CN109801257A - No reference DIBR generates image quality evaluating method - Google Patents
No reference DIBR generates image quality evaluating method Download PDFInfo
- Publication number
- CN109801257A CN109801257A CN201811544691.6A CN201811544691A CN109801257A CN 109801257 A CN109801257 A CN 109801257A CN 201811544691 A CN201811544691 A CN 201811544691A CN 109801257 A CN109801257 A CN 109801257A
- Authority
- CN
- China
- Prior art keywords
- image
- lbp
- value
- quality
- region
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 14
- 238000013442 quality metrics Methods 0.000 claims abstract description 11
- 238000005259 measurement Methods 0.000 claims abstract description 9
- 238000001514 detection method Methods 0.000 claims abstract description 8
- 238000005070 sampling Methods 0.000 claims description 4
- 238000013441 quality evaluation Methods 0.000 description 4
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000009877 rendering Methods 0.000 description 2
- 239000004576 sand Substances 0.000 description 2
- 238000003786 synthesis reaction Methods 0.000 description 2
- 238000003646 Spearman's rank correlation coefficient Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to a kind of no reference DIBR to generate image quality evaluating method, the following steps are included: 1) local cavity region detection and quality metric: given image I, calculate its local binary patterns figure LBP, and LBP is schemed to carry out binaryzation, defining the region that LBP value is 8 is hole area, using non-hole region and the accounting of general image area as the quality metric Q of hole region1；2) local elongation region detection and quality metric: to LBP figure again binaryzation, the pixel value that regulation LBP value is 8 is 1, remaining corresponding pixel value of LBP value is 0, phase adduction is carried out to the pixel value of each column of whole image and takes mean value, regulation mean value is classified as stretch zones greater than 0.2, and the gradient similitude for defining the region of stretch zones homalographic adjacent thereto is the measurement Q of stretch zones intensity2；3) global Fuzzy Quality measurement；4) total quality is assessed.
Description
Technical field
The invention belongs to field of image processings, and the reference-free quality evaluation side of image is generated more particularly, to a kind of DIBR
Method.
Background technique
In recent years, stereo-picture and video gradually come into daily life with multiple display modes, as glasses are seen
See formula, naked eye viewing formula etc..Wherein, the Auto-stereo display based on naked eye viewing needs to capture Same Scene from multiple angles
Multiple images, this has initiated huge challenge to storing and transmitting for data.In order to solve this problem, multiple views and depth map
(MVD, Multi-view-Video-plus-Depth) format is adopted and is applied to stereo data storage.Using MVD format,
In transmitting terminal only need that limited texture view and its relevant depth map are encoded and transmitted.In receiving end, by connecing
Rendering (DIBR, the Depth-Image-Based-Rendering) algorithm based on depth image of device side is received from decoded texture
View and depth map synthesize remaining virtual view.Unfortunately, DIBR is not a kind of perfect technology, can be produced to synthesis view
Raw a variety of distortions.Low quality DIBR synthesis view may cause irritating sensory experience, but conventional images quality evaluation solution
Certainly scheme is helpless in terms of accurately estimation DIBR generates distortion.The present invention analyzes the distortion feature that DIBR generates figure,
Measurement based on localized distortion and global distortion, proposes a kind of non-reference picture quality appraisement method.
Summary of the invention
The present invention generates the distortion feature of image for DIBR, proposes that a kind of no reference DIBR generates image quality evaluation side
Method, this method and subjective assessment score have higher consistency.Technical solution is as follows:
A kind of no reference DIBR generation image quality evaluating method, comprising the following steps:
1) local cavity region detection and quality metric
A given DIBR generates image I, calculates its local binary patterns figure LBP, and schemes to carry out binaryzation to LBP, fixed
The region that adopted LBP value is 8 is hole area, using non-hole region and the accounting of general image area as the quality degree of hole region
Measure Q1；
2) local elongation region detection and quality metric
To LBP figure again binaryzation, it is specified that the pixel value that LBP value is 8 is 1, remaining corresponding pixel value of LBP value is 0, right
The pixel value of each column of whole image carries out phase adduction and takes mean value, it is specified that mean value is classified as stretch zones greater than 0.2, and definition stretches
The gradient similitude in the region of region homalographic adjacent thereto is the measurement Q of stretch zones intensity2；
3) global Fuzzy Quality measurement
DIBR generation image I is divided into n image blocks of a size, calculates the variance of each image block；To DIBR
Generation image I carries out the down-sampling with 2 for scale, is equally divided into the image block of the sizes such as n, calculates the side of each image block
Difference, the overall situation for defining image obscure as the mean value Q of original image and down-sampled images block variance difference3；
4) total quality is assessed
The total quality Q of image is defined as: Q=0.9787Q1+0.0143Q2+0.007Q3。
The present invention proposes that a kind of DIBR generates the quality evaluating method of image, without necessarily referring to the intervention of image, Ke Yiyou
Evaluation DIBR in effect ground generates picture quality.
Detailed description of the invention
Fig. 1 algorithm frame
Specific embodiment
The present invention proposes that a kind of no reference DIBR generates image quality evaluating method, and frame is as shown in Figure 1.
(1) local cavity region detection and quality metric
An image I is given, calculates its LBP figure first
Wherein, niIndicate center pixel ncCircular periphery ith pixel, I (ni) and I (nc) respectively indicate its corresponding picture
Element value.P indicates the neighboring pixel number to be considered, P=8 is arranged among the present invention.Scheme when being calculated as LBPAfterwards, to its into
Row binarization operation:
Wherein, DrIndicate the binary picture being calculated.Dr=0 region indicates that hole region, the present invention define hole area
The quality metric Q in domain1Are as follows:
Wherein, K0The number of pixels of the number of pixels and whole image in hole region is respectively indicated with K.
(2) local elongation region detection and quality metric
For the LBP figure that step (1) acquires, the present invention carries out binarization operation to it again:
Wherein, DsFor binarization operation result.For DsEach column of figure carry out phase adduction and take mean value, define mean value and are greater than
0.2 is listed in stretch zones.Since stretch zones tend to occur at image left end or right end, present invention provide that, only
There are the column for being in the same join domain with image left end or right end to be just defined as stretching column.Finally, all stretchings
Column collectively constitute stretch zones.It is influenced caused by picture quality for measurement tensile strength, the present invention calculates stretch zones and its
The gradient similitude S in the region of the sizes such as adjacent areag:
Wherein, GsAnd GnRespectively indicate stretch zones and its etc. sizes adjacent area gradient map.In the present invention, gradient map
There is following formula to be calculated:
Wherein pxAnd pyIt is Prewitt operator in the horizontal and vertical directions respectively.It is noted that calculating GsAnd Gn
When, the I among above-mentioned formula is substituted for the gray value of corresponding region respectively.Finally, the distortion matter in image stretch region
Measure Q2Is defined as:
Wherein,Indicate the mean value of stretch zones inside gradient similarity, SjIndicate j-th of SgElement, J indicate stretch zones
The number of interior pixel.
(3) global Fuzzy Quality measurement
For image I, it is divided into the non-coincidence image block of 32 × 32 sizes first, it is assumed here that finally divide
Image block number is N, calculates its variances sigma to each image block0,n,1≤n≤N；Down-sampling is carried out for scale with 2 to image I, and
To the obtained image of sampling equally to carry out piecemeal (size of block is 16 × 16) in non-coincidence mode, and calculate each piece
Variances sigma1,n,1≤n≤N.The overall situation of image is fuzzy can to measure as Q3:
(4) total quality is assessed
The total quality Q of image is defined as:
Q=0.9787Q1+0.0143Q2+0.007Q3\*MERGEFORMAT(13)
For verification algorithm validity, the present invention selects IRCCyN/IVC database to verify.The database includes 120
A different DIBR generates image (resolution ratio is 1024 × 768).
For the performance of checking image quality evaluation algorithm, using Spearman rank correlation coefficient (Spearman Rank-
Order Correlation Coefficient, SRCC) it is used as assessment level.SRCC shows proposition of the present invention closer to 1
Algorithm there is better performance, and subjective marking consistency is higher.
The experimental results showed that the present invention can obtain SRCC=0.652's as a result, illustrating the evaluating objective quality of this method
Correlation between predicted value and subjective scoring is high, shows that the method for the present invention and human visual system have preferable consistency.
Claims (1)
1. a kind of no reference DIBR generates image quality evaluating method, comprising the following steps:
1) local cavity region detection and quality metric
A given DIBR generates image I, calculates its local binary patterns figure LBP, and schemes to carry out binaryzation to LBP, defines LBP
The region that value is 8 is hole area, using non-hole region and the accounting of general image area as the quality metric Q of hole region1。
2) local elongation region detection and quality metric
To LBP figure again binaryzation, it is specified that the pixel value that LBP value is 8 is 1, remaining corresponding pixel value of LBP value is 0, to entire
The pixel value of each column of image carries out phase adduction and takes mean value, it is specified that mean value is classified as stretch zones, definition stretch zones greater than 0.2
The gradient similitude in the region of homalographic adjacent thereto is the measurement Q of stretch zones intensity2；
3) global Fuzzy Quality measurement
DIBR generation image I is divided into n image blocks of a size, calculates the variance of each image block；DIBR is generated
Image I carries out the down-sampling with 2 for scale, is equally divided into the image block of the sizes such as n, calculates the variance of each image block,
The overall situation for defining image obscures as the mean value Q of original image and down-sampled images block variance difference3；
4) total quality is assessed
The total quality Q of image is defined as: Q=0.9787Q1+0.0143Q2+0.007Q3。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811544691.6A CN109801257A (en) | 2018-12-17 | 2018-12-17 | No reference DIBR generates image quality evaluating method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811544691.6A CN109801257A (en) | 2018-12-17 | 2018-12-17 | No reference DIBR generates image quality evaluating method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109801257A true CN109801257A (en) | 2019-05-24 |
Family
ID=66556934
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811544691.6A Pending CN109801257A (en) | 2018-12-17 | 2018-12-17 | No reference DIBR generates image quality evaluating method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109801257A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110415223A (en) * | 2019-07-17 | 2019-11-05 | 西安邮电大学 | A kind of the stitching image quality evaluating method and system of no reference |
| CN110636282A (en) * | 2019-09-24 | 2019-12-31 | 宁波大学 | No-reference asymmetric virtual viewpoint three-dimensional video quality evaluation method |
| CN111507933A (en) * | 2019-12-16 | 2020-08-07 | 曲阜师范大学 | DIBR synthetic image quality evaluation method based on cavity and contour amplification |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106600597A (en) * | 2016-12-22 | 2017-04-26 | 华中科技大学 | Non-reference color image quality evaluation method based on local binary pattern |
| CN107767363A (en) * | 2017-09-05 | 2018-03-06 | 天津大学 | It is a kind of based on natural scene without refer to high-dynamics image quality evaluation algorithm |
| CN108257125A (en) * | 2018-01-24 | 2018-07-06 | 中国矿业大学 | A kind of depth image quality based on natural scene statistics is without with reference to evaluation method |
-
2018
- 2018-12-17 CN CN201811544691.6A patent/CN109801257A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106600597A (en) * | 2016-12-22 | 2017-04-26 | 华中科技大学 | Non-reference color image quality evaluation method based on local binary pattern |
| CN107767363A (en) * | 2017-09-05 | 2018-03-06 | 天津大学 | It is a kind of based on natural scene without refer to high-dynamics image quality evaluation algorithm |
| CN108257125A (en) * | 2018-01-24 | 2018-07-06 | 中国矿业大学 | A kind of depth image quality based on natural scene statistics is without with reference to evaluation method |
Non-Patent Citations (1)
| Title |
|---|
| GUANGHUI YUE: "Combining Local and Global Measures for DIBR-Synthesized Image Quality Evaluation", 《IEEE》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110415223A (en) * | 2019-07-17 | 2019-11-05 | 西安邮电大学 | A kind of the stitching image quality evaluating method and system of no reference |
| CN110415223B (en) * | 2019-07-17 | 2021-10-26 | 西安邮电大学 | No-reference spliced image quality evaluation method and system |
| CN110636282A (en) * | 2019-09-24 | 2019-12-31 | 宁波大学 | No-reference asymmetric virtual viewpoint three-dimensional video quality evaluation method |
| CN111507933A (en) * | 2019-12-16 | 2020-08-07 | 曲阜师范大学 | DIBR synthetic image quality evaluation method based on cavity and contour amplification |
| CN111507933B (en) * | 2019-12-16 | 2023-08-01 | 曲阜师范大学 | DIBR synthetic image quality evaluation method based on cavity and contour amplification |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sandić-Stanković et al. | Multi–scale synthesized view assessment based on morphological pyramids | |
| Narvekar et al. | A no-reference image blur metric based on the cumulative probability of blur detection (CPBD) | |
| US7418131B2 (en) | Image-capturing device and method for removing strangers from an image | |
| Jacobson et al. | A novel approach to FRUC using discriminant saliency and frame segmentation | |
| CN101873509B (en) | Method for eliminating background and edge shake of depth map sequence | |
| CN109801257A (en) | No reference DIBR generates image quality evaluating method | |
| US8565513B2 (en) | Image processing method for providing depth information and image processing system using the same | |
| CN109345525B (en) | Ghost-removed high-dynamic-range image quality evaluation method | |
| Zhang et al. | Sparse representation-based video quality assessment for synthesized 3D videos | |
| CN112950596B (en) | Tone mapping omnidirectional image quality evaluation method based on multiple areas and multiple levels | |
| CN109510981B (en) | Stereo image comfort degree prediction method based on multi-scale DCT | |
| CN109345502A (en) | A kind of stereo image quality evaluation method based on disparity map stereochemical structure information extraction | |
| Richter et al. | Robust super-resolution for mixed-resolution multiview image plus depth data | |
| Ekmekcioglu et al. | Depth based perceptual quality assessment for synthesised camera viewpoints | |
| JP2010152521A (en) | Apparatus and method for performing stereographic processing to image | |
| CN108447059A (en) | It is a kind of to refer to light field image quality evaluating method entirely | |
| CN114648482A (en) | Quality evaluation method and system for three-dimensional panoramic image | |
| Wen et al. | Visual quality assessment for perceptually encrypted light field images | |
| Fang et al. | Perceptual quality assessment for asymmetrically distorted stereoscopic video by temporal binocular rivalry | |
| Fang et al. | Blind 3D-synthesized image quality measurement by analysis of local and global statistical properties | |
| JP2015224909A (en) | Three-dimensional measurement device and three-dimensional measurement method | |
| Kim et al. | Measurement of critical temporal inconsistency for quality assessment of synthesized video | |
| US8077999B2 (en) | Image processing apparatus and method for reducing blocking effect and Gibbs effect | |
| Seitner et al. | Trifocal system for high-quality inter-camera mapping and virtual view synthesis | |
| Oelbaum et al. | Building a reduced reference video quality metric with very low overhead using multivariate data analysis |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190524 |