CN109522960A - Image evaluation method, device, electronic equipment and computer-readable medium - Google Patents
Image evaluation method, device, electronic equipment and computer-readable medium Download PDFInfo
- Publication number
- CN109522960A CN109522960A CN201811393400.8A CN201811393400A CN109522960A CN 109522960 A CN109522960 A CN 109522960A CN 201811393400 A CN201811393400 A CN 201811393400A CN 109522960 A CN109522960 A CN 109522960A
- Authority
- CN
- China
- Prior art keywords
- image
- characteristic pattern
- laplacian
- gradient features
- original 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
Abstract
This disclosure relates to a kind of image evaluation method, device, electronic equipment and computer-readable medium.This method comprises: pre-processing to original image, standard picture is generated；Obtain the Gradient Features figure and Laplacian characteristic pattern of the standard picture；Obtain the feature vector of the Gradient Features figure Yu the Laplacian characteristic pattern；And the scoring for the original image being obtained in described eigenvector input picture Rating Model.This disclosure relates to image evaluation method, device, electronic equipment and computer-readable medium, fast and accurately picture quality can be assessed.
Description
Technical field
This disclosure relates to computer information processing field, in particular to a kind of image evaluation method, device, electronics
Equipment and computer-readable medium.
Background technique
With the arrival of information age, digital picture is able to satisfy growing show as a kind of important information carrier
Generationization business demand.The development of computer and networks extends image application field, image acquisition, compression, processing, transmission,
Different degrees of and type distortion can be brought during storage etc., directly affect the acquisition of information.Therefore, it image procossing and is answering
Effective image quality evaluation system is established with field to be of great significance.
Image quality evaluating method has two major classes at present.(1) subjective evaluation method.By manually commenting picture quality
Point, people is that end user's subjective quality assessment of image is the most accurate, reliable image quality evaluating method.But due to
Its time-consuming, valuableness is not properly suited for big data era to the processing requirement of data.(2) method for objectively evaluating.With simple, real
When, it is repeatable and easy of integration the advantages that, recent decades development quickly, becomes the research hotspot in image quality evaluation system.Benefit
Picture quality is measured with mathematics and engineering method, compensates for the deficiency of subjective evaluation method.Due to people be image most
Whole receptor, objectively evaluate be with the consistency of subjective evaluation result method for objectively evaluating quality important measurement index.In conjunction with figure
As the method for own characteristic and the physiology of human visual system and psychological characteristic becomes the hot spot studied now.
Therefore, it is necessary to a kind of method, apparatus that can be accurately assessed to image, electronic equipment and computer-readable
Medium.
Above- mentioned information are only used for reinforcing the understanding to the background of the disclosure, therefore it disclosed in the background technology part
It may include the information not constituted to the prior art known to persons of ordinary skill in the art.
Summary of the invention
In view of this, the disclosure provides a kind of image evaluation method, device, electronic equipment and computer-readable medium, energy
It is enough that fast and accurately picture quality is assessed.
Other characteristics and advantages of the disclosure will be apparent from by the following detailed description, or partially by the disclosure
Practice and acquistion.
According to the one side of the disclosure, a kind of image evaluation method is proposed, this method comprises: being located in advance to original image
Reason generates standard picture；Obtain the Gradient Features figure and Laplacian characteristic pattern of the standard picture；Obtain the gradient
The feature vector of characteristic pattern and the Laplacian characteristic pattern；And it will be in described eigenvector input picture Rating Model
To obtain the scoring of the original image.
In a kind of exemplary embodiment of the disclosure, further includes: carried out by training image to supporting vector machine model
Training, to obtain described image Rating Model.
In a kind of exemplary embodiment of the disclosure, supporting vector machine model is trained by training image, with
Obtain described image Rating Model comprise determining that training image the Gradient Features figure and the Laplacian feature
Figure；The feature vector of the training image is determined according to the characteristic pattern；Determine the scoring of the training image；And according to institute
The feature vector of the scoring and the training image of stating training image is trained supporting vector machine model, to obtain the figure
As Rating Model.
In a kind of exemplary embodiment of the disclosure, according to the scoring of the training image and the spy of the training image
Sign vector is trained supporting vector machine model, includes: to input the training image to obtain described image Rating Model
In supporting vector machine model；The parameter in the supporting vector machine model is adjusted by the scoring of the training image；
And the parameter in the supporting vector machine model generates described image Rating Model when meeting threshold value.
In a kind of exemplary embodiment of the disclosure, original image is pre-processed, generating standard picture includes: pair
The original image is adjusted resolution processes；Original image after adjustment resolution processes is standardized；
Original image after standardization is subjected to color space conversion process, generates the standard picture.
In a kind of exemplary embodiment of the disclosure, the original image after standardization is subjected to color space and is turned
Changing processing includes: to be transformed into the original image after standardization in YcbCr color space by RGB color.
In a kind of exemplary embodiment of the disclosure, the Gradient Features figure and Gauss La Pula of the standard picture are obtained
This characteristic pattern includes: that the standard picture is carried out gradient transformation to generate the Gradient Features figure；And by the Gradient Features
Figure carries out Laplacian and generates the Laplacian characteristic pattern.
In a kind of exemplary embodiment of the disclosure, the Gradient Features figure and the Laplacian feature are obtained
The feature vector of figure includes: to determine the Gradient Features figure and the Laplacian characteristic pattern respectively by gray scale setting
Gray scale；The Feature Mapping figure of the Gradient Features figure Yu the Laplacian characteristic pattern is obtained respectively；And according to feature
Mapping graph determines the feature vector of the Gradient Features figure Yu the Laplacian characteristic pattern.
In a kind of exemplary embodiment of the disclosure, the Gradient Features figure and the height are determined according to Feature Mapping figure
The feature vector of this Laplce's characteristic pattern comprises determining that feature vector coefficient；And according to described eigenvector coefficient and spy
Sign mapping graph determines the feature vector of the Gradient Features figure Yu the Laplacian characteristic pattern.
According to the one side of the disclosure, propose that a kind of image evaluation device, the device include: original image processing module,
For pre-processing to original image, standard picture is generated；Standard picture processing module, for obtaining the standard picture
Gradient Features figure and Laplacian characteristic pattern；Feature vector module, for obtaining the Gradient Features figure and the Gauss
The feature vector of Laplce's characteristic pattern；And grading module, for by described eigenvector input picture Rating Model with
Obtain the quality score of the original image.
In a kind of exemplary embodiment of the disclosure, further includes: model training module, for passing through training image to branch
It holds vector machine model to be trained, to obtain described image Rating Model.
According to the one side of the disclosure, a kind of electronic equipment is proposed, which includes: one or more processors；
Storage device, for storing one or more programs；When one or more programs are executed by one or more processors, so that one
A or multiple processors realize such as methodology above.
According to the one side of the disclosure, it proposes a kind of computer-readable medium, is stored thereon with computer program, the program
Method as mentioned in the above is realized when being executed by processor.
According to the image evaluation method of the disclosure, device, electronic equipment and computer-readable medium, original is obtained by calculating
The Gradient Features figure and Laplacian characteristic pattern of beginning image；And then according to Gradient Features figure and Laplacian characteristic pattern
The mode of the scoring of the original image is determined with input picture Rating Model, and fast and accurately picture quality can be commented
Estimate.
It should be understood that the above general description and the following detailed description are merely exemplary, this can not be limited
It is open.
Detailed description of the invention
Its example embodiment is described in detail by referring to accompanying drawing, above and other target, feature and the advantage of the disclosure will
It becomes more fully apparent.Drawings discussed below is only some embodiments of the present disclosure, for the ordinary skill of this field
For personnel, without creative efforts, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is the system block diagram of a kind of image evaluation method and device shown according to an exemplary embodiment.
Fig. 2 is the application scenarios schematic diagram of a kind of image evaluation method and device shown according to an exemplary embodiment.
Fig. 3 is the application scenarios schematic diagram of a kind of image evaluation method and device shown according to an exemplary embodiment.
Fig. 4 is a kind of flow chart of image evaluation method shown according to an exemplary embodiment.
Fig. 5 is a kind of flow chart of the image evaluation method shown according to another exemplary embodiment.
Fig. 6 is a kind of block diagram of image evaluation device shown according to an exemplary embodiment.
Fig. 7 is a kind of block diagram of the image evaluation device shown according to another exemplary embodiment.
Fig. 8 is the block diagram of a kind of electronic equipment shown according to an exemplary embodiment.
Fig. 9 is that a kind of computer readable storage medium schematic diagram is shown according to an exemplary embodiment.
Specific embodiment
Example embodiment is described more fully with reference to the drawings.However, example embodiment can be real in a variety of forms
It applies, and is not understood as limited to embodiment set forth herein；On the contrary, thesing embodiments are provided so that the disclosure will be comprehensively and complete
It is whole, and the design of example embodiment is comprehensively communicated to those skilled in the art.Identical appended drawing reference indicates in figure
Same or similar part, thus repetition thereof will be omitted.
In addition, described feature, structure or characteristic can be incorporated in one or more implementations in any suitable manner
In example.In the following description, many details are provided to provide and fully understand to embodiment of the disclosure.However,
It will be appreciated by persons skilled in the art that can with technical solution of the disclosure without one or more in specific detail,
Or it can be using other methods, constituent element, device, step etc..In other cases, it is not shown in detail or describes known side
Method, device, realization or operation are to avoid fuzzy all aspects of this disclosure.
Block diagram shown in the drawings is only functional entity, not necessarily must be corresponding with physically separate entity.
I.e., it is possible to realize these functional entitys using software form, or realized in one or more hardware modules or integrated circuit
These functional entitys, or these functional entitys are realized in heterogeneous networks and/or processor device and/or microcontroller device.
Flow chart shown in the drawings is merely illustrative, it is not necessary to including all content and operation/step,
It is not required to execute by described sequence.For example, some operation/steps can also decompose, and some operation/steps can close
And or part merge, therefore the sequence actually executed is possible to change according to the actual situation.
It should be understood that although herein various assemblies may be described using term first, second, third, etc., these groups
Part should not be limited by these terms.These terms are to distinguish a component and another component.Therefore, first group be discussed herein below
Part can be described as the second component without departing from the teaching of disclosure concept.As used herein, term " and/or " include associated
All combinations for listing any of project and one or more.
It will be understood by those skilled in the art that attached drawing is the schematic diagram of example embodiment, module or process in attached drawing
Necessary to not necessarily implementing the disclosure, therefore it cannot be used for the protection scope of the limitation disclosure.
The disclosure inventors have found that in existing picture quality objective evaluation algorithm, mainly include two parts: having
The image quality measure method of reference and image quality measure method without reference.Wherein have the image quality measure of reference due to
There is corresponding reference standard, the quality of image can be accurately judged by the method for machine learning.And without reference
Image quality measure needs more to be known by means of image procossing, mode due to not having input picture the knowledge of any priori
The relevant technologies such as other and machine learning.
It mainly include being assessed based on traditional feature extracting method image in existing technical solution, such
Method usually has more satisfactory as a result, being unable to satisfy actual engineering demand to small sample test set.
Further, the final purpose of image quality measure is to be quickly found out not meeting industry from large nuber of images information
The image of business demand solves the difficulty of staff's post-processing.Image quality measure algorithm is as a kind of important pretreatment
Means, people conduct extensive research image quality evaluation algorithm.But existing non-reference picture quality appraisement method
Be all based on special scenes progress apriority greatly is assumed to be precondition, and scalability is poor；In prior art it is also proposed that
Some picture quality predictor methods based on neural network model, but due to lacking a large amount of marker samples, actually answering
It is also unsatisfactory with middle performance.In large-scale field of image processing, robustness is also poor.
Based on problem above, the present disclosure proposes a kind of image evaluation method and devices.This method solves to a certain extent
Some technical problems that prior art of having determined encounters.
Fig. 1 is the system block diagram of a kind of image evaluation method and device shown according to an exemplary embodiment.
As shown in Figure 1, system architecture 100 may include terminal device 101,102,103, network 104 and server 105.
Network 104 between terminal device 101,102,103 and server 105 to provide the medium of communication link.Network 104 can be with
Including various connection types, such as wired, wireless communication link or fiber optic cables etc..
User can be used terminal device 101,102,103 and be interacted by network 104 with server 105, to receive or send out
Send message etc..Various telecommunication customer end applications, such as the application of shopping class, net can be installed on terminal device 101,102,103
The application of page browsing device, searching class application, instant messaging tools, mailbox client, social platform software etc..
Terminal device 101,102,103 can be the various electronic equipments with display screen and supported web page browsing, packet
Include but be not limited to smart phone, tablet computer, pocket computer on knee and desktop computer etc..
Server 105 can be to provide the server of various services, such as utilize terminal device 101,102,103 to user
Captured picture provides the background server supported.Server 105 can carry out the image data received the processing such as analyzing,
And processing result is fed back into terminal device.
User can obtain picture by terminal device 101,102,103, and terminal device 101,102,103 can be for example to original
Image is pre-processed, and standard picture is generated；The gradient that terminal device 101,102,103 can for example obtain the standard picture is special
Sign figure and Laplacian characteristic pattern；Terminal device 101,102,103 can for example obtain the Gradient Features figure and the height
The feature vector of this Laplce's characteristic pattern；And described eigenvector can be inputted for example and be schemed by terminal device 101,102,103
Scoring as obtaining the original image in Rating Model.
Terminal device 101,102,103 can also for example be trained supporting vector machine model by training image, to obtain
Take described image Rating Model.
User can obtain picture by terminal device 101,102,103, and terminal device 101,102,103 can be for example by picture
It is forwarded in server 105, server 105 can for example pre-process original image, generate standard picture；Server 105
The Gradient Features figure and Laplacian characteristic pattern of the standard picture can for example be obtained；Server 105 can for example obtain institute
State the feature vector of Gradient Features figure Yu the Laplacian characteristic pattern；And server 105 can be for example by the feature
The scoring of the original image is obtained in vector input picture Rating Model.Server 105 also for example can directly export image
Appraisal result.Server 105 also for example image appraisal result can return to terminal device 101,102,103.
Server 105 can also for example be trained supporting vector machine model by training image, to obtain described image
Rating Model.
Server 105 can be the server of an entity, also may be, for example, multiple server compositions, needs to illustrate
It is that image evaluation method provided by the embodiment of the present disclosure can be held by server 105 and/or terminal device 101,102,103
Row, correspondingly, image evaluation device can be set in server 105 and/or terminal device 101,102,103.And it is supplied to
The receiving end that user obtains picture is normally in terminal device 101,102,103.
According to the image evaluation method and device of the disclosure, the Gradient Features figure and Gauss of original image are obtained by calculating
Laplce's characteristic pattern；And then institute is determined according to Gradient Features figure and Laplacian characteristic pattern and input picture Rating Model
The mode of the scoring of original image is stated, fast and accurately picture quality can be assessed.
Fig. 2 is the application scenarios schematic diagram of a kind of image evaluation method and device shown according to an exemplary embodiment.
Fig. 2 illustratively illustrates the concrete application scene of the image evaluation method in insurance field, the disclosure.
In insurance field, client requires to upload the photographed image-related informations such as identity card during buying Related product,
And these information need to be saved in background data base, since image data may during acquisition, transimission and storage
There are the fuzzy all phenomenons of certain distortion.With the increase of information content, and manually check one by one verification also become be increasingly difficult to
With control, in order to improve the efficiency and real-time of storage, the certificate image by treating storage carries out image quality measure, can
It efficiently solves the problems, such as that very much data is put in storage, while saving a large amount of human cost.
In insurance field, client requires to upload the photographed image-related informations such as identity card during buying Related product,
And these information need to be saved in background data base, since image data may during acquisition, transimission and storage
There are the fuzzy all phenomenons of certain distortion.With the increase of information content, and manually check one by one verification also become be increasingly difficult to
With control.
In order to improve the efficiency and real-time of storage, it can obtain terminal acquisition client's by first passing through the certificate of setting in advance
The images such as identity card, certificate obtain terminal identity card image and are pre-processed, and generate standard picture；Determine the standard picture
Gradient Features figure and Laplacian characteristic pattern；And then determine the Gradient Features figure and the Laplacian characteristic pattern
Feature vector；Described eigenvector input certificate is obtained in the image Rating Model prestored in terminal to obtain certificate figure again
The scoring of picture.Certificate image by treating storage carries out image quality measure, can efficiently solve very much data storage
Problem, while saving a large amount of human cost.
Fig. 3 is the application scenarios schematic diagram of a kind of image evaluation method and device shown according to an exemplary embodiment.
Fig. 3 illustratively illustrates the concrete application scene of the image evaluation method in insurance benefits field, the disclosure.
In insurance benefits field, client usually requires to submit Claims Resolution data used, and the data of these paperys need with
The form of text is entered into database to check verification etc. afterwards.And the typing of data will usually be later than the upper of data
It passes, once there is the case where image document is fuzzy to be uploaded, generally requires to expend huge manpower and material resources and remove number from magnanimity
According to middle some initial data of lookup.
For example the image in the matching data of client can be uploaded onto the server end in real time, server locates image in advance
Reason, and then the scoring of image is obtained in input picture Rating Model.It is estimated in real time by the image document to upload
Judgement, can avoid to the full extent the appearance of such case, to greatly save cost of labor.
Fig. 4 is a kind of flow chart of image evaluation method shown according to an exemplary embodiment.Image evaluation method 40
Including at least step S402 to S408.
As shown in figure 4, pre-processing in S402 to original image, standard picture is generated.It can be for example, to the original
Beginning image is adjusted resolution processes；Original image after adjustment resolution processes is standardized；By standard
Original image after change processing carries out color space conversion process, generates the standard picture.
In one embodiment, being adjusted resolution processes to the original image includes: to less than predetermined resolution
Image carry out image enhancement processing, increase the resolution ratio of original image；Image pressure is carried out to the image for being greater than predetermined resolution
Contracting processing, to reduce the resolution ratio of original image, convenient for other subsequent calculating.
In one embodiment, being standardized to the original image after adjustment resolution processes includes: image
Standardization may be, for example, that the format of image is subjected to unified arrangement, for example can uniformly be converted into the image of jpg format with
Carry out subsequent processing.Also for example the outer dimension of image can be handled, by Image Adjusting into scheduled size.
In one embodiment, by after standardization original image carry out color space conversion process include: by
Original image after standardization is transformed into YcbCr color space by RGB color.Wherein YCbCr is color sky
Between one kind, it will usually in film image continuous processing or digital photographic systems in.Y is the brightness of color
(luma) ingredient and CB and CR are then blue and red concentration excursion amount composition.As geometrically being described with coordinate space
Coordinate set, color space describe color set with mathematical way.The basic colour model of common 3 is RGB, CMYK and
YUV.YCbCr is then a part in world's number organizing video standard development, is that YUV process scales and what is deviated turns in fact
Version.Wherein Y is consistent with the Y meaning in YUV, and Cb, Cr equally refer to color, only different on representation method.In YUV family
In race, YCbCr is in computer systems using most members, and application field is very extensive, and JPEG, MPEG are all made of this lattice
Formula.The YUV that general people are said refers to YCbCr mostly.
In one embodiment, image is transformed into YCbCr color space from RGB color, conversion formula can be such as (1)
It is shown:
Wherein R, G, B respectively indicate the value of obstructed Color Channel on RGB color.
For obtained YCbCr image, it can extract the channel image Y and handle image as standard and be labeled as I (x, y).
In S404, the Gradient Features figure and Laplacian characteristic pattern of the standard picture are obtained.It can be such as: by institute
It states standard picture and carries out the gradient transformation generation Gradient Features figure；And the Gradient Features figure is subjected to Laplacian
Generate the Laplacian characteristic pattern.
Wherein, since image is stored in the form of digital picture in a computer, i.e., image is discrete number letter
Number, the differential in continuous signal is replaced using difference to the gradient of digital picture.
Image can be regarded as two-dimensional discrete function, image gradient is exactly the derivation of this two-dimensional discrete function in fact:
Image gradient: G (x, y)=dx (i, j)+dy (i, j)；
Dx (i, j)=I (i+1, j)-I (i, j)；
Dy (i, j)=I (i, j+1)-I (i, j)；
Wherein, I is the value of image pixel, and (i, j) is the coordinate of pixel.
Image gradient can also generally use intermediate value difference:
Dx (i, j)=[I (i+1, j)-I (i-1, j)]/2；
Dy (i, j)=[I (i, j+1)-I (i, j-1)]/2；
Image border is typically all to be realized by carrying out gradient algorithm to image.
In one embodiment, for standard picture I (x, y), the Gradient Features figure of image can be calculated by formula (2),
And it is labeled as GI。
WhereinIndicate linear convolution operation, hd(d ∈ (x, y)) respectively indicates the direction x and the gaussian filtering in the direction y is calculated
Son, shown in calculation method such as formula (3).
Wherein, Laplace operator is a kind of high-pass filter, is image greyscale function in two vertical direction Second Order Partials
The sum of derivative.In the case where discrete digital image, two under continuous situation are directly replaced with the second differnce of image greyscale grade
Rank partial derivative, it is very sensitive to noise, it often will appear pseudo-edge response when extracting edge.To overcome Laplace operator not
Foot preferably first carries out low-pass filtering to digitized video, inhibits noise.Gaussian function is a kind of normalization low-pass filter well,
It can be used for carrying out digitized video low-pass filtering to reduce the influence of noise, recycle Laplace operator to extract on this basis
Edge, here it is Gauss-Laplace, also known as LOG (Laplacian of Gaussian) operators.
In one embodiment, for standard picture I (x, y), the Gauss-La Pula of image can be calculated by formula (4)
This characteristic pattern is simultaneously labeled as LI。
Wherein Gauss-Laplace operator can be indicated by formula (5), specifically
In S406, the feature vector of the Gradient Features figure Yu the Laplacian characteristic pattern is obtained.Can for example,
Determine the gray scale of the Gradient Features figure Yu the Laplacian characteristic pattern respectively by gray scale setting；Described in obtaining respectively
The Feature Mapping figure of Gradient Features figure and the Laplacian characteristic pattern；And the gradient is determined according to Feature Mapping figure
The feature vector of characteristic pattern and the Laplacian characteristic pattern.
In one embodiment, according to the Gradient Features figure and the Laplacian characteristic pattern, we are by feature
Scheme GIGray level be set as M, characteristic pattern LIGray level be set as N.Then corresponding Feature Mapping figure can pass through formula (6)
It indicates.
KM, n=P (G=gm, L=ln), m=1 ..., M；N=1 ... N. (6)
By obtained Feature Mapping figure,
According to KM, n, defining corresponding independent coefficient is DM, n.Corresponding calculation formula is as shown below:
Then corresponding feature vector can be defined as
In S408, the scoring of the original image will be obtained in described eigenvector input picture Rating Model.
In one embodiment, supporting vector machine model can be trained by training image, to obtain described image
Rating Model.It can be for example, determining the Gradient Features figure and the Laplacian characteristic pattern of training image；According to described
Characteristic pattern determines the feature vector of the training image；Determine the scoring of the training image；And according to the training image
Scoring supporting vector machine model is trained with the feature vector of the training image, scored mould with obtaining described image
Type.The content of model training will be described in detail in the corresponding embodiment of Fig. 5.
By that described eigenvector will be inputted in trained image Rating Model, the available original image
Scoring.User can carry out subsequent processing based on the scoring of described image, and the application is not limited.
According to the image evaluation method of the disclosure, the Gradient Features figure and Gauss La Pula of original image are obtained by calculating
This characteristic pattern；It is determined in turn with Laplacian characteristic pattern with input picture Rating Model according to Gradient Features figure described original
The mode of the scoring of image can fast and accurately assess picture quality.
It will be clearly understood that the present disclosure describes how to form and use particular example, but the principle of the disclosure is not limited to
These exemplary any details.On the contrary, the introduction based on disclosure disclosure, these principles can be applied to many other
Embodiment.
Fig. 5 is a kind of flow chart of the image evaluation method shown according to another exemplary embodiment.Image shown in fig. 5
Appraisal procedure 50 " is trained supporting vector machine model by training image, to obtain described image Rating Model." it is detailed
Thin description,
As shown in figure 5, in S502, determine training image the Gradient Features figure and the Laplacian feature
Figure.Training image can be obtained for example by various image sources, can also be generated by historical image data, the application not with
This is limited.
In S504, the feature vector of the training image is determined according to the characteristic pattern.
In S506, the scoring of the training image is determined.
In S508, according to the feature vector of the scoring of the training image and the training image to support vector machines mould
Type is trained, to obtain described image Rating Model.For example the training image can be inputted in supporting vector machine model；It is logical
The scoring for crossing the training image is adjusted the parameter in the supporting vector machine model；And the support vector machines mould
When parameter in type meets threshold value, described image Rating Model is generated.
It can according to the method described above, by the feature vector, X (Q of the tag image of acquisitionG, QL, PG, PL) carry out image
Then image and corresponding picture quality scoring Y are inputted supporting vector machine model by quality score.By using supporting vector
The method that machine returns is trained data, and obtains corresponding image Rating Model.
The image evaluation method of the disclosure can have by the method that gradient and Gauss-Laplace Fusion Features count
The blind image quality measure of the solution of effect.
It will be appreciated by those skilled in the art that realizing that all or part of the steps of above-described embodiment is implemented as being executed by CPU
Computer program.When the computer program is executed by CPU, above-mentioned function defined by the above method that the disclosure provides is executed
Energy.The program can store in a kind of computer readable storage medium, which can be read-only memory, magnetic
Disk or CD etc..
Further, it should be noted that above-mentioned attached drawing is only the place according to included by the method for disclosure exemplary embodiment
Reason schematically illustrates, rather than limits purpose.It can be readily appreciated that above-mentioned processing shown in the drawings is not indicated or is limited at these
The time sequencing of reason.In addition, be also easy to understand, these processing, which can be, for example either synchronously or asynchronously to be executed in multiple modules.
Following is embodiment of the present disclosure, can be used for executing embodiments of the present disclosure.It is real for disclosure device
Undisclosed details in example is applied, embodiments of the present disclosure is please referred to.
Fig. 6 is a kind of block diagram of image evaluation device shown according to an exemplary embodiment.As shown in fig. 6, image is commented
Estimating device 60 includes: original image processing module 602, standard picture processing module 604, feature vector module 606, and scoring
Module 608.
Original image processing module 602 generates standard picture for pre-processing to original image；Can for example, to institute
It states original image and is adjusted resolution processes；Original image after adjustment resolution processes is standardized；It will
Original image after standardization carries out color space conversion process, generates the standard picture.
Standard picture processing module 604 is used to obtain the Gradient Features figure and Laplacian feature of the standard picture
Figure；It can be such as: the standard picture being subjected to gradient transformation and generates the Gradient Features figure；And by the Gradient Features figure into
Row Laplacian generates the Laplacian characteristic pattern.
Feature vector module 606 be used to obtain the feature of the Gradient Features figure and the Laplacian characteristic pattern to
Amount；It can be for example, determining the gray scale of the Gradient Features figure Yu the Laplacian characteristic pattern respectively by gray scale setting；Point
The Feature Mapping figure of the Gradient Features figure Yu the Laplacian characteristic pattern is not obtained；And it is true according to Feature Mapping figure
The feature vector of fixed the Gradient Features figure and the Laplacian characteristic pattern.
Grading module 608 is used to that the matter of the original image will to be obtained in described eigenvector input picture Rating Model
Amount scoring.
According to the image evaluation device of the disclosure, the Gradient Features figure and Gauss La Pula of original image are obtained by calculating
This characteristic pattern；It is determined in turn with Laplacian characteristic pattern with input picture Rating Model according to Gradient Features figure described original
The mode of the scoring of image can fast and accurately assess picture quality.
Fig. 7 is a kind of block diagram of the image evaluation device shown according to another exemplary embodiment.Image as shown in Figure 7
Device 70 is assessed on the basis of image evaluation device 60 further include: model training module 702.
Model training module 702 is for being trained supporting vector machine model by training image, to obtain the figure
As Rating Model.It can be for example, determining the Gradient Features figure and the Laplacian characteristic pattern of training image；According to institute
State the feature vector that characteristic pattern determines the training image；Determine the scoring of the training image；And schemed according to the training
The scoring of picture and the feature vector of the training image are trained supporting vector machine model, to obtain described image scoring mould
Type.
Fig. 8 is the block diagram of a kind of electronic equipment shown according to an exemplary embodiment.
The electronic equipment 200 of this embodiment according to the disclosure is described referring to Fig. 8.The electronics that Fig. 8 is shown
Equipment 200 is only an example, should not function to the embodiment of the present disclosure and use scope bring any restrictions.
As shown in figure 8, electronic equipment 200 is showed in the form of universal computing device.The component of electronic equipment 200 can wrap
It includes but is not limited to: at least one processing unit 210, at least one storage unit 220, (including the storage of the different system components of connection
Unit 220 and processing unit 210) bus 230, display unit 240 etc..
Wherein, the storage unit is stored with program code, and said program code can be held by the processing unit 210
Row, so that the processing unit 210 executes described in this specification above-mentioned electronic prescription circulation processing method part according to this
The step of disclosing various illustrative embodiments.For example, the processing unit 210 can be executed such as Fig. 4, walked shown in Fig. 5
Suddenly.
The storage unit 220 may include the readable medium of volatile memory cell form, such as random access memory
Unit (RAM) 2201 and/or cache memory unit 2202 can further include read-only memory unit (ROM) 2203.
The storage unit 220 can also include program/practical work with one group of (at least one) program module 2205
Tool 2204, such program module 2205 includes but is not limited to: operating system, one or more application program, other programs
It may include the realization of network environment in module and program data, each of these examples or certain combination.
Bus 230 can be to indicate one of a few class bus structures or a variety of, including storage unit bus or storage
Cell controller, peripheral bus, graphics acceleration port, processing unit use any bus structures in a variety of bus structures
Local bus.
Electronic equipment 200 can also be with one or more external equipments 300 (such as keyboard, sensing equipment, bluetooth equipment
Deng) communication, can also be enabled a user to one or more equipment interact with the electronic equipment 200 communicate, and/or with make
Any equipment (such as the router, modulation /demodulation that the electronic equipment 200 can be communicated with one or more of the other calculating equipment
Device etc.) communication.This communication can be carried out by input/output (I/O) interface 250.Also, electronic equipment 200 can be with
By network adapter 260 and one or more network (such as local area network (LAN), wide area network (WAN) and/or public network,
Such as internet) communication.Network adapter 260 can be communicated by bus 230 with other modules of electronic equipment 200.It should
Understand, although not shown in the drawings, other hardware and/or software module can be used in conjunction with electronic equipment 200, including but unlimited
In: microcode, device driver, redundant processing unit, external disk drive array, RAID system, tape drive and number
According to backup storage system etc..
Through the above description of the embodiments, those skilled in the art is it can be readily appreciated that example described herein is implemented
Mode can also be realized by software realization in such a way that software is in conjunction with necessary hardware.Therefore, according to the disclosure
The technical solution of embodiment can be embodied in the form of software products, which can store non-volatile at one
Property storage medium (can be CD-ROM, USB flash disk, mobile hard disk etc.) in or network on, including some instructions are so that a calculating
Equipment (can be personal computer, server or network equipment etc.) executes the above method according to disclosure embodiment.
Fig. 9 schematically shows a kind of computer readable storage medium schematic diagram in disclosure exemplary embodiment.
Refering to what is shown in Fig. 9, describing the program product for realizing the above method according to embodiment of the present disclosure
400, can using portable compact disc read only memory (CD-ROM) and including program code, and can in terminal device,
Such as it is run on PC.However, the program product of the disclosure is without being limited thereto, in this document, readable storage medium storing program for executing can be with
To be any include or the tangible medium of storage program, the program can be commanded execution system, device or device use or
It is in connection.
Described program product can be using any combination of one or more readable mediums.Readable medium can be readable letter
Number medium or readable storage medium storing program for executing.Readable storage medium storing program for executing for example can be but be not limited to electricity, magnetic, optical, electromagnetic, infrared ray or
System, device or the device of semiconductor, or any above combination.The more specific example of readable storage medium storing program for executing is (non exhaustive
List) include: electrical connection with one or more conducting wires, portable disc, hard disk, random access memory (RAM), read-only
Memory (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc read only memory
(CD-ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.
The computer readable storage medium may include in a base band or the data as the propagation of carrier wave a part are believed
Number, wherein carrying readable program code.The data-signal of this propagation can take various forms, including but not limited to electromagnetism
Signal, optical signal or above-mentioned any appropriate combination.Readable storage medium storing program for executing can also be any other than readable storage medium storing program for executing
Readable medium, the readable medium can send, propagate or transmit for by instruction execution system, device or device use or
Person's program in connection.The program code for including on readable storage medium storing program for executing can transmit with any suitable medium, packet
Include but be not limited to wireless, wired, optical cable, RF etc. or above-mentioned any appropriate combination.
Can with any combination of one or more programming languages come write for execute the disclosure operation program
Code, described program design language include object oriented program language-Java, C++ etc., further include conventional
Procedural programming language-such as " C " language or similar programming language.Program code can be fully in user
It calculates and executes in equipment, partly executes on a user device, being executed as an independent software package, partially in user's calculating
Upper side point is executed on a remote computing or is executed in remote computing device or server completely.It is being related to far
Journey calculates in the situation of equipment, and remote computing device can pass through the network of any kind, including local area network (LAN) or wide area network
(WAN), it is connected to user calculating equipment, or, it may be connected to external computing device (such as utilize ISP
To be connected by internet).
Above-mentioned computer-readable medium carries one or more program, when said one or multiple programs are by one
When the equipment executes, so that the computer-readable medium implements function such as: being pre-processed to original image, generate standard drawing
Picture；Obtain the Gradient Features figure and Laplacian characteristic pattern of the standard picture；Obtain the Gradient Features figure with it is described
The feature vector of Laplacian characteristic pattern；And by described eigenvector input picture Rating Model to obtain the original
The scoring of beginning image.
It will be appreciated by those skilled in the art that above-mentioned each module can be distributed in device according to the description of embodiment, it can also
Uniquely it is different from one or more devices of the present embodiment with carrying out corresponding change.The module of above-described embodiment can be merged into
One module, can also be further split into multiple submodule.
By the description of above embodiment, those skilled in the art is it can be readily appreciated that example embodiment described herein
It can also be realized in such a way that software is in conjunction with necessary hardware by software realization.Therefore, implemented according to the disclosure
The technical solution of example can be embodied in the form of software products, which can store in a non-volatile memories
In medium (can be CD-ROM, USB flash disk, mobile hard disk etc.) or on network, including some instructions are so that a calculating equipment (can
To be personal computer, server, mobile terminal or network equipment etc.) it executes according to the method for the embodiment of the present disclosure.
It is particularly shown and described the exemplary embodiment of the disclosure above.It should be appreciated that the present disclosure is not limited to
Detailed construction, set-up mode or implementation method described herein；On the contrary, disclosure intention covers included in appended claims
Various modifications and equivalence setting in spirit and scope.
In addition, structure shown by this specification Figure of description, ratio, size etc., only to cooperate specification institute
Disclosure, for skilled in the art realises that be not limited to the enforceable qualifications of the disclosure with reading, therefore
Do not have technical essential meaning, the modification of any structure, the change of proportionate relationship or the adjustment of size are not influencing the disclosure
Under the technical effect and achieved purpose that can be generated, it should all still fall in technology contents disclosed in the disclosure and obtain and can cover
In the range of.Meanwhile cited such as "upper" in this specification, " first ", " second " and " one " term, be also only and be convenient for
Narration is illustrated, rather than to limit the enforceable range of the disclosure, relativeness is altered or modified, without substantive change
Under technology contents, when being also considered as the enforceable scope of the disclosure.
Claims (13)
1. a kind of image evaluation method characterized by comprising
Original image is pre-processed, standard picture is generated；
Obtain the Gradient Features figure and Laplacian characteristic pattern of the standard picture；
Obtain the feature vector of the Gradient Features figure Yu the Laplacian characteristic pattern；And
The scoring of the original image will be obtained in described eigenvector input picture Rating Model.
2. the method as described in claim 1, which is characterized in that further include:
Supporting vector machine model is trained by training image, to obtain described image Rating Model.
3. method according to claim 2, which is characterized in that supporting vector machine model is trained by training image,
Include: to obtain described image Rating Model
Determine the Gradient Features figure and the Laplacian characteristic pattern of training image；
The feature vector of the training image is determined according to the characteristic pattern；
Determine the scoring of the training image；And
Supporting vector machine model is trained according to the feature vector of the scoring of the training image and the training image, with
Obtain described image Rating Model.
4. method as claimed in claim 3, which is characterized in that according to the scoring of the training image and the training image
Feature vector is trained supporting vector machine model, includes: to obtain described image Rating Model
The training image is inputted in supporting vector machine model；
The parameter in the supporting vector machine model is adjusted by the scoring of the training image；And
When parameter in the supporting vector machine model meets threshold value, described image Rating Model is generated.
5. the method as described in claim 1, which is characterized in that pre-processed to original image, generating standard picture includes:
Resolution processes are adjusted to the original image；
Original image after adjustment resolution processes is standardized；
Original image after standardization is subjected to color space conversion process, generates the standard picture.
6. method as claimed in claim 5, which is characterized in that the original image after standardization is carried out color space
Conversion process includes:
Original image after standardization is transformed into YcbCr color space by RGB color.
7. the method as described in claim 1, which is characterized in that Gradient Features figure and the Gauss drawing for obtaining the standard picture are general
Lust's sign figure includes:
The standard picture is subjected to gradient transformation and generates the Gradient Features figure；And
The standard picture is subjected to Laplacian and generates the Laplacian characteristic pattern.
8. the method as described in claim 1, which is characterized in that obtain the Gradient Features figure and the Laplacian is special
The feature vector of sign figure includes:
Determine the gray scale of the Gradient Features figure Yu the Laplacian characteristic pattern respectively by gray scale setting；
The Feature Mapping figure of the Gradient Features figure Yu the Laplacian characteristic pattern is obtained respectively；And
The feature vector of the Gradient Features figure Yu the Laplacian characteristic pattern is determined according to Feature Mapping figure.
9. the method as described in claim 1, which is characterized in that according to Feature Mapping figure determine the Gradient Features figure with it is described
The feature vector of Laplacian characteristic pattern includes:
Determine feature vector coefficient；And
The Gradient Features figure and the Laplacian feature are determined according to described eigenvector coefficient and Feature Mapping figure
The feature vector of figure.
10. a kind of image evaluation device characterized by comprising
Original image processing module generates standard picture for pre-processing to original image；
Standard picture processing module, for obtaining the Gradient Features figure and Laplacian characteristic pattern of the standard picture；
Feature vector module, for obtaining the feature vector of the Gradient Features figure Yu the Laplacian characteristic pattern；With
And
Grading module, for will be commented in described eigenvector input picture Rating Model with the quality for obtaining the original image
Point.
11. device as claimed in claim 10, which is characterized in that further include:
Model training module, for being trained by training image to supporting vector machine model, to obtain described image scoring
Model.
12. a kind of electronic equipment characterized by comprising
One or more processors；
Storage device, for storing one or more programs；
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
The now method as described in any in claim 1-9.
13. a kind of computer-readable medium, is stored thereon with computer program, which is characterized in that described program is held by processor
The method as described in any in claim 1-9 is realized when row.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811393400.8A CN109522960A (en) | 2018-11-21 | 2018-11-21 | Image evaluation method, device, electronic equipment and computer-readable medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811393400.8A CN109522960A (en) | 2018-11-21 | 2018-11-21 | Image evaluation method, device, electronic equipment and computer-readable medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109522960A true CN109522960A (en) | 2019-03-26 |
Family
ID=65778148
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811393400.8A Pending CN109522960A (en) | 2018-11-21 | 2018-11-21 | Image evaluation method, device, electronic equipment and computer-readable medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109522960A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109949250A (en) * | 2019-03-29 | 2019-06-28 | 北京奇艺世纪科技有限公司 | A kind of image processing method and device |
| CN110428368A (en) * | 2019-07-31 | 2019-11-08 | 北京金山云网络技术有限公司 | A kind of algorithm evaluation method, device, electronic equipment and readable storage medium storing program for executing |
| CN111145153A (en) * | 2019-12-25 | 2020-05-12 | 上海肇观电子科技有限公司 | Image processing method, circuit, visual impairment assisting device, electronic device, and medium |
| WO2020253773A1 (en) * | 2019-06-21 | 2020-12-24 | 腾讯科技（深圳）有限公司 | Medical image classification method, model training method, computing device and storage medium |
| CN112288699A (en) * | 2020-10-23 | 2021-01-29 | 北京百度网讯科技有限公司 | Method, device, equipment and medium for evaluating relative definition of image |
| CN112712550A (en) * | 2019-10-24 | 2021-04-27 | 马上消费金融股份有限公司 | Image quality evaluation method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101980242A (en) * | 2010-09-30 | 2011-02-23 | 徐勇 | Human face discrimination method and system and public safety system |
| CN104658002A (en) * | 2015-03-10 | 2015-05-27 | 浙江科技学院 | Non-reference image objective quality evaluation method |
| CN105160667A (en) * | 2015-08-26 | 2015-12-16 | 西安交通大学 | Blind image quality evaluation method based on combining gradient signal and Laplacian of Gaussian (LOG) signal |
| CN106709916A (en) * | 2017-01-19 | 2017-05-24 | 泰康保险集团股份有限公司 | Image quality assessment method and device |
-
2018
- 2018-11-21 CN CN201811393400.8A patent/CN109522960A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101980242A (en) * | 2010-09-30 | 2011-02-23 | 徐勇 | Human face discrimination method and system and public safety system |
| CN104658002A (en) * | 2015-03-10 | 2015-05-27 | 浙江科技学院 | Non-reference image objective quality evaluation method |
| CN105160667A (en) * | 2015-08-26 | 2015-12-16 | 西安交通大学 | Blind image quality evaluation method based on combining gradient signal and Laplacian of Gaussian (LOG) signal |
| CN106709916A (en) * | 2017-01-19 | 2017-05-24 | 泰康保险集团股份有限公司 | Image quality assessment method and device |
Non-Patent Citations (2)
| Title |
|---|
| WUFENG XUE ET AL.: ""Blind Image Quality Assessment Using Joint Statistics of Gradient Magnitude and Laplacian Features"", 《IEEE TRANSACTIONS ON IMAGE PROCESSING》 * |
| 谭红宝: ""基于机器学习的无参考立体图像质量评价方法研究"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109949250A (en) * | 2019-03-29 | 2019-06-28 | 北京奇艺世纪科技有限公司 | A kind of image processing method and device |
| CN109949250B (en) * | 2019-03-29 | 2021-05-18 | 北京奇艺世纪科技有限公司 | Image processing method and device |
| WO2020253773A1 (en) * | 2019-06-21 | 2020-12-24 | 腾讯科技（深圳）有限公司 | Medical image classification method, model training method, computing device and storage medium |
| US11954852B2 (en) | 2019-06-21 | 2024-04-09 | Tencent Technology (Shenzhen) Company Limited | Medical image classification method, model training method, computing device, and storage medium |
| CN110428368A (en) * | 2019-07-31 | 2019-11-08 | 北京金山云网络技术有限公司 | A kind of algorithm evaluation method, device, electronic equipment and readable storage medium storing program for executing |
| CN112712550A (en) * | 2019-10-24 | 2021-04-27 | 马上消费金融股份有限公司 | Image quality evaluation method and device |
| CN111145153A (en) * | 2019-12-25 | 2020-05-12 | 上海肇观电子科技有限公司 | Image processing method, circuit, visual impairment assisting device, electronic device, and medium |
| CN111145153B (en) * | 2019-12-25 | 2023-10-03 | 上海肇观电子科技有限公司 | Image processing method, circuit, vision-impaired auxiliary equipment, electronic equipment and medium |
| CN112288699A (en) * | 2020-10-23 | 2021-01-29 | 北京百度网讯科技有限公司 | Method, device, equipment and medium for evaluating relative definition of image |
| EP3879454A3 (en) * | 2020-10-23 | 2022-02-16 | Beijing Baidu Netcom Science and Technology Co., Ltd. | Method and apparatus for evaluating image relative definition, device and medium |
| CN112288699B (en) * | 2020-10-23 | 2024-02-09 | 北京百度网讯科技有限公司 | Method, device, equipment and medium for evaluating relative definition of image |
| US11921276B2 (en) | 2020-10-23 | 2024-03-05 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for evaluating image relative definition, device and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109522960A (en) | Image evaluation method, device, electronic equipment and computer-readable medium | |
| Kundu et al. | No-reference quality assessment of tone-mapped HDR pictures | |
| CN108197618B (en) | Method and device for generating human face detection model | |
| Bellavia et al. | Dissecting and reassembling color correction algorithms for image stitching | |
| CN108140130A (en) | The bilateral image procossing that edge perceives | |
| CA3065062C (en) | Simulating image capture | |
| CN110222694B (en) | Image processing method, image processing device, electronic equipment and computer readable medium | |
| TW202026948A (en) | Methods and devices for biological testing and storage medium thereof | |
| US11514263B2 (en) | Method and apparatus for processing image | |
| CN107622504B (en) | Method and device for processing pictures | |
| Waheed et al. | Multifocus watermarking approach based on discrete cosine transform | |
| Wu et al. | Color-to-grayscale conversion through weighted multiresolution channel fusion | |
| CN106709916A (en) | Image quality assessment method and device | |
| CN113658085B (en) | Image processing method and device | |
| Pereira et al. | Gamut mapping spatially varying reflectance with an improved BRDF similarity metric | |
| CN110335330A (en) | Image simulation generation method and its system, deep learning algorithm training method and electronic equipment | |
| CN116665228B (en) | Image processing method and device | |
| CN109493296A (en) | Image enchancing method, device, electronic equipment and computer-readable medium | |
| CN113706400A (en) | Image correction method, image correction device, microscope image correction method, and electronic apparatus | |
| Krasula et al. | Objective evaluation of naturalness, contrast, and colorfulness of tone-mapped images | |
| CN116912648A (en) | Method, device, equipment and storage medium for generating material parameter identification model | |
| JP2021189527A (en) | Information processing device, information processing method, and program | |
| CN110569698B (en) | Image target detection and semantic segmentation method and device | |
| Li et al. | Instrument reading recognition by deep learning of capsules network model for digitalization in Industrial Internet of Things | |
| US10255674B2 (en) | Surface reflectance reduction in images using non-specular portion replacement |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190326 |