopt local tertiary mode operation to process the left and right viewpoint Fusion Features image of the stereo-picture of distortion to be evaluated, obtain the upper mode image of its local tertiary mode and lower mode image; Afterwards, adopt statistics with histogram method to carry out statistical operation to upper mode image and lower mode image respectively, under correspondence obtains the upper pattern image histogram statistical nature vector sum of the stereo-picture of distortion to be evaluated, pattern image histogram statistical nature is vectorial; Finally, according to pattern image histogram statistical nature vector under the upper pattern image histogram statistical nature vector sum of the stereo-picture of distortion to be evaluated, support vector regression prediction is adopted to obtain the evaluating objective quality predicted value of the stereo-picture of distortion to be evaluated.
Of the present invention without reference three-dimensional image objective quality evaluation method, it comprises the following steps:
1. S is made
disrepresent the stereo-picture of distortion to be evaluated, by S
disleft visual point image be designated as { L
dis(x, y) }, by S
disright visual point image be designated as { R
dis(x, y) }, wherein, 1≤x≤W, 1≤y≤H, W represents S
diswidth, H represents S
disheight, L
dis(x, y) represents { L
dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y), R
dis(x, y) represents { R
dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y).
2. adopt prior art to { L
dis(x, y) } implement Gauss's gradient filtering, obtain { L
dis(x, y) } magnitude image and phase image, correspondence is designated as { G
l_dis(x, y) } and { P
l_dis(x, y) }; Equally, to { R
dis(x, y) } implement Gauss's gradient filtering, obtain { R
dis(x, y) } magnitude image and phase image, correspondence is designated as { G
r_dis(x, y) } and { P
r_dis(x, y) }; Wherein, G
l_dis(x, y) represents { G
l_dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y), P
l_dis(x, y) represents { P
l_dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y), G
r_dis(x, y) represents { G
r_dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y), P
r_dis(x, y) represents { P
r_dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y).
In the present embodiment, the scale parameter σ in Gauss's gradient filtering can value be σ=0.5.
3. existing block matching method is adopted to calculate { L
dis(x, y) } and { R
dis(x, y) } between anaglyph, be designated as { d
dis(x, y) }, wherein, d
dis(x, y) represents { d
dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y).
4. according to { G
l_dis(x, y) } and { P
l_dis(x, y) }, { G
r_dis(x, y) } and { P
r_dis(x, y) }, { d
dis(x, y) }, calculate S
disleft and right viewpoint Fusion Features image, be designated as { F
dis(x, y) }, by { F
dis(x, y) } in coordinate position be that the pixel value of the pixel of (x, y) is designated as F
dis(x, y),
, wherein, G
r_dis(x+d
dis(x, y), y) represents { G
r_dis(x, y) } in coordinate position be (x+d
dis(x, y), the pixel value of pixel y),
p
r_dis(x+d
dis(x, y), y) represents { P
r_dis(x, y) } in coordinate position be (x+d
dis(x, y), the pixel value of pixel y), cos () is for getting cosine function.
5. adopt the tertiary mode operation of existing local to { F
dis(x, y) } process, obtain { F
dis(x, y) } the upper mode image of local tertiary mode and lower mode image, correspondence is designated as { LTP
u(x, y) } and { LTP
d(x, y) }, wherein, LTP
u(x, y) represents { LTP
u(x, y) } in coordinate position be the pixel value of the pixel of (x, y), LTP
d(x, y) represents { LTP
d(x, y) } in coordinate position be the pixel value of the pixel of (x, y).
In the present embodiment, field parameter P value in the tertiary mode operation of local is 8, local radius parameter R value is 1, adaptive thresholding value matrix { T (x, y) value T (x, the y) value being designated as the element of (x, y) under } is α × F
dis(x, y), wherein, α is intensity factor, gets α=0.05.
6. adopt existing statistics with histogram method to { LTP
u(x, y) } carry out statistical operation, obtain S
disupper pattern image histogram statistical nature vector, be designated as { H
u(m) }; Equally, adopt statistics with histogram method to { LTP
d(x, y) } carry out statistical operation, obtain S
dislower pattern image histogram statistical nature vector, be designated as { H
d(m) }; Wherein, { H
u(m) } dimension be 1 × m' dimension, H
um () represents { H
u(m) } in m element, { H
d(m) } dimension be 1 × m' dimension, H
dm () represents { H
d(m) } in m element, 1≤m≤m', m '=P+2, P represent local tertiary mode operation in field parameter.
7. n is adopted " an original undistorted stereo-picture, set up its distortion stereo-picture set under the different distortion level of different type of distortion, this distortion stereo-picture set comprises several distortion stereo-pictures; Then utilize existing subjective quality assessment method evaluation to go out the subjective scoring of the every width distortion stereo-picture in this distortion stereo-picture set, the subjective scoring of the jth width distortion stereo-picture in this distortion stereo-picture set is designated as DMOS
j; Again according to step 1. to step operation 6., under obtaining the upper pattern image histogram statistical nature vector sum of the every width distortion stereo-picture in this distortion stereo-picture set in an identical manner, pattern image histogram statistical nature is vectorial, and pattern image histogram statistical nature vector correspondence under the upper pattern image histogram statistical nature vector sum of the jth width distortion stereo-picture in this distortion stereo-picture set is designated as { H
u,j(m) } and { H
d,j(m) }; Wherein, n " >1, as got n "=the initial value of 3, j is 1,1≤j≤N', N' represent total width number of the distortion stereo-picture comprised in this distortion stereo-picture set, 0≤DMOS
j≤ 100, { H
u,j(m) } and { H
d,j(m) } dimension be 1 × m' dimension, H
u,jm () represents { H
u,j(m) } in m element, H
d,jm () represents { H
d,j(m) } in m element, 1≤m≤m', m '=P+2, P represent local tertiary mode operation in field parameter.
8. support vector regression (Support Vector Regression, SVR) be new machine learning method and the statistical theory of structure based risk minimization criterion, it can suppress over-fitting problem effectively, therefore the present invention using this distorted image set as training set; Then support vector regression is utilized to train pattern image histogram statistical nature vector under the subjective scoring of all distortion stereo-pictures in training set and upper pattern image histogram statistical nature vector sum, make through training the error between regression function value and subjective scoring obtained minimum, matching obtains optimum weighted vector W
optwith the bias term b of optimum
opt; Then W is utilized
optand b
optstructure obtains support vector regression training pattern; Again according to support vector regression training pattern, to S
disupper pattern image histogram statistical nature vector { H
u(m) } and lower pattern image histogram statistical nature vector { H
d(m) } to test, prediction obtains S
disevaluating objective quality predicted value, be designated as Q, Q=f (x),
wherein, Q is the function of x, and f () is function representation form, and x is input, and x represents S
disupper pattern image histogram statistical nature vector { H
u(m) } and lower pattern image histogram statistical nature vector { H
d(m) }, (W
opt)
tfor W
opttransposed vector,
for the linear function of x.
In order to verify feasibility and the validity of the inventive method further, test.
At this, the correlation adopting LIVE stereo-picture distortion storehouse to come the evaluating objective quality predicted value of the stereo-picture of the distortion that analysis and utilization the inventive method obtains and mean subjective to mark between difference.Here, utilize 3 of evaluate image quality evaluating method conventional objective parameters as evaluation index, namely Pearson correlation coefficient (the Pearson linear correlation coefficient under nonlinear regression condition, PLCC), Spearman coefficient correlation (Spearman rank order correlation coefficient, SROCC), mean square error (root mean squared error, RMSE), PLCC and RMSE reflects the accuracy of the evaluating objective quality predicted value of the stereo-picture of distortion, and SROCC reflects its monotonicity.
Utilize the inventive method to calculate the evaluating objective quality predicted value of the every width distortion stereo-picture in LIVE stereo-picture distortion storehouse, recycle the mean subjective scoring difference that existing subjective evaluation method obtains the every width distortion stereo-picture in LIVE stereo-picture distortion storehouse.The evaluating objective quality predicted value of the distortion stereo-picture calculated by the inventive method is done five parameter Logistic function nonlinear fittings, PLCC and SROCC value is higher, and the correlation that the objective evaluation result of the lower explanation method for objectively evaluating of RMSE value and mean subjective are marked between difference is better.PLCC, SROCC and RMSE coefficient correlation of the quality evaluation performance of reflection the inventive method as listed in table 1.From the data listed by table 1, final evaluating objective quality predicted value and the mean subjective correlation of marking between difference of the distortion stereo-picture obtained by the inventive method are good, show that the result of objective evaluation result and human eye subjective perception is more consistent, be enough to feasibility and validity that the inventive method is described.
The correlation that the evaluating objective quality predicted value of the stereo-picture of the distortion that table 1 utilizes the inventive method to obtain and mean subjective are marked between difference
Claims (4)
1. the nothing reference three-dimensional image objective quality evaluation method based on local tertiary mode, it is characterized in that its processing procedure is: first, Gauss's gradient filtering is implemented respectively to the left visual point image of the stereo-picture of distortion to be evaluated and right visual point image, obtain respective magnitude image and phase image, and calculate the anaglyph between the left visual point image of the stereo-picture of distortion to be evaluated and right visual point image; Secondly, according to the respective magnitude image of the left visual point image of the stereo-picture of distortion to be evaluated and right visual point image and phase image, and the anaglyph between left visual point image and right visual point image, calculate the left and right viewpoint Fusion Features image of the stereo-picture of distortion to be evaluated; Then, adopt local tertiary mode operation to process the left and right viewpoint Fusion Features image of the stereo-picture of distortion to be evaluated, obtain the upper mode image of its local tertiary mode and lower mode image; Afterwards, adopt statistics with histogram method to carry out statistical operation to upper mode image and lower mode image respectively, under correspondence obtains the upper pattern image histogram statistical nature vector sum of the stereo-picture of distortion to be evaluated, pattern image histogram statistical nature is vectorial; Finally, according to pattern image histogram statistical nature vector under the upper pattern image histogram statistical nature vector sum of the stereo-picture of distortion to be evaluated, support vector regression prediction is adopted to obtain the evaluating objective quality predicted value of the stereo-picture of distortion to be evaluated.
2. the nothing based on local tertiary mode according to claim 1 is with reference to three-dimensional image objective quality evaluation method, it is characterized in that comprising the following steps:
1. S is made
disrepresent the stereo-picture of distortion to be evaluated, by S
disleft visual point image be designated as { L
dis(x, y) }, by S
disright visual point image be designated as { R
dis(x, y) }, wherein, 1≤x≤W, 1≤y≤H, W represents S
diswidth, H represents S
disheight, L
dis(x, y) represents { L
dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y), R
dis(x, y) represents { R
dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y);
2. to { L
dis(x, y) } implement Gauss's gradient filtering, obtain { L
dis(x, y) } magnitude image and phase image, correspondence is designated as { G
l_dis(x, y) } and { P
l_dis(x, y) }; Equally, to { R
dis(x, y) } implement Gauss's gradient filtering, obtain { R
dis(x, y) } magnitude image and phase image, correspondence is designated as { G
r_dis(x, y) } and { P
r_dis(x, y) }; Wherein, G
l_dis(x, y) represents { G
l_dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y), P
l_dis(x, y) represents { P
l_dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y), G
r_dis(x, y) represents { G
r_dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y), P
r_dis(x, y) represents { P
r_dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y);
3. block matching method is adopted to calculate { L
dis(x, y) } and { R
dis(x, y) } between anaglyph, be designated as { d
dis(x, y) }, wherein, d
dis(x, y) represents { d
dis(x, y) } in coordinate position be the pixel value of the pixel of (x, y);
4. according to { G
l_dis(x, y) } and { P
l_dis(x, y) }, { G
r_dis(x, y) } and { P
r_dis(x, y) }, { d
dis(x, y) }, calculate S
disleft and right viewpoint Fusion Features image, be designated as { F
dis(x, y) }, by { F
dis(x, y) } in coordinate position be that the pixel value of the pixel of (x, y) is designated as F
dis(x, y),
, wherein, G
r_dis(x+d
dis(x, y), y) represents { G
r_dis(x, y) } in coordinate position be (x+d
dis(x, y), the pixel value of pixel y),
p
r_dis(x+d
dis(x, y), y) represents { P
r_dis(x, y) } in coordinate position be (x+d
dis(x, y), the pixel value of pixel y), cos () is for getting cosine function;
5. adopt local tertiary mode operation to { F
dis(x, y) } process, obtain { F
dis(x, y) } the upper mode image of local tertiary mode and lower mode image, correspondence is designated as { LTP
u(x, y) } and { LTP
d(x, y) }, wherein, LTP
u(x, y) represents { LTP
u(x, y) } in coordinate position be the pixel value of the pixel of (x, y), LTP
d(x, y) represents { LTP
d(x, y) } in coordinate position be the pixel value of the pixel of (x, y);
6. adopt statistics with histogram method to { LTP
u(x, y) } carry out statistical operation, obtain S
disupper pattern image histogram statistical nature vector, be designated as { H
u(m) }; Equally, adopt statistics with histogram method to { LTP
d(x, y) } carry out statistical operation, obtain S
dislower pattern image histogram statistical nature vector, be designated as { H
d(m) }; Wherein, { H
u(m) } dimension be 1 × m' dimension, H
um () represents { H
u(m) } in m element, { H
d(m) } dimension be 1 × m' dimension, H
dm () represents { H
d(m) } in m element, 1≤m≤m', m '=P+2, P represent local tertiary mode operation in field parameter;
7. n is adopted " an original undistorted stereo-picture, set up its distortion stereo-picture set under the different distortion level of different type of distortion, this distortion stereo-picture set comprises several distortion stereo-pictures; Then utilize subjective quality assessment method evaluation to go out the subjective scoring of the every width distortion stereo-picture in this distortion stereo-picture set, the subjective scoring of the jth width distortion stereo-picture in this distortion stereo-picture set is designated as DMOS
j; Again according to step 1. to step operation 6., under obtaining the upper pattern image histogram statistical nature vector sum of the every width distortion stereo-picture in this distortion stereo-picture set in an identical manner, pattern image histogram statistical nature is vectorial, and pattern image histogram statistical nature vector correspondence under the upper pattern image histogram statistical nature vector sum of the jth width distortion stereo-picture in this distortion stereo-picture set is designated as { H
u,j(m) } and { H
d,j(m) }; Wherein, n " initial value of >1, j is 1,1≤j≤N', N' represent total width number of the distortion stereo-picture comprised in this distortion stereo-picture set, 0≤DMOS
j≤ 100, { H
u,j(m) } and { H
d,j(m) } dimension be 1 × m' dimension, H
u,jm () represents { H
u,j(m) } in m element, H
d,jm () represents { H
d,j(m) } in m element, 1≤m≤m', m '=P+2, P represent local tertiary mode operation in field parameter;
8. using this distorted image set as training set; Then support vector regression is utilized to train pattern image histogram statistical nature vector under the subjective scoring of all distortion stereo-pictures in training set and upper pattern image histogram statistical nature vector sum, make through training the error between regression function value and subjective scoring obtained minimum, matching obtains optimum weighted vector W
optwith the bias term b of optimum
opt; Then W is utilized
optand b
optstructure obtains support vector regression training pattern; Again according to support vector regression training pattern, to S
disupper pattern image histogram statistical nature vector { H
u(m) } and lower pattern image histogram statistical nature vector { H
d(m) } to test, prediction obtains S
disevaluating objective quality predicted value, be designated as Q, Q=f (x),
wherein, Q is the function of x, and f () is function representation form, and x is input, and x represents S
disupper pattern image histogram statistical nature vector { H
u(m) } and lower pattern image histogram statistical nature vector { H
d(m) }, (W
opt)
tfor W
opttransposed vector,
for the linear function of x.
3. the nothing based on local tertiary mode according to claim 2 is with reference to three-dimensional image objective quality evaluation method, it is characterized in that the scale parameter σ value in the Gauss's gradient filtering during described step is 2. 0.5.
4. the nothing based on local tertiary mode according to Claims 2 or 3 is with reference to three-dimensional image objective quality evaluation method, the field parameter P value that it is characterized in that in the local tertiary mode operation during described step is 5. 8, local radius parameter R value is 1, adaptive thresholding value matrix { T (x, y) (x is designated as under }, value T (x, the y) value of element y) is α × F
dis(x, y), wherein, α is intensity factor, gets α=0.05.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510310558.4A CN104902268B (en) | 2015-06-08 | 2015-06-08 | Based on local tertiary mode without with reference to three-dimensional image objective quality evaluation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510310558.4A CN104902268B (en) | 2015-06-08 | 2015-06-08 | Based on local tertiary mode without with reference to three-dimensional image objective quality evaluation method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104902268A true CN104902268A (en) | 2015-09-09 |
| CN104902268B CN104902268B (en) | 2016-12-07 |
Family
ID=54034628
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510310558.4A Active CN104902268B (en) | 2015-06-08 | 2015-06-08 | Based on local tertiary mode without with reference to three-dimensional image objective quality evaluation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104902268B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105282543A (en) * | 2015-10-26 | 2016-01-27 | 浙江科技学院 | Total blindness three-dimensional image quality objective evaluation method based on three-dimensional visual perception |
| CN105357519A (en) * | 2015-12-02 | 2016-02-24 | 浙江科技学院 | Quality objective evaluation method for three-dimensional image without reference based on self-similarity characteristic |
| CN105488792A (en) * | 2015-11-26 | 2016-04-13 | 浙江科技学院 | No-reference stereo image quality evaluation method based on dictionary learning and machine learning |
| CN105574901A (en) * | 2016-01-18 | 2016-05-11 | 浙江科技学院 | General reference-free image quality evaluation method based on local contrast mode |
| CN105979253A (en) * | 2016-05-06 | 2016-09-28 | 浙江科技学院 | Generalized regression neural network based non-reference stereoscopic image quality evaluation method |
| CN106683079A (en) * | 2016-12-14 | 2017-05-17 | 浙江科技学院 | No-reference image objective quality evaluation method based on structural distortion |
| CN107040775A (en) * | 2017-03-20 | 2017-08-11 | 宁波大学 | A kind of tone mapping method for objectively evaluating image quality based on local feature |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120262549A1 (en) * | 2011-04-15 | 2012-10-18 | Tektronix, Inc. | Full Reference System For Predicting Subjective Quality Of Three-Dimensional Video |
| CN103996202A (en) * | 2014-06-11 | 2014-08-20 | 北京航空航天大学 | Stereo matching method based on hybrid matching cost and adaptive window |
| CN104243976A (en) * | 2014-09-23 | 2014-12-24 | 浙江科技学院 | Stereo image objective quality evaluation method |
-
2015
- 2015-06-08 CN CN201510310558.4A patent/CN104902268B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120262549A1 (en) * | 2011-04-15 | 2012-10-18 | Tektronix, Inc. | Full Reference System For Predicting Subjective Quality Of Three-Dimensional Video |
| CN103996202A (en) * | 2014-06-11 | 2014-08-20 | 北京航空航天大学 | Stereo matching method based on hybrid matching cost and adaptive window |
| CN104243976A (en) * | 2014-09-23 | 2014-12-24 | 浙江科技学院 | Stereo image objective quality evaluation method |
Non-Patent Citations (1)
| Title |
|---|
| 张春风: "基于多特征的行人检测技术研究", 《万方》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105282543A (en) * | 2015-10-26 | 2016-01-27 | 浙江科技学院 | Total blindness three-dimensional image quality objective evaluation method based on three-dimensional visual perception |
| CN105488792B (en) * | 2015-11-26 | 2017-11-28 | 浙江科技学院 | Based on dictionary learning and machine learning without referring to stereo image quality evaluation method |
| CN105488792A (en) * | 2015-11-26 | 2016-04-13 | 浙江科技学院 | No-reference stereo image quality evaluation method based on dictionary learning and machine learning |
| CN105357519B (en) * | 2015-12-02 | 2017-05-24 | 浙江科技学院 | Quality objective evaluation method for three-dimensional image without reference based on self-similarity characteristic |
| CN105357519A (en) * | 2015-12-02 | 2016-02-24 | 浙江科技学院 | Quality objective evaluation method for three-dimensional image without reference based on self-similarity characteristic |
| CN105574901A (en) * | 2016-01-18 | 2016-05-11 | 浙江科技学院 | General reference-free image quality evaluation method based on local contrast mode |
| CN105574901B (en) * | 2016-01-18 | 2018-10-16 | 浙江科技学院 | A kind of general non-reference picture quality appraisement method based on local contrast pattern |
| CN105979253A (en) * | 2016-05-06 | 2016-09-28 | 浙江科技学院 | Generalized regression neural network based non-reference stereoscopic image quality evaluation method |
| CN105979253B (en) * | 2016-05-06 | 2017-11-28 | 浙江科技学院 | Based on generalized regression nerve networks without with reference to stereo image quality evaluation method |
| CN106683079A (en) * | 2016-12-14 | 2017-05-17 | 浙江科技学院 | No-reference image objective quality evaluation method based on structural distortion |
| CN106683079B (en) * | 2016-12-14 | 2019-05-17 | 浙江科技学院 | A kind of non-reference picture method for evaluating objective quality based on structure distortion |
| CN107040775A (en) * | 2017-03-20 | 2017-08-11 | 宁波大学 | A kind of tone mapping method for objectively evaluating image quality based on local feature |
| CN107040775B (en) * | 2017-03-20 | 2019-01-15 | 宁波大学 | A kind of tone mapping method for objectively evaluating image quality based on local feature |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104902268B (en) | 2016-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104658001B (en) | Non-reference asymmetric distorted stereo image objective quality assessment method | |
| CN104902268A (en) | Non-reference three-dimensional image objective quality evaluation method based on local ternary pattern | |
| CN104902267B (en) | No-reference image quality evaluation method based on gradient information | |
| CN105376563B (en) | No-reference three-dimensional image quality evaluation method based on binocular fusion feature similarity | |
| CN105979253B (en) | Based on generalized regression nerve networks without with reference to stereo image quality evaluation method | |
| CN105357519B (en) | Quality objective evaluation method for three-dimensional image without reference based on self-similarity characteristic | |
| CN103347196B (en) | Method for evaluating stereo image vision comfort level based on machine learning | |
| CN105282543B (en) | Total blindness three-dimensional image quality objective evaluation method based on three-dimensional visual perception | |
| CN104243976B (en) | A kind of three-dimensional image objective quality evaluation method | |
| CN104658002B (en) | Non-reference image objective quality evaluation method | |
| CN106791822B (en) | It is a kind of based on single binocular feature learning without reference stereo image quality evaluation method | |
| CN101610425B (en) | Method for evaluating stereo image quality and device | |
| CN105407349A (en) | No-reference objective three-dimensional image quality evaluation method based on binocular visual perception | |
| CN105574901B (en) | A kind of general non-reference picture quality appraisement method based on local contrast pattern | |
| CN104036501A (en) | Three-dimensional image quality objective evaluation method based on sparse representation | |
| CN104361583B (en) | A kind of method determining asymmetric distortion three-dimensional image objective quality | |
| CN104408716A (en) | Three-dimensional image quality objective evaluation method based on visual fidelity | |
| CN104581143A (en) | Reference-free three-dimensional picture quality objective evaluation method based on machine learning | |
| CN104954778A (en) | Objective stereo image quality assessment method based on perception feature set | |
| CN106023152B (en) | It is a kind of without with reference to objective evaluation method for quality of stereo images | |
| CN105488792A (en) | No-reference stereo image quality evaluation method based on dictionary learning and machine learning | |
| CN105069794A (en) | Binocular rivalry based totally blind stereo image quality evaluation method | |
| CN105898279B (en) | A kind of objective evaluation method for quality of stereo images | |
| CN108848365B (en) | A kind of reorientation stereo image quality evaluation method | |
| CN103914835A (en) | Non-reference quality evaluation method for fuzzy distortion three-dimensional images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210716 Address after: No.1063, building 13, industrial zone, Wuhan, Hubei 430000 Patentee after: Wuhan Tuozhijia Information Technology Co.,Ltd. Address before: 310023 No. 318 stay Road, Xihu District, Zhejiang, Hangzhou Patentee before: ZHEJIANG University OF SCIENCE AND TECHNOLOGY |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20211125 Address after: 314500 02, No. 4, South Zaoqiang street, No. 1, Nanmen Gongnong Road, Chongfu Town, Tongxiang City, Jiaxing City, Zhejiang Province Patentee after: Jiaxing Zhixu Information Technology Co.,Ltd. Address before: No.1063, building 13, industrial zone, Wuhan, Hubei 430000 Patentee before: Wuhan Tuozhijia Information Technology Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240201 Address after: Room E1403, Building 1, No. 1378 Wenyi West Road, Cangqian Street, Yuhang District, Hangzhou City, Zhejiang Province, 310000 Patentee after: Hangzhou Shijia Culture Media Co.,Ltd. Country or region after: China Address before: 314500 02, No. 4, South Zaoqiang street, No. 1, Nanmen Gongnong Road, Chongfu Town, Tongxiang City, Jiaxing City, Zhejiang Province Patentee before: Jiaxing Zhixu Information Technology Co.,Ltd. Country or region before: China |