CN113850764A - Method, system and equipment for controlling image quality of claim material based on Gaussian blur - Google Patents
Method, system and equipment for controlling image quality of claim material based on Gaussian blur Download PDFInfo
- Publication number
- CN113850764A CN113850764A CN202111053937.1A CN202111053937A CN113850764A CN 113850764 A CN113850764 A CN 113850764A CN 202111053937 A CN202111053937 A CN 202111053937A CN 113850764 A CN113850764 A CN 113850764A
- Authority
- CN
- China
- Prior art keywords
- image
- image quality
- algorithm
- definition
- quality control
- 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
- 239000000463 material Substances 0.000 title claims abstract description 155
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 111
- 238000012545 processing Methods 0.000 claims abstract description 30
- 238000004458 analytical method Methods 0.000 claims abstract description 20
- 238000011156 evaluation Methods 0.000 claims abstract description 15
- 238000003908 quality control method Methods 0.000 claims description 41
- 238000004364 calculation method Methods 0.000 claims description 26
- 230000008859 change Effects 0.000 claims description 19
- 238000010606 normalization Methods 0.000 claims description 17
- 230000006870 function Effects 0.000 claims description 15
- 241001156002 Anthonomus pomorum Species 0.000 claims description 7
- 101100517192 Arabidopsis thaliana NRPD1 gene Proteins 0.000 claims description 7
- 101150094905 SMD2 gene Proteins 0.000 claims description 7
- 238000007477 logistic regression Methods 0.000 claims description 6
- 238000010801 machine learning Methods 0.000 claims description 6
- 230000003993 interaction Effects 0.000 claims description 3
- 238000004590 computer program Methods 0.000 description 13
- 230000008569 process Effects 0.000 description 6
- 238000013441 quality evaluation Methods 0.000 description 5
- 238000012706 support-vector machine Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000012216 screening Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000012954 risk control Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
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
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/02—Computing arrangements based on specific mathematical models using fuzzy logic
-
- 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)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Automation & Control Theory (AREA)
- Mathematical Analysis (AREA)
- Biomedical Technology (AREA)
- Fuzzy Systems (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Evolutionary Biology (AREA)
- Algebra (AREA)
- Computational Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Quality & Reliability (AREA)
- Medical Informatics (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention provides a method for controlling the quality of an image of a claim material based on Gaussian blur, which comprises the following steps: acquiring an image of the input claim material; establishing a claim material image quality scoring model, comprehensively processing an image definition evaluation index algorithm, and executing computer vision image quality scoring of the claim material image; obtaining an image quality grading result of the claim material image quality grading model, performing Gaussian fuzzy comparison on the claim material image, and performing non-reference image quality index analysis based on computer vision; and outputting a clear judgment result of the claim material image.
Description
Technical Field
The invention relates to the field of image processing, in particular to a method, a system and equipment for controlling the image quality of a claim material based on Gaussian blur.
Background
With the insurance field enabled by financial science and technology, some business scenes are very suitable for being replaced by artificial intelligence due to the characteristics of high repetition rate and experience guidance, and the method plays a role in reducing cost and improving efficiency for insurance enterprises in the future intelligent upgrading process of the system. For insurance claims, the image quality control of claim materials directly influences the timeliness of the subsequent processes of loss assessment, claim verification and the like of the claims, and has important significance for anti-fraud of risk control inside insurance companies. The insurance industry belongs to the personnel intensive industry to a certain extent, the labor cost of China insurance enterprises accounts for about 30% of the total cost, the image quality control of claim settlement materials through system detection saves a large amount of repeated labor in a certain sense, and the profit level of the enterprises is directly reduced. Therefore, the image quality control of the claim settlement material has important significance for the intellectualization of the claim settlement system of the insurance company.
The image quality evaluation means that the degree of visual distortion of an image is evaluated by analyzing the correlation characteristics of image signals. In image sharpness determination, sharpness determination is often low due to various factors of generated images (such as brightness and hue at shooting angles) and the difference between the sharpness determination and human subjective perception is too large.
And because the claim materials are mostly shot by people (clients), the uniformity and the normalization cannot be realized, and whether the pictures uploaded by the clients can be used or not can be judged by the traditional industry only through the manual judgment of business personnel, the image definition of the claim materials is judged to be the reference-free image quality judgment.
In the conventional image quality judgment, due to the above various reasons, full-reference image quality evaluation and half-reference image quality evaluation are mostly used in practical application, and such methods mostly quantize the difference between a reference image and a distorted image through modes such as feature extraction, and often need the specification and unification of the images.
Disclosure of Invention
One of the objectives of the present invention is to provide a method, a system, and an apparatus for controlling image quality of claim material based on gaussian blur, which can improve accuracy of image definition determination in the field for reference-free image quality analysis of claim material, assist service personnel to follow up the process better, and improve efficiency.
One of the objectives of the present invention is to provide a method, a system and a device for controlling image quality of claim material based on gaussian blur, which can achieve an optimal solution for resolution determination, subjective determination and human determination by combining a plurality of algorithms for image resolution, and also for the field of claim material, thereby greatly improving the accuracy and practical application efficiency of clear image screening.
In order to achieve at least one of the above objects, the present invention provides a method for controlling image quality of claim material based on gaussian blur, which comprises the following steps:
acquiring an image of the input claim material;
establishing a claim material image quality scoring model, comprehensively processing an image definition evaluation index algorithm, and executing computer vision image quality scoring of the claim material image; and
acquiring an image quality grading result of a claim material image of the claim material image quality grading model, performing Gaussian fuzzy comparison on the claim material image, and performing non-reference image quality index analysis based on computer vision;
and outputting a clear judgment result of the claim material image.
In some embodiments, the step of analyzing the no-reference image quality index based on computer vision further comprises the following steps: and performing Gaussian fuzzy comparison on the claim material image, performing vertical fuzzy comparison on the same image for multiple times, obtaining image indexes of the image to be detected and the fuzzy image obtained by calculation through a comprehensive algorithm of a claim material image quality grading model, determining the definition value according to the change conditions of the indexes, wherein the image is clearer when the change of the calculation result is smaller, and is fuzzy when the change of the calculation result is smaller.
In some embodiments, the method for controlling image quality of claim material based on gaussian blur further comprises the following steps: executing normalization processing of various algorithm indexes of final image definition; and calculating a final image quality score in a weighted manner.
In some embodiments, the method for controlling image quality of claim material based on gaussian blur further comprises the following steps:
acquiring comprehensive definition scores of all input claims material images; and
classifying the input images into clear or non-clear pictures according to a threshold value of a definition level, establishing an image definition database and recording the definition score of each image;
default blurring times for executing multiple times of blurring on the claim material image to be detected by adopting Gaussian blurring is preset to be 4-5 times;
wherein the threshold of the definition level is preset and divided into 2, 3 and 5 levels;
when the threshold is judged, the obtained change value of each basic algorithm calculates the final image clear value in a weighting mode, and the weight value obtains corresponding weight parameter by executing the machine learning-based logistic regression and SVM algorithm.
In some embodiments, wherein the image sharpness evaluation indicator algorithm is selected from one or more of Brenner gradient algorithm, Laplacian gradient algorithm, SMD grayscale variance algorithm, SMD2 grayscale variance product algorithm, energy gradient algorithm, Vollath function algorithm, entropy function algorithm, EAV point sharpness algorithm, NRSS gradient structure similarity.
According to another aspect of the present invention, there is also provided a claim material image quality control apparatus based on gaussian blur, including:
a memory for storing a software application,
and the processor is used for executing the software application programs, and each program of the software application programs correspondingly executes the steps in the method for controlling the image quality of the claim material based on the Gaussian blur.
According to another aspect of the present invention, there is also provided a gaussian blur based claim material image quality control system, including a claim material image quality control client subsystem and a claim material image quality control service subsystem, the claim material image quality control client subsystem being provided with a human-computer interaction unit for inputting an image of a claim material, the claim material image quality control service subsystem being configured to: acquiring an image of the input claim material; establishing a claim material image quality scoring model, comprehensively processing an image definition evaluation index algorithm, and executing computer vision image quality scoring of the claim material image; obtaining an image quality grading result of the claim material image quality grading model, performing Gaussian fuzzy comparison on the claim material image, and performing non-reference image quality index analysis based on computer vision; the claim material image quality control service subsystem outputs a clear judgment result of a claim material image to the claim material image quality control client subsystem.
In some embodiments, the claim material image quality control service subsystem comprises an image quality scoring unit and a no-reference image quality index analysis unit, the image quality scoring unit establishes a claim material image quality scoring model, and comprehensively processes an image sharpness evaluation index algorithm, wherein the algorithm is selected from one or more of Brenner gradient algorithm, Laplacian gradient algorithm, SMD grayscale variance algorithm, SMD2 grayscale variance product algorithm, energy gradient algorithm, Vollath function algorithm, entropy function algorithm, EAV point sharpness algorithm, NRSS gradient structure similarity; the non-reference image quality index analysis unit comprises a Gaussian fuzzy comparison image module which is used for carrying out Gaussian fuzzy comparison on claim material images, carrying out vertical multi-time fuzzy comparison on the same image, obtaining image indexes of the image to be detected and the fuzzy image which are obtained through the comprehensive algorithm calculation of the claim material image quality scoring model, determining the definition value according to the variation condition of the indexes, wherein the image is clearer when the variation of the calculation result is smaller, and is fuzzy when the variation of the calculation result is smaller.
In some embodiments, the no-reference image quality index analysis unit further includes an algorithm index normalization module, a weighting calculation module, and an image sharpness determination result processing module, where the algorithm index normalization module performs normalization processing of various final image sharpness algorithm indexes according to a change condition of indexes thereof by vertical comparison of image quality of the gaussian fuzzy comparison image module; the weighting calculation module obtains the variation value of each basic algorithm in the image quality scoring unit, calculates the final image clear value in a weighting mode, and obtains corresponding weight parameters through the logistic regression based on machine learning and the SVM algorithm; the image definition judgment result processing module acquires the comprehensive definition scores of all the images and sends the comprehensive definition scores to the claim material image quality control client subsystem; the image definition judgment result processing module also classifies the input images into clear or non-clear pictures according to the threshold value of the definition level, and simultaneously establishes an image definition score database to record the definition score of each image.
In some embodiments, the default blurring number of times in which the claim material image quality control service subsystem performs blurring on the claim material image to be tested a plurality of times using gaussian blurring is preset to 4-5 times; when the threshold determination is performed, the threshold of the definition level is preset and divided into 2, 3 and 5 levels.
Drawings
Fig. 1 is a flowchart of the steps of a method for controlling the image quality of claim materials based on gaussian blur according to an embodiment of the present invention.
Fig. 2 is various index algorithms used when the image definition evaluation index algorithm is comprehensively processed in the method for controlling the image quality of the claims material based on gaussian blur according to the above embodiment of the present invention.
Detailed Description
The following description is presented to disclose the invention so as to enable any person skilled in the art to practice the invention. The preferred embodiments in the following description are given by way of example only, and other obvious variations will occur to those skilled in the art. The basic principles of the invention, as defined in the following description, may be applied to other embodiments, variations, modifications, equivalents, and other technical solutions without departing from the spirit and scope of the invention.
It is understood that the terms "a" and "an" should be interpreted as meaning that a number of one element or element is one in one embodiment, while a number of other elements is one in another embodiment, and the terms "a" and "an" should not be interpreted as limiting the number.
The present invention relates to a computer program. Fig. 1 is a flow chart of a method for controlling image quality of claim material based on gaussian blur according to the present invention, which illustrates a solution for controlling or processing an external object or an internal object of a computer by executing a computer program prepared according to the above flow on the basis of a computer program processing flow to solve the problems proposed by the present invention. By the method for controlling the image quality of the claim material based on the Gaussian blur, the accuracy of image definition judgment in the field can be improved by using a computer system aiming at the quality analysis of a non-reference image of the claim material, business personnel can be assisted to follow the process better, the efficiency is improved, the definition judgment and subjective and artificial judgment can be optimized by combining a plurality of algorithms aiming at the image definition, and the accuracy and the actual application efficiency of clear image screening can be greatly improved aiming at the field of the claim material. It should be understood that the term "computer" as used herein refers not only to desktop computers, notebook computers, tablet computers, etc., but also includes other intelligent electronic devices capable of operating according to programs and processing data.
Specifically, the method for controlling the image quality of the claim material based on the Gaussian blur comprises the following steps:
acquiring an image of the input claim material;
establishing a claim material image quality scoring model, comprehensively processing an image definition evaluation index algorithm, and executing computer vision image quality scoring of the claim material image, wherein the image definition evaluation index algorithm is selected from one or more of a Brenner gradient algorithm, a Laplacian gradient algorithm, an SMD gray variance algorithm, an SMD2 gray variance product algorithm, an energy gradient algorithm, a Vollant function algorithm, an entropy function algorithm, an EAV point sharpness algorithm and NRSS gradient structure similarity;
and acquiring an image quality grading result of the claim material image quality grading model, and executing non-reference image quality index analysis based on computer vision.
Specifically, in the step of analyzing no-reference image quality index based on computer vision, the method further comprises the following steps:
performing Gaussian fuzzy comparison on claim material images, performing vertical fuzzy comparison on the same image for multiple times, obtaining image indexes of the image to be detected and the fuzzy image obtained through the comprehensive algorithm calculation of the claim material image quality scoring model, determining the definition value according to the change conditions of the indexes, wherein the image is clearer when the change of the calculation result is smaller, and is fuzzy when the change of the calculation result is smaller;
executing normalization processing of various algorithm indexes of final image definition;
calculating the final image quality score in a weighted mode;
and outputting a clear judgment result of the claim material image.
The method for controlling the image quality of the claim material based on the Gaussian blur further comprises the following steps:
acquiring comprehensive definition scores of all input claims material images;
classifying the input images into clear or non-clear pictures according to the threshold value of the definition level, simultaneously establishing an image definition database, and recording the definition score of each image.
As shown in fig. 2, in a specific embodiment, the algorithm for evaluating the sharpness of the synthetically processed image in the model for scoring the image quality of the claim material includes, but is not limited to Brenner gradient algorithm, Laplacian gradient algorithm, SMD grayscale variance algorithm, SMD2 grayscale variance product algorithm, energy gradient algorithm, Vollath function algorithm, entropy function algorithm, EAV point sharpness algorithm, NRSS gradient structure similarity. Because the image quality is affected by various factors, it mainly includes: brightness, contrast, hue, boundary, noise, blur, etc., so that combining the above algorithms can reduce the difference between different environments and subjective judgments.
In the quality evaluation of the non-reference image, the definition of the image is an important index for measuring the quality of the image, and the image can better correspond to the subjective feeling of people, and the image is not high in definition and shows the blurring of the image. In a specific embodiment, the definition of the image of the claim material can be improved in the step of analyzing the no-reference image quality index based on computer vision.
Because of the uncertainty and non-normalization of the reference image, different images are difficult to have a unified standard of normalization, and therefore, the definition index cannot be obtained through the lateral comparison of multiple images. In the specific embodiment of the method for controlling the image quality of the claim material based on the Gaussian blur, the vertical secondary or even multiple times of blur processing on the same image is adopted to obtain image indexes of an image to be detected and a blurred image which are obtained through calculation of various algorithms, and the definition value is determined according to the change conditions of the indexes: the image is clearer when the change of the calculation result is smaller, otherwise, the image is fuzzy, so that the uncertainty and non-normalization of the non-reference image can be reduced, different images have standard unification, and the definition index is obtained.
In the image of the claim material, the photograph is mostly taken in a static manner. The shooting of staff such as a claimant is removed, and the shooting is uploaded for the customer in a remote mode, so that strict specifications are difficult to define for angles, directions, sizes and the like of the staff. According to the image quality control method of the claim material based on the Gaussian blur, after the required claim image material is collected, the image to be detected can be blurred for two times or more times by adopting the Gaussian blur, and the blurring times can be determined according to actual conditions. The more the blurring times are, the more accurate the definition change value obtained theoretically is, but the higher the blurring times are, the more blurred the generated degraded image is, and the reference value of the degraded image is lower and lower. Finally, according to technical tests, the default fuzzy times are set to be 4-5 times, and can be changed according to actual requirements.
In a specific embodiment of the method for controlling the image quality of the claim material based on the gaussian blur, when the threshold is determined, the normalization processing of various algorithm indexes of the final image definition can be performed according to the change condition of the indexes by the vertical comparison of the image quality. The threshold value can be divided into 2, 3 and 5 levels aiming at the problem of whether the image is clear or not, and the specific grading can be made according to the actual requirement. For the judgment of the threshold, the obtained change value of each basic algorithm can be used for calculating a final image clear value in a weighting mode, and the weight value can be used for obtaining a corresponding weight parameter through a machine learning-based logistic regression and a Support Vector Machine (SVM) algorithm.
Further, by the method for controlling the image quality of the claim material based on the Gaussian blur, comprehensive definition scores of all input images can be obtained, the input images can be classified into clear or non-clear pictures according to the threshold value of the definition level, and the definition score of each image is recorded in the background.
According to the method for controlling the image quality of the claim material based on the Gaussian blur, a universal evaluation method can be adopted in the non-reference image quality evaluation, and meanwhile, in an actual application system, the problems of difference of application scenes, difference of customer requirements and complexity of image definition judgment factors can be solved. According to the method for controlling the image quality of the claim material based on Gaussian blur, disclosed by the invention, the definition judgment, the subjective judgment and the manual judgment are carried out to achieve the optimal solution by combining a plurality of algorithms aiming at the definition of the image and aiming at the field of the claim material, so that the accuracy rate and the practical application efficiency of clear image screening are greatly improved.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects.
Those skilled in the art will understand that the method for controlling image quality of claim material based on gaussian blur according to the present invention can be implemented by hardware, software, or a combination of hardware and software. The method for controlling the image quality of claim material based on gaussian blur of the present invention can be realized in a centralized manner in at least one computer system or in a decentralized manner by different parts distributed in several interconnected computer systems. Any kind of computer system or other apparatus adapted for carrying out the methods described herein is suited. A typical combination of hardware and software could be a general purpose computer system with a computer program that, when being loaded and executed, controls the computer system such that it carries out the methods described herein.
The present invention can be embedded in a computer program product, which comprises all the features enabling the implementation of the methods described herein. The computer program product is embodied in one or more computer-readable storage media having computer-readable program code embodied therein. According to another aspect of the invention, there is also provided a computer-readable storage medium having stored thereon a computer program capable, when executed by a processor, of performing the steps of the method of the invention. Computer storage media is media in computer memory for storage of some discrete physical quantity. Computer storage media includes, but is not limited to, semiconductors, magnetic disk storage, magnetic cores, magnetic drums, magnetic tape, laser disks, and the like. It will be appreciated by persons skilled in the art that computer storage media are not limited by the foregoing examples, which are intended to be illustrative only and not limiting of the invention.
According to another aspect of the present invention, there is also provided a gaussian blur based claim material image quality control apparatus, including: a software application, a memory for storing the software application, and a processor for executing the software application. Each program of the software application program can correspondingly execute the steps in the method for controlling the image quality of the claim material based on the Gaussian blur.
A typical combination of hardware and software could be a general purpose computer system with a computer program that, when being loaded and executed, controls the computer system such that it carries out the method for image quality control of claim material based on gaussian blur as disclosed herein.
It will be understood by those skilled in the art that the image quality control device based on the gaussian blur may be embodied as a desktop computer, a notebook computer, a mobile intelligent device, etc., but the foregoing is merely an example, and other intelligent analysis devices loaded with the software application of the present invention are also included.
Corresponding to the embodiment of the method, according to another aspect of the invention, a system for controlling the quality of the image of the claim material based on the gaussian blur is also provided, and the system for controlling the quality of the image of the claim material based on the gaussian blur is an application of the method for controlling the quality of the image of the claim material based on the gaussian blur in the improvement of computer programs.
Specifically, the system for controlling the image quality of the claim material based on the gaussian blur comprises a customer subsystem for controlling the image quality of the claim material and a service subsystem for controlling the image quality of the claim material, wherein the customer subsystem for controlling the image quality of the claim material is provided with a human-computer interaction unit for inputting an image of the claim material, and the service subsystem for controlling the image quality of the claim material is configured to: acquiring an image of the input claim material; establishing a claim material image quality scoring model, comprehensively processing an image definition evaluation index algorithm, and executing computer vision image quality scoring of the claim material image; obtaining an image quality grading result of the claim material image quality grading model, performing Gaussian fuzzy comparison on the claim material image, and performing non-reference image quality index analysis based on computer vision; the claim material image quality control service subsystem outputs a clear judgment result of a claim material image to the claim material image quality control client subsystem.
More specifically, the claim material image quality control service subsystem comprises an image quality scoring unit and a no-reference image quality index analysis unit, wherein the image quality scoring unit establishes a claim material image quality scoring model and comprehensively processes an image definition evaluation index algorithm, wherein the algorithm is selected from one or more of Brenner gradient algorithm, Laplacian gradient algorithm, SMD grayscale variance algorithm, SMD2 grayscale variance product algorithm, energy gradient algorithm, Vollath function algorithm, entropy function algorithm, EAV point sharpness algorithm and NRSS gradient structure similarity; the non-reference image quality index analysis unit comprises a Gaussian fuzzy comparison image module which is used for carrying out Gaussian fuzzy comparison on claim material images, carrying out vertical multi-time fuzzy comparison on the same image, obtaining image indexes of the image to be detected and the fuzzy image which are obtained through the comprehensive algorithm calculation of the claim material image quality scoring model, determining the definition value according to the variation condition of the indexes, wherein the image is clearer when the variation of the calculation result is smaller, and is fuzzy when the variation of the calculation result is smaller.
More specifically, the no-reference image quality index analysis unit further comprises an algorithm index normalization module, a weighting calculation module and an image sharpness judgment result processing module, wherein the algorithm index normalization module performs normalization processing of various final image sharpness algorithm indexes according to the change condition of indexes of the image quality of the Gaussian blur comparison image module through vertical comparison of the image quality of the Gaussian blur comparison image module; the weighting calculation module obtains the variation value of each basic algorithm in the image quality scoring unit, calculates the final image clear value in a weighting mode, and obtains corresponding weight parameters through the logistic regression based on machine learning and the SVM algorithm; the image definition judgment result processing module acquires the comprehensive definition scores of all the images and sends the comprehensive definition scores to the claim material image quality control client subsystem; the image definition judgment result processing module also classifies the input images into clear or non-clear pictures according to the threshold value of the definition level, and simultaneously establishes an image definition score database to record the definition score of each image.
More specifically, in a specific embodiment, the default blurring number of times of performing multiple blurring on the image of the claim material to be detected by adopting gaussian blurring is preset to be 4-5 times; when the no-reference image quality index analysis unit performs threshold determination, the threshold of the sharpness level is preset to be divided into 2, 3, 5 levels.
It will be appreciated by those skilled in the art that the present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, systems and computer program products according to the invention. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart and/or block diagram block or blocks.
It will be appreciated by persons skilled in the art that the embodiments of the invention described above and shown in the drawings are given by way of example only and are not limiting of the invention. The objects of the invention have been fully and effectively accomplished. The functional and structural principles of the present invention have been shown and described in the examples, and any variations or modifications of the embodiments of the present invention may be made without departing from the principles.
Claims (10)
1. The method for controlling the image quality of the claim material based on the Gaussian blur is characterized by comprising the following steps of:
acquiring an image of the input claim material;
establishing a claim material image quality scoring model, comprehensively processing an image definition evaluation index algorithm, and executing computer vision image quality scoring of the claim material image; and
acquiring an image quality grading result of a claim material image of the claim material image quality grading model, performing Gaussian fuzzy comparison on the claim material image, and performing non-reference image quality index analysis based on computer vision;
and outputting a clear judgment result of the claim material image.
2. The method for controlling image quality of claims material based on gaussian blur according to claim 1, wherein the step of analyzing quality index of reference-free image based on computer vision further comprises the following steps: and performing Gaussian fuzzy comparison on the claim material image, performing vertical fuzzy comparison on the same image for multiple times, obtaining image indexes of the image to be detected and the fuzzy image obtained by calculation through a comprehensive algorithm of a claim material image quality grading model, determining the definition value according to the change conditions of the indexes, wherein the image is clearer when the change of the calculation result is smaller, and is fuzzy when the change of the calculation result is smaller.
3. The method for claim material image quality control based on gaussian blur according to claim 2, wherein the method for claim material image quality control based on gaussian blur further comprises the steps of: executing normalization processing of various algorithm indexes of final image definition; and calculating a final image quality score in a weighted manner.
4. The method for claim material image quality control based on gaussian blur according to claim 2, wherein the method for claim material image quality control based on gaussian blur further comprises the steps of:
acquiring comprehensive definition scores of all input claims material images; and
classifying the input images into clear or non-clear pictures according to a threshold value of a definition level, establishing an image definition database and recording the definition score of each image;
default blurring times for executing multiple times of blurring on the claim material image to be detected by adopting Gaussian blurring is preset to be 4-5 times;
wherein the threshold of the definition level is preset and divided into 2, 3 and 5 levels;
when the threshold is judged, the obtained change value of each basic algorithm calculates the final image clear value in a weighting mode, and the weight value obtains corresponding weight parameter by executing the machine learning-based logistic regression and SVM algorithm.
5. The method for image quality control of claims material based on gaussian blur according to any one of claims 1 to 4, wherein the image sharpness evaluation index algorithm is selected from one or more of Brenner gradient algorithm, Laplacian gradient algorithm, SMD grayscale variance algorithm, SMD2 grayscale variance product algorithm, energy gradient algorithm, Vollath function algorithm, entropy function algorithm, EAV point sharpness algorithm, NRSS gradient structure similarity.
6. A claim material image quality control apparatus based on gaussian blur, comprising:
a memory for storing a software application,
a processor for executing the software applications, wherein each program of the software applications correspondingly executes the steps of the method for controlling the image quality of the claims material based on the gaussian blur according to claims 1 to 5.
7. The claim material image quality control system based on Gaussian blur is characterized by comprising a claim material image quality control client subsystem and a claim material image quality control service subsystem, wherein the claim material image quality control client subsystem is provided with a human-computer interaction unit for inputting images of claim materials, and the claim material image quality control service subsystem is configured to: acquiring an image of the input claim material; establishing a claim material image quality scoring model, comprehensively processing an image definition evaluation index algorithm, and executing computer vision image quality scoring of the claim material image; obtaining an image quality grading result of the claim material image quality grading model, performing Gaussian fuzzy comparison on the claim material image, and performing non-reference image quality index analysis based on computer vision; the claim material image quality control service subsystem outputs a clear judgment result of a claim material image to the claim material image quality control client subsystem.
8. The gaussian blur based claimed material image quality control system as claimed in claim 7, wherein said claimed material image quality control service subsystem comprises an image quality scoring unit establishing a claimed material image quality scoring model, a comprehensive processing image sharpness evaluation index algorithm, wherein the algorithm is selected from one or more of Brenner gradient algorithm, Laplacian gradient algorithm, SMD grayscale variance algorithm, SMD2 grayscale variance product algorithm, energy gradient algorithm, Vollath function algorithm, entropy function algorithm, EAV point sharpness algorithm, NRSS gradient structure similarity, and a no reference image quality index analysis unit; the non-reference image quality index analysis unit comprises a Gaussian fuzzy comparison image module which is used for carrying out Gaussian fuzzy comparison on claim material images, carrying out vertical multi-time fuzzy comparison on the same image, obtaining image indexes of the image to be detected and the fuzzy image which are obtained through the comprehensive algorithm calculation of the claim material image quality scoring model, determining the definition value according to the variation condition of the indexes, wherein the image is clearer when the variation of the calculation result is smaller, and is fuzzy when the variation of the calculation result is smaller.
9. The image quality control system for claims material based on gaussian blur according to claim 8, wherein the no-reference image quality index analysis unit further comprises an algorithm index normalization module, a weighting calculation module and an image sharpness judgment result processing module, wherein the algorithm index normalization module performs normalization processing of various types of algorithm indexes of final image sharpness according to the change condition of the indexes thereof by vertical comparison of the image quality of the gaussian blur comparison image module; the weighting calculation module obtains the variation value of each basic algorithm in the image quality scoring unit, calculates the final image clear value in a weighting mode, and obtains corresponding weight parameters through the logistic regression based on machine learning and the SVM algorithm; the image definition judgment result processing module acquires the comprehensive definition scores of all the images and sends the comprehensive definition scores to the claim material image quality control client subsystem; the image definition judgment result processing module also classifies the input images into clear or non-clear pictures according to the threshold value of the definition level, and simultaneously establishes an image definition score database to record the definition score of each image.
10. The gaussian blur-based claim material image quality control system as claimed in any one of claims 7 to 9, wherein a default number of blurrings by the claim material image quality control service subsystem to perform multiple blurrings on a claim material image to be tested using gaussian blur is preset to 4-5 times; when the threshold determination is performed, the threshold of the definition level is preset and divided into 2, 3 and 5 levels.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111053937.1A CN113850764A (en) | 2021-09-09 | 2021-09-09 | Method, system and equipment for controlling image quality of claim material based on Gaussian blur |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111053937.1A CN113850764A (en) | 2021-09-09 | 2021-09-09 | Method, system and equipment for controlling image quality of claim material based on Gaussian blur |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113850764A true CN113850764A (en) | 2021-12-28 |
Family
ID=78973629
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111053937.1A Pending CN113850764A (en) | 2021-09-09 | 2021-09-09 | Method, system and equipment for controlling image quality of claim material based on Gaussian blur |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113850764A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101996406A (en) * | 2010-11-03 | 2011-03-30 | 中国科学院光电技术研究所 | No-reference structure sharpness image quality assessment method |
| CN107369134A (en) * | 2017-06-12 | 2017-11-21 | 上海斐讯数据通信技术有限公司 | A kind of image recovery method of blurred picture |
| CN110738656A (en) * | 2019-10-28 | 2020-01-31 | 公安部交通管理科学研究所 | Method for evaluating definition of certificate photos, storage medium and processor |
| CN112001904A (en) * | 2020-08-21 | 2020-11-27 | 珀乐(北京)信息科技有限公司 | Comprehensive evaluation module and evaluation method for remote sensing image quality definition |
| CN112330469A (en) * | 2020-11-04 | 2021-02-05 | 阳光保险集团股份有限公司 | Pre-examination method and device for medical insurance claim settlement materials |
-
2021
- 2021-09-09 CN CN202111053937.1A patent/CN113850764A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101996406A (en) * | 2010-11-03 | 2011-03-30 | 中国科学院光电技术研究所 | No-reference structure sharpness image quality assessment method |
| CN107369134A (en) * | 2017-06-12 | 2017-11-21 | 上海斐讯数据通信技术有限公司 | A kind of image recovery method of blurred picture |
| CN110738656A (en) * | 2019-10-28 | 2020-01-31 | 公安部交通管理科学研究所 | Method for evaluating definition of certificate photos, storage medium and processor |
| CN112001904A (en) * | 2020-08-21 | 2020-11-27 | 珀乐(北京)信息科技有限公司 | Comprehensive evaluation module and evaluation method for remote sensing image quality definition |
| CN112330469A (en) * | 2020-11-04 | 2021-02-05 | 阳光保险集团股份有限公司 | Pre-examination method and device for medical insurance claim settlement materials |
Non-Patent Citations (2)
| Title |
|---|
| FUNNY AI: "模糊图像检测-无参考图像的清晰度评价", pages 6, Retrieved from the Internet <URL:https://zhuanlan.zhihu.com/p/97024018> * |
| 王冠军;吴志勇;云海姣;梁敏华;杨华;: "结合图像二次模糊范围和奇异值分解的无参考模糊图像质量评价", 计算机辅助设计与图形学学报, no. 04, 15 April 2016 (2016-04-15), pages 124 - 132 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108229321B (en) | Face recognition model, and training method, device, apparatus, program, and medium therefor | |
| CN109284673B (en) | Object tracking method and device, electronic equipment and storage medium | |
| JP2015087903A (en) | Apparatus and method for information processing | |
| KR20200087297A (en) | Defect inspection method and apparatus using image segmentation based on artificial neural network | |
| WO2021153633A1 (en) | Metal structure phase classification method, metal structure phase classification device, metal structure phase learning method, metal structure phase learning device, material property prediction method for metal material, and material property prediction device for metal material | |
| KR102141302B1 (en) | Object detection method based 0n deep learning regression model and image processing apparatus | |
| CN111932510A (en) | Method and device for determining image definition | |
| CN108961239A (en) | Continuous casting billet quality detection method, device, electronic equipment and storage medium | |
| CN117314826B (en) | A performance detection method for display screen | |
| CN116740728B (en) | Dynamic acquisition method and system for wafer code reader | |
| CN111784658A (en) | Quality analysis method and system for face image | |
| CN112215831B (en) | Method and system for evaluating quality of face image | |
| CN111259926A (en) | Meat freshness detection method and device, computing equipment and storage medium | |
| CN118552550B (en) | Wear detection method, system and equipment for multi-caliper disc type elevator brake | |
| CN108257117B (en) | Image exposure evaluation method and device | |
| CN118644758B (en) | Nuclear reactor pressure vessel inner surface defect identification method and system | |
| CN113850764A (en) | Method, system and equipment for controlling image quality of claim material based on Gaussian blur | |
| CN117474916A (en) | Image detection method, electronic equipment and storage medium | |
| CN113870210B (en) | Image quality assessment method, device, equipment and storage medium | |
| Li et al. | Sharpness and brightness quality assessment of face images for recognition | |
| CN112949431A (en) | Video tampering detection method and system, and storage medium | |
| CN117522966B (en) | Physical quantity position information management method and system | |
| CN117274084B (en) | Real-time self-adaptive sharpening method for expressway video monitoring | |
| RU2774624C1 (en) | Method and system for determining synthetic changes of faces in video | |
| CN113420716B (en) | Illegal behavior identification and early warning method based on improved Yolov3 algorithm |
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 |