CN113643262A - No-reference panoramic image quality evaluation method, system, equipment and medium - Google Patents
No-reference panoramic image quality evaluation method, system, equipment and medium Download PDFInfo
- Publication number
- CN113643262A CN113643262A CN202110946915.1A CN202110946915A CN113643262A CN 113643262 A CN113643262 A CN 113643262A CN 202110946915 A CN202110946915 A CN 202110946915A CN 113643262 A CN113643262 A CN 113643262A
- Authority
- CN
- China
- Prior art keywords
- panoramic image
- characteristic
- gaussian distribution
- hsv
- image
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30168—Image quality inspection
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method, a system, equipment and a medium for evaluating the quality of a non-reference panoramic image, wherein the method comprises the following steps: down-sampling the panoramic image to generate images with different scales; calculating HSV (hue, saturation and value) characteristics of the images of different scales; calculating the MSCN coefficient of the panoramic image, fitting the MSCN coefficient into generalized Gaussian distribution and asymmetric generalized Gaussian distribution, and extracting BRI SQUE characteristics from the generalized Gaussian distribution and the asymmetric generalized Gaussian distribution; and combining the HSV characteristic and the BRI SQUE characteristic of the obtained images with different scales, and inputting the combined HSV characteristic and BRI SQUE characteristic as an integral characteristic into a regression model for training and predicting to obtain the final quality score of the panoramic image. The invention combines the statistical characteristic of the panoramic image and the observation characteristic of the human visual system, and has better accuracy compared with other non-reference image quality evaluation.
Description
Technical Field
The invention relates to the technical field of panoramic image quality evaluation, in particular to a method, a system, equipment and a medium for evaluating the quality of a non-reference panoramic image.
Background
Panoramic images, also known as 360 °, spherical or omnidirectional images, are a new type of multimedia that can bring immersive experiences to the audience. In recent years, with the rapid development of Virtual Reality (VR) technology, panoramic images/videos have gradually come into our daily lives as one of the main forms of VR contents, and have attracted great attention. Unlike conventional two-dimensional (2D) images, which can only cover a limited plane, the content of the panoramic image can cover the entire 360 ° × 180 ° viewing range, i.e., the panoramic image can seamlessly surround the viewer and occupy the entire field of view of the viewer. Also on the viewing mechanism, the viewer needs to wear the head mounted display to freely focus the viewport in the direction of the line of sight, which is very similar to the way that human beings capture image content in different directions by head movement in the real world, and thus immersive and even interactive experiences are achieved.
But at the same time, new challenges are also presented to the quality of the panoramic image in order not to affect the viewer experience. Only the high-quality panoramic image can deeply restore the scene information, so that the scene information is falsified and truthful, and the real scene feeling is brought to people. However, in the processes of acquisition and splicing, projection transformation, coding compression, transmission, decoding, back projection, playing and displaying of panoramic images, the problem of image distortion is brought to the process which is difficult to avoid. A reasonable quality evaluation system can accurately reflect the distortion degree of the panoramic image, so that key factors influencing the quality of the panoramic image are found out, and guidance is provided for improving the quality of the panoramic image. Particularly in the aspect of compression of panoramic images, quality evaluation is needed to evaluate the quality reduction caused by compression, so that improvement of a compression algorithm is guided.
The panoramic image evaluation is classified into subjective quality evaluation and objective quality evaluation, as in the planar image evaluation. The subjective quality evaluation needs the participation of people, and the audience gives subjective scores according to the scoring standard after watching the panoramic image, and the method needs a large amount of manpower and time, so the method is usually only used for manufacturing a database and is used for measuring the effect of the objective quality evaluation algorithm. The objective quality evaluation does not need human participation, and the quality score of the panoramic image is directly calculated through a designed quality evaluation model.
Objective algorithms can be divided into three types: full reference, half reference and no reference. The full-reference method requires evaluation by means of all information of the original image, the half-reference method requires partial information of the original image, and the no-reference method does not require original image information. In the prior art, some methods are based on PSNR (peak signal-to-noise ratio). However, the PSNR-based method does not consider visual characteristics of human eyes, and thus ssim (structural similarity) -based methods have been proposed. These methods are all full reference algorithms, and it is difficult to obtain a reference image without distortion in the actual evaluation process. Therefore, a no-reference evaluation method based on learning has been proposed thereafter. However, none of these methods takes into account the effect of image sharpness on the subjective perception of the viewer. Although the resolution used by the mainstream panoramic video standard generally exceeds 2K, the distance of the same object from the panoramic camera in the shooting scene will inevitably affect its sharpness during viewing, thus affecting the subjective perception of the observer. In addition, the performance of different playback devices will affect the clarity of the picture actually seen by the viewer.
Therefore, how to incorporate the influence of the definition on the quality score in the design of the objective quality evaluation method so that the established model can more accurately predict the image quality score is a problem which needs to be further solved at present.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a method, a system, equipment and a medium for evaluating the quality of a non-reference panoramic image, which incorporate the influence of definition on quality scores in design, and simultaneously combine the color characteristics of the panoramic image, thereby having better accuracy.
In order to solve the technical problems, the invention is realized by the following technical scheme:
in a first aspect of the present invention, a method for evaluating quality of a reference-free panoramic image is provided, which includes:
down-sampling the panoramic image to generate images with different scales;
calculating HSV (hue, saturation and value) characteristics of the images of different scales;
calculating the MSCN coefficient of the panoramic image, fitting the MSCN coefficient into generalized Gaussian distribution and asymmetric generalized Gaussian distribution, and extracting BRISQUE characteristics from the generalized Gaussian distribution and the asymmetric generalized Gaussian distribution;
and combining the HSV characteristic and the BRISQE characteristic of the obtained images with different scales, and inputting the combined HSV characteristic and BRISQE characteristic as integral characteristics into a regression model for training and prediction to obtain the final quality score of the panoramic image.
Optionally, the panoramic image is downsampled, wherein a downsampling factor is a multiple of 2.
Optionally, the calculating, for the images of different scales, a feature of each scale image HSV includes:
converting RGB into HSV flux for each scale image;
and then, average values of the H flux, the S flux and the V flux of the image are respectively calculated to obtain the color characteristic vector of the image.
Optionally, the extracting the BRISQUE feature therefrom includes:
and fitting generalized Gaussian distribution to obtain 2-dimensional characteristics, fitting non-generalized Gaussian distribution to obtain 4 x 4-dimensional characteristics, and splicing together to obtain 18-dimensional BRISQUE characteristics.
Optionally, the regression model is SVR and RBF is selected as the kernel function.
In a second aspect of the present invention, there is provided a no-reference panoramic image quality evaluation system, including:
a down-sampling module: down-sampling the original image to obtain images with different scales;
an HSV feature acquisition module: calculating the hue, saturation and brightness of the image to obtain HSV (hue, saturation and value) characteristics of the image;
a BRISQUE feature acquisition module: calculating the MSCN coefficient of the image, fitting the MSCN coefficient to Gaussian distribution to obtain BRISQUE characteristics;
model training and prediction module: and combining the HSV characteristic and the BRISQE characteristic of the images with different scales, inputting the combined HSV characteristic and BRISQE characteristic as an integral characteristic into a regression model for training and predicting to obtain the final quality score of the panoramic image.
In a third aspect of the present invention, there is provided a no-reference panoramic image quality rating apparatus, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor is configured to execute the no-reference panoramic image quality rating method when executing the program.
In a fourth aspect of the present invention, there is provided a computer-readable storage medium having stored thereon a computer program for executing the method for rating a quality of a non-reference panoramic image described above when the program is executed by a processor.
Compared with the prior art, the embodiment of the invention has at least one of the following advantages:
(1) the invention provides a method and a system for evaluating the quality of a non-reference panoramic image, which are characterized in that images with a plurality of scales are obtained through down sampling, HSV (hue, saturation, value) characteristics and BRISQE (hue, saturation, value) characteristics of the images are extracted and combined into an overall characteristic to be input into a model for training and prediction, the statistical characteristics of the panoramic image and the observation characteristics of a human visual system are combined, and the performance of the model comprises the following steps: the Spearman rank-order correlation coefficient (SROCC), the Pearson Linear Correlation Coefficient (PLCC) and the Root Mean Square Error (RMSE) are superior to other existing technologies;
(2) according to the method and the system for evaluating the quality of the non-reference panoramic image, provided by the invention, the images with multiple scales are obtained through down sampling, HSV (hue, saturation, value) characteristics and BRISQE (hue, saturation, value) characteristics of the images are extracted and combined into an overall characteristic which is input into a model for training and prediction, the calculation is simple, the characteristic dimension is low, frequency conversion and convolution operation are not involved, the speed is high, and the accuracy is high.
Drawings
Embodiments of the invention are further described below with reference to the accompanying drawings:
fig. 1 is a flowchart of a method for evaluating quality of a non-reference panoramic image according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating the relationship between MOS and prediction score for each test picture according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a no-reference panoramic image quality evaluation system according to an embodiment of the present invention.
Description of reference numerals: the system comprises a 1-down sampling module, a 2-HSV feature obtaining module, a 3-BRISQUE feature obtaining module and a 4-model training and predicting module.
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that variations and modifications can be made by persons skilled in the art without departing from the spirit of the invention. All falling within the scope of the present invention.
Fig. 1 is a flowchart of a method for evaluating quality of a non-reference panoramic image according to an embodiment of the present invention.
Referring to fig. 1, the method for evaluating quality of a non-reference panoramic image of the present embodiment includes:
s11: downsampling an original image to generate images with different scales; the down-sampling factor is set to be an integer multiple of 2;
s12: extracting Hue, Saturation and brightness of each scale image to obtain HSV (Hue: Hue; Saturation: Value: brightness) characteristics of the image;
s13: calculating MSCN (Mean filtered Contrast Normalized coefficients) of the panoramic Image, fitting the MSCN to Generalized Gaussian Distribution (GGD) and Asymmetric Generalized Gaussian Distribution (AGGD), and extracting BRISQUE (blank/referred Image Spatial Quality Evaluator) characteristics from the GGD and AGGD;
s14: and combining the HSV characteristic and the BRISQE characteristic of the images with different scales, inputting the combined HSV characteristic and BRISQE characteristic as an integral characteristic into a regression model for training and predicting to obtain the final quality score of the panoramic image.
In the embodiment, images with multiple scales are obtained through down-sampling, HSV (hue, saturation, value) features and BRISQUE (brosy) features of the images are extracted to be combined into an integral feature, the integral feature is input into a model to be trained and predicted, the statistical features of the panoramic image and the observation characteristics of a human visual system are combined, the influence of definition on the quality score is brought into the design of the objective quality evaluation method, and the established model can be used for more accurately predicting the image quality score.
In some preferred embodiments, when S12 is executed, RGB to HSV fluxes are performed on each scale image, and then average values of H flux, S flux and V flux of the image are calculated respectively to obtain a color feature vector of the image, which is denoted as FHSV_k，Wherein k represents the kth scale image, and the default original image is the 1 st scale image.
In some preferred embodiments, when S13 is executed, the extracted briske feature may specifically refer to the following operations as a spatial domain feature of the image:
for a pixel point (I, j) on the image, the pixel value is I (I, j), and the MSCN coefficient is calculated according to the following formula
In the above formula, i ∈ 1, 2., M, j ∈ 1, 2., N (M is the height of the image, N is the width of the image). μ (i, j) represents the local mean and σ (i, j) represents the local mean square error. C is a constant term, ensuring that the denominator is not 0, where C is taken to be 1. The local mean μ is the gaussian blur of the original image and the local mean square error is the gaussian blur of the square of the difference between the original image and μ. In the following equation, W is a gaussian blur window function.
μ＝W*I (2)
The generalized Gaussian distribution (AGG) density function with zero mean is as follows:
wherein the content of the first and second substances,
fitting the MSCN coefficient into AGG to obtain two parameters (alpha, sigma)2) As two characteristic quantities.
On the other hand, to describe the statistical relationship between adjacent pixels, the MSCN coefficients between adjacent pixels are calculated for four directions, horizontal (H), vertical (V), major diagonal (D1), minor diagonal (D2), resulting in 4 coefficient matrices:
wherein i belongs to 1, 2, the.
The non-generalized Gaussian distribution (AGGD) density function is as follows:
fitting the four coefficient matrixes into AGGD to obtain 4 parameters
As a characteristic feature, among others,
the 2-dimensional features are obtained by fitting (AGG), the 4 × 4-dimensional features are obtained by fitting (AGGD), and the BRISQUE features of 18 dimensions are obtained by splicing together.
S14: and combining the HSV characteristic and the BRISQE characteristic of the images with different scales, inputting the combined HSV characteristic and BRISQE characteristic as an integral characteristic into a regression model for training and predicting to obtain the final quality score of the panoramic image.
According to the embodiment of the invention, the images of multiple scales are obtained by down-sampling the panoramic image, the HSV characteristic and the BRISQUE characteristic of the images of different scales are calculated, the influence of the definition on subjective perception is considered, the statistical characteristic of the panoramic image is combined, and the accuracy is better compared with other quality evaluation without reference images.
In the preferred embodiment, the original image is downsampled in S11 for the panoramic image I with size of m × n1Selecting the sampling times to be 4, setting the down-sampling factors to be 2, 4, 8 and 16, and obtaining the size of the down-sampling of four times
Respectively, are recorded asI2、I3、I4、I5And, together with the original image, there are a total of five scales of images. Of course, in other embodiments, other sampling arrangements may be used, and are not limited to the preferred embodiment.
Based on the above preferred embodiment, in S14, the HSV feature and the BRISQUE feature of five scales are combined together, each scale having 3 dimensions for HSV feature and 18 dimensions for BRISQUE feature, and these features are pieced together to obtain an overall feature with 105 dimensions, and input into the regression model. The regression model in S14 is further svr (support vector regression), and it selects rbf (radial basis function) as the kernel function.
Score＝SVR_model{I} (15)
Experiments were performed on the OIQA dataset to evaluate the model proposed by the present invention. The experimental environment of the experiment is a win10 system and a matlab2018a experimental platform. The evaluation indexes of the experiment are SROCC (specific rank-order correlation coefficient), PLCC (Pearson linear correlation coefficient), RMSE (root mean squared error). The higher the SRCC and PLCC, the lower the RMSE, indicating better model performance. The data set has 336 panoramic images, containing 16 scenes. In the experiment, 12 scenes are randomly divided to be used as a training set, and 4 scenes are left to be used as a testing set. And repeating 1000 times of training-testing experiments, and taking the median of the obtained results as the final experiment result. Before calculating PLCC and RMSE for different objective quality assessment methods, the predicted quality scores need to be mapped into a common scale using a five-parameter logistic nonlinear fitting function, as follows:
wherein { betaiI | (1, 2.·, 5): is 5 parameters to be fitted, x represents an original score calculated by an objective quality evaluation algorithm, and g (x) represents a quality score obtained by nonlinear mapping of x.
Table 1 shows the comparison of the present invention embodiments with other full reference/no reference quality evaluation algorithms, from a single distortion type and an overall distortion type, respectively. Various indexes of the method provided by the embodiment of the invention are superior to those of the existing algorithm.
TABLE 1
As shown in fig. 2, the MOS of each test picture is related to the prediction score, and the fitting degree between the MOS and the prediction score is good.
The experiment shows that the method can well extract the statistical characteristics of the panoramic image, thereby predicting the human perception quality score.
Fig. 3 is a schematic diagram of a no-reference panoramic image quality evaluation system according to an embodiment of the present invention.
Referring to fig. 3, the no-reference panoramic image quality evaluation system of the present embodiment includes: the system comprises a down-sampling module 1, an HSV feature obtaining module 2, a BRISQUE feature obtaining module 3 and a model training and predicting module 4. The down-sampling module 1 is configured to down-sample an original image to obtain images with different scales.
The HSV feature obtaining module 2 is used for calculating hue, saturation and brightness of the image and obtaining HSV features of the image. The BRISQUE characteristic obtaining module 3 is used for calculating the MSCN coefficient of the image, fitting the MSCN coefficient to Gaussian distribution and obtaining the BRISQUE characteristic; and the model training set prediction module 4 is used for combining HSV (hue, saturation, value) characteristics and BRISQUE (brosy) characteristics of the images with different scales, inputting the combined HSV characteristics and BRISQUE characteristics as overall characteristics into the regression model for training and prediction to obtain the final quality score of the panoramic image. In the embodiment, images of multiple scales are obtained by down-sampling the panoramic image, HSV (hue, saturation, value) characteristics and BRISQUE (brosy) characteristics of the images of different scales are calculated, the influence of definition on subjective perception is considered, the statistical characteristics of the panoramic image are combined, and the performance of the system is superior to that of the existing other technologies.
In a preferred embodiment, the down-sampling factor of the down-sampling module is set to be an integer multiple of 2, for example, the down-sampling factors are set to be 2, 4, 8, and 16, including the original image, to obtain five images with different scales.
In a preferred embodiment, the regression model in the model training set prediction module may be an SVR and it selects an RBF as the kernel function.
Based on the above embodiments, in another embodiment, the present invention further provides a no-reference panoramic image quality rating apparatus, including a memory, a processor, and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the program to perform the no-reference panoramic image quality rating method in any one of the above embodiments.
Based on the above-mentioned embodiments, in another embodiment, the present invention further provides a computer-readable storage medium having stored thereon a computer program, which when executed by a processor, is configured to perform the no-reference panoramic image quality rating method in any of the above-mentioned embodiments.
The non-reference panoramic image quality evaluation system provided by the embodiment of the invention obtains images of multiple scales through down sampling, extracts HSV (hue, saturation, value) characteristics and BRISQE (brosy) characteristics of the images, combines the HSV characteristics and the BRISQE characteristics into an overall characteristic, inputs the overall characteristic into a model for training and prediction, is simple in calculation, has low characteristic dimensionality, and does not relate to frequency conversion and convolution operation.
It should be noted that, the steps in the method provided by the present invention may be implemented by using corresponding modules, devices, units, and the like in the system, and those skilled in the art may implement the composition of the system by referring to the technical solution of the method, that is, the embodiment in the method may be understood as a preferred example for constructing the system, and will not be described herein again.
Those skilled in the art will appreciate that, in addition to implementing the system and its various devices provided by the present invention in purely computer readable program code means, the method steps can be fully programmed to implement the same functions by implementing the system and its various devices in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Therefore, the system and various devices thereof provided by the present invention can be regarded as a hardware component, and the devices included in the system and various devices thereof for realizing various functions can also be regarded as structures in the hardware component; means for performing the functions may also be regarded as structures within both software modules and hardware components for performing the methods.
The foregoing disclosure of the preferred embodiments of the present invention has been made in an effort to provide a better understanding of the principles of the invention and its practical application, and is not intended to limit the invention to the specific embodiments disclosed herein. Any modifications and variations within the scope of the description, which may occur to those skilled in the art, are intended to be within the scope of the invention.
Claims (10)
1. A no-reference panoramic image quality evaluation method is characterized by comprising the following steps:
down-sampling the panoramic image to generate images with different scales;
calculating HSV (hue, saturation and value) characteristics of the images of different scales;
calculating the MSCN coefficient of the panoramic image, fitting the MSCN coefficient into generalized Gaussian distribution and asymmetric generalized Gaussian distribution, and extracting BRISQUE characteristics from the generalized Gaussian distribution and the asymmetric generalized Gaussian distribution;
and combining the HSV characteristic and the BRISQE characteristic of the obtained images with different scales, and inputting the combined HSV characteristic and BRISQE characteristic as integral characteristics into a regression model for training and prediction to obtain the final quality score of the panoramic image.
2. The method of claim 1, wherein the panoramic image is downsampled by a factor of 2.
3. The method for evaluating the quality of the non-reference panoramic image according to claim 1, wherein the step of calculating HSV characteristics of the images of different scales comprises the following steps:
converting RGB into HSV flux for each scale image;
and then, average values of the H flux, the S flux and the V flux of the image are respectively calculated to obtain the color characteristic vector of the image.
4. The method for evaluating the quality of the non-reference panoramic image according to claim 1, wherein the fitting of the MSCN coefficients into a generalized gaussian distribution specifically comprises:
wherein
Two parameters alpha, sigma obtained by fitting2As two characteristic quantities;
in the above formula: equation (4) is a generalized Gaussian distribution density function with zero mean, alpha, sigma2These two parameters may be calculated with a time-based matching algorithm.
5. The method of claim 1, wherein fitting the MSCN coefficients to an asymmetric generalized Gaussian distribution comprises:
the MSCN coefficients between adjacent pixels are calculated to describe the statistical relationship between adjacent pixels, for four directions: obtaining four coefficient matrixes by using a horizontal H, a vertical V, a main diagonal D1 and a secondary diagonal D2;
fitting the four coefficient matrixes into a non-generalized Gaussian distribution:
the 4 parameters η, v,
as a characteristic feature, among others,
in the above formula: equation (10) is a non-generalized gaussian distribution density function, v,
these three parameters may be calculated by a time-based matching algorithm.
6. The no-reference panoramic image quality evaluation method according to claim 5, wherein the extracting BRISQUE features therefrom includes:
and fitting generalized Gaussian distribution to obtain 2-dimensional characteristics, fitting non-generalized Gaussian distribution to obtain 4 x 4-dimensional characteristics, and splicing together to obtain 18-dimensional BRISQUE characteristics.
7. The no-reference panoramic image quality evaluation method according to any one of claims 1 to 6, characterized in that the regression model is SVR and RBF is selected as a kernel function.
8. A no-reference panoramic image quality rating system, comprising:
a down-sampling module: down-sampling the original image to obtain images with different scales;
an HSV feature acquisition module: calculating the hue, saturation and brightness of the image to obtain HSV (hue, saturation and value) characteristics of the image;
a BRISQUE feature acquisition module: calculating the MSCN coefficient of the image, fitting the MSCN coefficient to Gaussian distribution to obtain BRISQUE characteristics;
model training and prediction module: and combining the HSV characteristic and the BRISQE characteristic of the images with different scales, inputting the combined HSV characteristic and BRISQE characteristic as an integral characteristic into a regression model for training and predicting to obtain the final quality score of the panoramic image.
9. A non-reference panoramic image quality rating apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor is configured to execute the non-reference panoramic image quality rating method of any one of claims 1 to 7 when executing the program.
10. A computer-readable storage medium having stored thereon a computer program, characterized in that the program, when being executed by a processor, is adapted to carry out the method of quality evaluation of a non-reference panoramic image of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110946915.1A CN113643262A (en) | 2021-08-18 | 2021-08-18 | No-reference panoramic image quality evaluation method, system, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110946915.1A CN113643262A (en) | 2021-08-18 | 2021-08-18 | No-reference panoramic image quality evaluation method, system, equipment and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113643262A true CN113643262A (en) | 2021-11-12 |
Family
ID=78422651
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110946915.1A Pending CN113643262A (en) | 2021-08-18 | 2021-08-18 | No-reference panoramic image quality evaluation method, system, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113643262A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105654433A (en) * | 2015-12-28 | 2016-06-08 | 桂林电子科技大学 | Color image enhancement method based on improved multi-scale Retinex |
| CN108765414A (en) * | 2018-06-14 | 2018-11-06 | 上海大学 | Based on wavelet decomposition and natural scene count without referring to stereo image quality evaluation method |
| CN108961206A (en) * | 2018-04-20 | 2018-12-07 | 北京航空航天大学 | A kind of defog effect without reference method for objectively evaluating |
| CN110853027A (en) * | 2019-11-18 | 2020-02-28 | 方玉明 | Three-dimensional synthetic image no-reference quality evaluation method based on local variation and global variation |
| CN111310774A (en) * | 2020-04-01 | 2020-06-19 | 江苏商贸职业学院 | PM2.5 concentration measurement method based on image quality |
| CN112950629A (en) * | 2021-04-02 | 2021-06-11 | 上海大学 | No-reference panoramic image quality evaluation method and system |
-
2021
- 2021-08-18 CN CN202110946915.1A patent/CN113643262A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105654433A (en) * | 2015-12-28 | 2016-06-08 | 桂林电子科技大学 | Color image enhancement method based on improved multi-scale Retinex |
| CN108961206A (en) * | 2018-04-20 | 2018-12-07 | 北京航空航天大学 | A kind of defog effect without reference method for objectively evaluating |
| CN108765414A (en) * | 2018-06-14 | 2018-11-06 | 上海大学 | Based on wavelet decomposition and natural scene count without referring to stereo image quality evaluation method |
| CN110853027A (en) * | 2019-11-18 | 2020-02-28 | 方玉明 | Three-dimensional synthetic image no-reference quality evaluation method based on local variation and global variation |
| CN111310774A (en) * | 2020-04-01 | 2020-06-19 | 江苏商贸职业学院 | PM2.5 concentration measurement method based on image quality |
| CN112950629A (en) * | 2021-04-02 | 2021-06-11 | 上海大学 | No-reference panoramic image quality evaluation method and system |
Non-Patent Citations (2)
| Title |
|---|
| 吴军: "基于视觉感知特性的图像质量评价方法研究", 《中国博士学位论文全文数据库 信息科技辑》, no. 04, pages 55 - 72 * |
| 杨蓝平: "基于真实失真图像无参考质量评价的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》, no. 03, pages 29 - 42 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Gu et al. | Hybrid no-reference quality metric for singly and multiply distorted images | |
| Ciancio et al. | No-reference blur assessment of digital pictures based on multifeature classifiers | |
| CN104079925B (en) | Ultra high-definition video image quality method for objectively evaluating based on vision perception characteristic | |
| Sun et al. | CVIQD: Subjective quality evaluation of compressed virtual reality images | |
| CN110751649B (en) | Video quality evaluation method and device, electronic equipment and storage medium | |
| Tian et al. | A multi-order derivative feature-based quality assessment model for light field image | |
| Yang et al. | Subjective quality assessment of screen content images | |
| Yang et al. | An objective assessment method based on multi-level factors for panoramic videos | |
| Li et al. | Subjective and objective quality assessment of compressed screen content videos | |
| Lu et al. | Deep neural network for blind visual quality assessment of 4K content | |
| CN112950596B (en) | Tone mapping omnidirectional image quality evaluation method based on multiple areas and multiple levels | |
| Ghadiyaram et al. | A no-reference video quality predictor for compression and scaling artifacts | |
| Men et al. | Empirical evaluation of no-reference VQA methods on a natural video quality database | |
| CN113298779B (en) | Video redirection quality objective evaluation method based on reverse reconstruction grid | |
| Liu et al. | Spatio-temporal interactive laws feature correlation method to video quality assessment | |
| Da et al. | Perceptual quality assessment of nighttime video | |
| Xian et al. | A content-oriented no-reference perceptual video quality assessment method for computer graphics animation videos | |
| CN113128517A (en) | Tone mapping image mixed visual feature extraction model establishment and quality evaluation method | |
| Pan et al. | No-reference video quality assessment based on modeling temporal-memory effects | |
| CN113643262A (en) | No-reference panoramic image quality evaluation method, system, equipment and medium | |
| Zhang et al. | Multi-layer and Multi-scale feature aggregation for DIBR-Synthesized image quality assessment | |
| Huang et al. | Sb-vqa: A stack-based video quality assessment framework for video enhancement | |
| Suen et al. | Spatial-temporal visual attention model for video quality assessment | |
| CN112950629A (en) | No-reference panoramic image quality evaluation method and system | |
| CN110276744B (en) | Image splicing quality evaluation method and device |
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 |