CN115392662A - Quality identification grading management system and method based on image data - Google Patents

Quality identification grading management system and method based on image data Download PDF

Info

Publication number
CN115392662A
CN115392662A CN202210967037.6A CN202210967037A CN115392662A CN 115392662 A CN115392662 A CN 115392662A CN 202210967037 A CN202210967037 A CN 202210967037A CN 115392662 A CN115392662 A CN 115392662A
Authority
CN
China
Prior art keywords
product
evaluation
image
module
quality
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.)
Withdrawn
Application number
CN202210967037.6A
Other languages
Chinese (zh)
Inventor
李元乔
杨沛泉
王温馨
左腊梅
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hunan Zuoyuan Information Technology Co ltd
Original Assignee
Hunan Zuoyuan Information Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hunan Zuoyuan Information Technology Co ltd filed Critical Hunan Zuoyuan Information Technology Co ltd
Priority to CN202210967037.6A priority Critical patent/CN115392662A/en
Publication of CN115392662A publication Critical patent/CN115392662A/en
Withdrawn legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06395—Quality analysis or management
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/205—Parsing
    • G06F40/216—Parsing using statistical methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/04—Inference or reasoning models
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/018—Certifying business or products
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Development Economics (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Educational Administration (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Tourism & Hospitality (AREA)
  • Computational Linguistics (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Evolutionary Computation (AREA)
  • Data Mining & Analysis (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Multimedia (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Probability & Statistics with Applications (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • General Health & Medical Sciences (AREA)
  • Image Analysis (AREA)

Abstract

The invention belongs to the technical field of quality identification, and discloses a quality identification grading management system and method based on image data, wherein the quality identification grading management system based on the image data comprises the following components: the system comprises a product image acquisition module, a main control module, an image enhancement module, an image feature extraction module, a product traceability module, a quality identification module, a quality evaluation module, a grade determination module and a display module. According to the invention, the darker part in the product image becomes brighter through the image enhancement module, more product image details are highlighted, the layering sense of the product image is increased, and the product image imaging effect is effectively improved; meanwhile, the quality evaluation module is used for matching the evaluation of the user on the target product with the quality index of the target product to obtain the specific opinion under the evaluation index, so that the problem of the quality of the target product can be traced, and the evaluation result of the quality of the target product is accurate.

Description

Quality identification grading management system and method based on image data
Technical Field
The invention belongs to the technical field of quality identification, and particularly relates to a quality identification grading management system and method based on image data.
Background
Product quality refers to the sum of the features and characteristics of a product that meet regulatory and potential needs. Any product is manufactured to meet the user's needs. For product quality, whether simple or complex, product quality characteristics or features should be used for description. The quality characteristics of the product are different according to the characteristics of the product, expressed parameters and indexes are also various, and the quality characteristics reflecting the use requirements of users generally have six aspects, namely performance, service life (namely durability), reliability and maintainability, safety, adaptability and economy. Quality of a product (Quality) refers to the necessary information disclosure throughout the course of planning, designing, manufacturing, detecting, metering, transporting, storing, selling, after-sales service, ecological recycling, etc., of a product by an enterprise according to specific standards in the commodity economy category. However, the existing quality identification grading management system and method based on image data adopts a product image enhancement method of histogram equalization processing, which has poor enhancement effect and is difficult to adapt to complex illumination change; meanwhile, the factor of a consumer is not fully considered when the evaluation data is obtained in the prior art, so that the data source is not wide enough, and the target product quality evaluation method provided by the prior art has the defect of inaccurate evaluation result.
Through the above analysis, the problems and defects of the prior art are as follows:
(1) The existing quality identification grading management system and method based on image data adopts a product image enhancement method of histogram equalization processing, which has poor enhancement effect and is difficult to adapt to complex illumination change.
(2) The factor of a consumer is not fully considered when the evaluation data is obtained in the prior art, so that the data source is not wide enough, and therefore, the target product quality evaluation method provided by the prior art has the defect of inaccurate evaluation result.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a quality identification grading management system and method based on image data.
The invention is realized in this way, a quality discrimination ranking management system based on image data includes:
the system comprises a product image acquisition module, a main control module, an image enhancement module, an image feature extraction module, a product traceability module, a quality identification module, a quality evaluation module, a grade determination module and a display module;
the product image acquisition module is connected with the main control module and is used for acquiring product image data through the camera equipment;
the main control module is connected with the product image acquisition module, the image enhancement module, the image feature extraction module, the product traceability module, the quality identification module, the quality evaluation module, the grade determination module and the display module and is used for controlling the normal work of each module;
the image enhancement module is connected with the main control module and is used for enhancing the acquired image through an image enhancement program;
the image feature extraction module is connected with the main control module and used for extracting the image features of the product through an extraction program;
the product source tracing module is connected with the main control module and is used for tracing the source of the product through a source tracing program;
the quality identification module is connected with the main control module and is used for identifying the product quality through an identification program;
the quality evaluation module is connected with the main control module and used for evaluating the product quality through an evaluation program;
the grade determining module is connected with the main control module and is used for determining the quality grade of the product;
and the display module is connected with the main control module and is used for displaying the product image, the product traceability result, the quality identification result, the quality evaluation result and the product grade.
A quality discrimination hierarchical management method based on image data comprises the following steps:
acquiring product image data by utilizing camera equipment through a product image acquisition module;
secondly, the main control module utilizes an image enhancement program to enhance the collected image through an image enhancement module;
extracting the image characteristics of the product by using an extraction program through an image characteristic extraction module; tracing the source of the product by using a tracing program through a product tracing module; identifying the product quality by using an identification program through a quality identification module;
evaluating the product quality by utilizing an evaluation program through a quality evaluation module; determining the quality grade of the product through a grade determining module;
and fifthly, displaying the product image, the product traceability result, the quality identification result, the quality evaluation result and the product grade through a display module.
Further, the image enhancement module enhancement method is as follows:
(1) Collecting a product image, and denoising the product image; counting the number of pixel points corresponding to each gray scale in the product image, thereby determining the occurrence frequency of each gray scale; solving the Lp norm of the frequency of each gray scale, thereby obtaining the Lp norm of the frequency of each gray scale;
(2) In the process of performing histogram equalization processing on a product image, when the number of times of occurrence of a certain gray scale increases, the rate of increase decreases with the increase of the number of times of occurrence of the gray scale, and histogram equalization is performed on the product image based on the Lp norm of the number of times of occurrence of each gray scale so as to realize enhancement processing on the product image;
the method for denoising the product image comprises the following steps:
estimating the noise intensity of the product image according to the pixel value of the pixel in the product image;
carrying out primary denoising treatment on the obtained product image to be processed to obtain a primary denoised product image;
calculating a residual quantity corresponding to a central pixel of each unit region on the product image to be processed, wherein the residual quantity is an absolute value of a difference value between a gray value of the product image to be processed and a gray value of the preliminary denoising product image;
calculating a weight matrix corresponding to each unit region by using the residual quantity, and performing non-local mean value calculation on the product image to be processed according to the weight matrix to realize the denoising processing of the product image to be processed, wherein the process of calculating the weight matrix comprises the following steps:
selecting any unit area on the product image to be processed, and determining a related area of the unit area on the product image to be processed;
calculating a weight value corresponding to each of the associated unit regions according to the distance value between the associated unit region and the residual amount to obtain the weight matrix, wherein the weight value corresponding to any associated unit region in the associated region is calculated by using the following formula:
w(n，m)＝e (-(d(n，m)+residuals(n，m))/h) ，
wherein w (n, m) is a weight value corresponding to any associated unit region (n, m), d (n, m) is a distance value between any associated unit region and any unit region, residual (n, m) is a residual amount corresponding to a central pixel of any unit region, h is a preset control coefficient,
calculating a distance value between said any associated unit area (n, m) and said any unit area (i, j) using the following formula:
Figure BDA0003793594860000041
Figure BDA0003793594860000042
where 2r +1 is the side length of any one of the associated unit areas, and T, k and T are intermediate values.
The Lp norm is obtained by the following method:
storing Lp norms corresponding to integers smaller than a first threshold in advance;
acquiring the Lp norm of the occurrence times of each gray scale of the product image by searching the Lp norm corresponding to the stored integer;
the storage process comprises: storing Lp norms corresponding to integers every interval of a second threshold number;
and the first threshold value is correspondingly set according to the resolution of the product image.
Further, the obtaining the Lp norm of the number of times of occurrence of each gray scale to obtain the Lp norm of the number of times of occurrence of each gray scale includes: when the occurrence frequency of the gray scale is greater than a first threshold, the Lp norm value of the occurrence frequency of the gray scale is the corresponding Lp norm value when the occurrence frequency of the gray scale is the first threshold;
the value range of p of the Lp norm is that p is more than or equal to 0 and less than 1;
carrying out histogram equalization on pixel points in the product image through the following formula:
Figure BDA0003793594860000051
wherein I (x, y) is a brightness value of a pixel point at a (x, y) position in the product image, j is an index value of a gray scale in the product image, M is a total number of gray scales in the product image, H × j is an Lp norm value of the number of times of occurrence of the gray scale j in the product image, and ψ (I (x, y)) is a brightness value of the pixel point at the (x, y) position after histogram equalization processing;
the product image is in a YUV format, and the brightness value of the pixel point is the Y component of the pixel point.
Further, the quality evaluation module evaluation method is as follows:
1) Constructing a product database; acquiring target product data and client comment data of a user on a target product; storing the acquired data into a product database;
2) Preprocessing the client comment data to obtain historical client comment data; acquiring an initial evaluation index based on the target product data; acquiring an index seed word based on the historical client comment data and the initial evaluation index; obtaining a word vector based on the historical customer comment data;
3) Obtaining an evaluation index based on the word vector and the index seed word; acquiring evaluation data based on the word vector and a BLSTM-CRFs model constructed in advance; matching the evaluation index with the evaluation data, and scoring the matched evaluation data;
4) Processing the scoring result based on an evidence reasoning method to obtain evaluation grade reliability distribution; converting the evaluation grade reliability distribution into a utility value; obtaining the evaluation of the quality of the target product based on the utility value;
the evidence-based reasoning method for processing the scoring result comprises the following steps:
presetting the evaluation grade of the evaluation index; converting the scoring result into the reliability of the evaluation grade of the corresponding evaluation index; giving a weight to the evaluation index, and acquiring basic probability distribution of the evaluation index based on the reliability and the weight;
based on an evidence reasoning method, combining the basic probability distribution, fusing the evaluation data to obtain evaluation grade reliability distribution;
based on an evidence reasoning method, combining the basic probability distribution and fusing the evaluation data to obtain the evaluation grade reliability distribution specifically comprises the following steps:
and selecting ER algorithm in an evidence reasoning method to fuse the evaluation data, wherein a fusion formula is as follows:
m n，I(k+1) ＝K I(1+1) (m n，I(k) m n，k+1 +m n，I(k) m H，k+1 +m H，I(k) m n，k+1 ) n＝1，2，…N
m H，I(k+1) ＝K I(k+1) m H，I(k) m H，k+1
Figure BDA0003793594860000061
wherein:
m n，I(k) indicates that the first k indexes are distributed to the level H after being fused n The probability of (d);
m H，I(k) representing the probability of the first k indexes being distributed to the complete set H after fusion;
n represents the number of evaluation levels, L represents the number of evaluation indexes,
k represents a normalization coefficient;
m obtained n，I(L) Reconversion to confidence distribution on evaluation level:
Figure BDA0003793594860000062
wherein:
β n indicates that L indexes are evaluated to the grade H after being synthesized n The confidence level of;
the final obtained evaluation grade reliability distribution is as follows:
S(y)＝{(H n ，β n )，n＝1，2，...，N}
wherein:
H n the evaluation scale is shown.
Further, the preprocessing the customer comment data includes:
deleting repeated customer comments, customer comments which are not filled with effective contents and customer comments of which the character length is lower than a preset value;
converting the complex form of the characters in the client comment data into a simplified form;
and performing sentence segmentation, word segmentation and stop word removal on the client comment data.
Further, the method for acquiring the index seed words comprises the following steps:
performing part-of-speech tagging on the historical client comment data; extracting high-frequency nouns and verbs, and filtering out nouns and verbs with frequency lower than a preset value and single words irrelevant to quality attributes; and matching the finally obtained words with the initial evaluation indexes to obtain index seed words corresponding to the initial evaluation indexes.
Further, the method for obtaining the word vector comprises the following steps:
and processing the historical client comment data based on a pre-trained word2vec model to obtain a word vector with a preset dimension.
Further, the method for acquiring the evaluation index includes:
vectorizing the index seed words based on the word vectors to obtain the correlation degrees of all words and words in the historical client comment data and the vectorized index seed words, selecting the first a words with the maximum correlation degrees to expand the index seed words, and removing repeated words to obtain expanded seed words;
processing the expansion seed words based on the KNN to obtain target product quality attribute words;
and calculating the frequency of the target product quality attribute words appearing in the historical customer comment data, and filtering out initial evaluation indexes corresponding to the target product quality attribute words with the frequency lower than a preset value to obtain the evaluation indexes.
Further, the method for acquiring the evaluation data comprises the following steps:
processing the historical client comment data based on a BLSTM-CRFs model and the word vector which are constructed in advance, extracting degree words and emotion words corresponding to the target product quality attribute words, and storing the degree words and the emotion words in a text form to obtain evaluation data;
the method for acquiring the pre-constructed BLSTM-CRFs model comprises the following steps:
building a BLSTM-CRFs model based on a TensorFlow tool; marking the historical customer comment data based on a BIO standard mode to obtain a data set; randomly disordering the sequence of the data set, training the data set for preset times, and selecting a model with the highest accuracy as a pre-constructed BLSTM-CRFs model;
the utility value obtaining method comprises the following steps:
Figure BDA0003793594860000081
wherein:
u(H n ) Indicates evaluation level H n The utility value of (c).
In combination with the technical solutions and the technical problems to be solved, please analyze the advantages and positive effects of the technical solutions to be protected in the present invention from the following aspects:
first, aiming at the technical problems existing in the prior art and the difficulty in solving the problems, the technical problems to be solved by the technical scheme of the present invention are closely combined with results, data and the like in the research and development process, and some creative technical effects are brought after the problems are solved. The specific description is as follows:
according to the method, the Lp norm of the occurrence frequency of each gray scale of the product image is obtained through the image enhancement module, and then the histogram equalization processing is carried out on the product image based on the Lp norm of the occurrence frequency of each gray scale, so that the enhancement processing of the product image is realized, the contrast of the product image can be effectively improved, the integral or local characteristics of the product image are effectively improved, the darker part in the product image becomes brighter, more product image details are highlighted, the layering sense of the product image is increased, and the imaging effect of the product image is effectively improved; meanwhile, the quality evaluation module is used for matching the evaluation of the user on the target product with the quality index of the target product to obtain the specific opinion under the evaluation index, so that the quality problem of the target product can be traced, and the quality evaluation result of the target product is accurate. The enterprise can improve the design of target products, adjust the sales strategy and service and the like in a targeted manner through the evaluation result, so that the profit of the enterprise is improved.
Secondly, considering the technical scheme as a whole or from the perspective of products, the technical effect and advantages of the technical scheme to be protected by the invention are specifically described as follows:
according to the method, the Lp norm of the occurrence frequency of each gray scale of the product image is obtained through the image enhancement module, and then the histogram equalization processing is carried out on the product image based on the Lp norm of the occurrence frequency of each gray scale, so that the enhancement processing of the product image is realized, the contrast of the product image can be effectively improved, the overall or local characteristics of the product image are effectively improved, the darker part in the product image becomes brighter, more product image details are highlighted, the layering sense of the product image is increased, and the imaging effect of the product image is effectively improved; meanwhile, the quality evaluation module is used for matching the evaluation of the user on the target product with the quality index of the target product to obtain the specific opinion under the evaluation index, so that the problem of the quality of the target product can be traced, and the evaluation result of the quality of the target product is accurate. The enterprise can improve the design of target products, adjust the sales strategy and service and the like in a targeted manner through the evaluation result, so that the profit of the enterprise is improved.
Drawings
Fig. 1 is a flowchart of a quality discrimination hierarchical management method based on image data according to an embodiment of the present invention.
Fig. 2 is a block diagram of a hierarchical management system for quality discrimination based on image data according to an embodiment of the present invention.
Fig. 3 is a flowchart of an image enhancement module enhancement method according to an embodiment of the present invention.
Fig. 4 is a flowchart of a quality evaluation module evaluation method according to an embodiment of the present invention.
In fig. 2: 1. a product image acquisition module; 2. a main control module; 3. an image enhancement module; 4. an image feature extraction module; 5. a product tracing module; 6. a quality identification module; 7. a quality evaluation module; 8. a rank determination module; 9. and a display module.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
1. Illustrative embodiments are explained. This section is an explanatory embodiment expanding on the claims so as to fully understand how the present invention is embodied by those skilled in the art.
As shown in fig. 1, the method for quality discrimination hierarchical management based on image data according to the present invention comprises the following steps:
s101, acquiring product image data by utilizing a camera device through a product image acquisition module;
s102, the main control module utilizes an image enhancement program to enhance the collected image through an image enhancement module;
s103, extracting the image characteristics of the product by using an extraction program through an image characteristic extraction module; tracing the source of the product by using a tracing program through a product tracing module; identifying the product quality by using an identification program through a quality identification module;
s104, evaluating the product quality by utilizing an evaluation program through a quality evaluation module; determining the quality grade of the product through a grade determining module;
and S105, displaying the product image, the product source tracing result, the quality identification result, the quality evaluation result and the product grade through a display module.
According to the method, the Lp norm of the occurrence frequency of each gray scale of the product image is obtained through the image enhancement module, and then the histogram equalization processing is carried out on the product image based on the Lp norm of the occurrence frequency of each gray scale, so that the enhancement processing of the product image is realized, the contrast of the product image can be effectively improved, the overall or local characteristics of the product image are effectively improved, the darker part in the product image becomes brighter, more product image details are highlighted, the layering sense of the product image is increased, and the imaging effect of the product image is effectively improved; meanwhile, the quality evaluation module is used for matching the evaluation of the user on the target product with the quality index of the target product to obtain the specific opinion under the evaluation index, so that the problem of the quality of the target product can be traced, and the evaluation result of the quality of the target product is accurate. The enterprise can improve the design of target products, adjust the sales strategy and service and the like in a targeted manner through the evaluation result, so that the profit of the enterprise is improved.
As shown in fig. 2, the system for quality-based quality-discrimination hierarchical management based on image data according to an embodiment of the present invention includes:
the system comprises a product image acquisition module 1, a main control module 2, an image enhancement module 3, an image feature extraction module 4, a product traceability module 5, a quality identification module 6, a quality evaluation module 7, a grade determination module 8 and a display module 9.
The product image acquisition module 1 is connected with the main control module 2 and is used for acquiring product image data through camera equipment;
the main control module 2 is connected with the product image acquisition module 1, the image enhancement module 3, the image feature extraction module 4, the product traceability module 5, the quality identification module 6, the quality evaluation module 7, the grade determination module 8 and the display module 9 and is used for controlling each module to work normally;
the image enhancement module 3 is connected with the main control module 2 and is used for enhancing the acquired image through an image enhancement program;
the image feature extraction module 4 is connected with the main control module 2 and is used for extracting the image features of the product through an extraction program;
the product source tracing module 5 is connected with the main control module 2 and is used for tracing the source of the product through a source tracing program;
the quality identification module 6 is connected with the main control module 2 and is used for identifying the product quality through an identification program;
the quality evaluation module 7 is connected with the main control module 2 and used for evaluating the product quality through an evaluation program;
the grade determining module 8 is connected with the main control module 2 and is used for determining the quality grade of the product;
and the display module 9 is connected with the main control module 2 and is used for displaying the product image, the product tracing result, the quality identification result, the quality evaluation result and the product grade.
According to the method, the Lp norm of the occurrence frequency of each gray scale of the product image is obtained through the image enhancement module, and then the histogram equalization processing is carried out on the product image based on the Lp norm of the occurrence frequency of each gray scale, so that the enhancement processing of the product image is realized, the contrast of the product image can be effectively improved, the overall or local characteristics of the product image are effectively improved, the darker part in the product image becomes brighter, more product image details are highlighted, the layering sense of the product image is increased, and the imaging effect of the product image is effectively improved; meanwhile, the quality evaluation module is used for matching the evaluation of the user on the target product with the quality index of the target product to obtain the specific opinion under the evaluation index, so that the problem of the quality of the target product can be traced, and the evaluation result of the quality of the target product is accurate. The enterprise can improve the design of target products, adjust the sales strategy and service and the like in a targeted manner through the evaluation result, so that the profit of the enterprise is improved.
As shown in fig. 3, the image enhancement module 3 provided by the present invention has the following enhancement method:
s201, collecting a product image and denoising the product image; counting the number of pixel points corresponding to each gray scale in the product image, thereby determining the occurrence frequency of each gray scale; solving the Lp norm of the frequency of each gray scale, thereby obtaining the Lp norm of the frequency of each gray scale;
s202, in the process of performing histogram equalization processing on a product image, when the number of times of occurrence of a certain gray scale increases, the increasing rate decreases along with the increase of the number of times of occurrence of the gray scale, and the histogram equalization is performed on the product image on the basis of the Lp norm of the number of times of occurrence of each gray scale so as to realize enhancement processing on the product image;
according to the method, the Lp norm of the occurrence frequency of each gray scale of the product image is obtained through the image enhancement module, and then the histogram equalization processing is carried out on the product image based on the Lp norm of the occurrence frequency of each gray scale, so that the enhancement processing of the product image is realized.
The method for denoising the product image comprises the following steps:
estimating the noise intensity of the product image according to the pixel value of the pixel in the product image;
carrying out primary denoising treatment on the obtained product image to be processed to obtain a primary denoised product image;
calculating residual quantity of a central pixel corresponding to each unit region on the product image to be processed, wherein the residual quantity is an absolute value of a difference value between a gray value of the product image to be processed and a gray value of the preliminary denoising product image;
calculating a weight matrix corresponding to each unit region by using the residual quantity, and performing non-local mean value calculation on the product image to be processed according to the weight matrix to realize the denoising processing of the product image to be processed, wherein the process of calculating the weight matrix comprises the following steps:
selecting any unit area on the product image to be processed, and determining the associated area of any unit area on the product image to be processed;
calculating a weight value corresponding to each of the associated unit regions according to the distance value between the associated unit region and the residual amount to obtain the weight matrix, wherein the weight value corresponding to any associated unit region in the associated region is calculated by using the following formula:
w(n，m)＝e (-(d(n，m)+residuals(n，m))/h) ，
wherein w (n, m) is a weight value corresponding to any associated unit region (n, m), d (n, m) is a distance value between the associated unit region and the unit region, residual (n, m) is a residual amount corresponding to a central pixel of the unit region, and h is a preset control coefficient,
calculating a distance value between said any associated unit area (n, m) and said any unit area (i, j) using the following formula:
Figure BDA0003793594860000131
Figure BDA0003793594860000132
where 2r +1 is the side length of any one of the associated unit areas, and T, k and T are intermediate values.
The Lp norm is obtained by the following method:
storing Lp norms corresponding to integers smaller than a first threshold in advance;
acquiring the Lp norm of the occurrence times of each gray scale of the product image by searching the Lp norm corresponding to the stored integer;
the storage process comprises: storing Lp norms corresponding to integers every interval of a second threshold number;
and the first threshold value is correspondingly set according to the resolution of the product image.
The invention provides an Lp norm for calculating the occurrence frequency of each gray scale, so as to obtain the Lp norm of the occurrence frequency of each gray scale, comprising the following steps: when the occurrence frequency of the gray scale is greater than a first threshold, the Lp norm value of the occurrence frequency of the gray scale is the corresponding Lp norm value when the occurrence frequency of the gray scale is the first threshold;
the value range of p of the Lp norm is that p is more than or equal to 0 and less than 1;
carrying out histogram equalization on pixel points in the product image through the following formula:
Figure BDA0003793594860000141
wherein I (x, y) is a brightness value of a pixel point at a (x, y) position in the product image, j is an index value of a gray scale in the product image, M is a total number of gray scales in the product image, H × j is an Lp norm value of the number of times of occurrence of the gray scale j in the product image, and ψ (I (x, y)) is a brightness value of the pixel point at the (x, y) position after histogram equalization processing;
the product image is in a YUV format, and the brightness value of the pixel point is the Y component of the pixel point.
As shown in fig. 4, the quality evaluation module 7 provided by the present invention has the following evaluation method:
s301, constructing a product database; acquiring target product data and client comment data of a user on a target product; storing the acquired data into a product database;
s302, preprocessing the client comment data to obtain historical client comment data; acquiring an initial evaluation index based on the target product data; acquiring an index seed word based on the historical client comment data and the initial evaluation index; obtaining a word vector based on the historical customer comment data;
s303, obtaining an evaluation index based on the word vector and the index seed word; acquiring evaluation data based on the word vector and a BLSTM-CRFs model constructed in advance; matching the evaluation index with the evaluation data, and scoring the matched evaluation data;
s304, processing the scoring result based on an evidence reasoning method to obtain evaluation grade reliability distribution; converting the evaluation grade reliability distribution into a utility value; obtaining the evaluation of the quality of the target product based on the utility value;
according to the invention, the quality evaluation module is used for matching the evaluation of the user on the target product with the quality index of the target product to obtain the specific opinion under the evaluation index, so that the problem of the quality of the target product can be traced, and the evaluation result of the quality of the target product is accurate. The enterprise can improve the design of target products, adjust the sales strategy and service and the like in a targeted manner through the evaluation result, so that the profit of the enterprise is improved.
The evidence-based reasoning method for processing the scoring result comprises the following steps:
presetting the evaluation grade of the evaluation index; converting the scoring result into the reliability of the evaluation grade of the corresponding evaluation index; giving a weight to the evaluation index, and acquiring basic probability distribution of the evaluation index based on the reliability and the weight;
based on an evidence reasoning method, combining the basic probability distribution and fusing the evaluation data to obtain evaluation grade reliability distribution;
based on an evidence reasoning method, combining the basic probability distribution and fusing the evaluation data to obtain the evaluation grade reliability distribution specifically comprises the following steps:
and selecting ER algorithm in an evidence reasoning method to fuse the evaluation data, wherein a fusion formula is as follows:
m n，I(k+1) ＝K I(k+1) (m n，I(k) m n，k+1 +m n，I(k) m H，k+1 +m H，I(k) m n，k+1 ) n＝1，2，…，N
m H，I(k+1) ＝K I(k+1) m H，I(k) m H，k+1
Figure BDA0003793594860000151
wherein:
m n，I(k) representing the probability of the first k indexes being assigned to the level Hn after fusion;
m H，I(k) representing the probability of the first k indexes being distributed to the complete set H after fusion;
n represents the number of evaluation levels, L represents the number of evaluation indexes,
k represents a normalization coefficient;
m obtained n，I(L) Reconversion to confidence distribution on evaluation level:
Figure BDA0003793594860000152
wherein:
β n indicates that L indexes are evaluated to the grade H after being synthesized n The confidence level of;
the final obtained evaluation grade reliability distribution is as follows:
S(y)＝{(H n ，β n )，n＝1，2，...，N}
wherein:
H n the evaluation grade is indicated.
The invention provides a method for preprocessing the client comment data, which comprises the following steps:
deleting repeated client comments, client comments which do not fill in effective content and client comments of which the character length is less than a preset value;
converting the complex form of the characters in the client comment data into a simplified form;
and performing sentence segmentation, word segmentation and stop word removal on the client comment data.
The method for acquiring the index seed words comprises the following steps:
performing part-of-speech tagging on the historical client comment data; extracting high-frequency nouns and verbs, and filtering out nouns and verbs with the frequency lower than a preset value and single words irrelevant to quality attributes; and matching the finally obtained words with the initial evaluation indexes to obtain index seed words corresponding to the initial evaluation indexes.
The method for acquiring the word vector provided by the invention comprises the following steps:
and processing the historical client comment data based on a pre-trained word2vec model to obtain a word vector with a preset dimension.
The method for acquiring the evaluation index comprises the following steps:
vectorizing the index seed words based on the word vectors to obtain the correlation degrees of all words and words in the historical client comment data and the vectorized index seed words, selecting the first a words with the maximum correlation degrees to expand the index seed words, and removing repeated words to obtain expanded seed words;
processing the extended seed words based on the KNN to obtain target product quality attribute words;
and calculating the frequency of the target product quality attribute words appearing in the historical customer comment data, and filtering out initial evaluation indexes corresponding to the target product quality attribute words with the frequency lower than a preset value to obtain the evaluation indexes.
The method for acquiring the evaluation data comprises the following steps:
processing the historical client comment data based on a BLSTM-CRFs model and the word vector which are constructed in advance, extracting degree words and emotion words corresponding to the target product quality attribute words, and storing the degree words and the emotion words in a text form to obtain evaluation data;
the method for acquiring the pre-constructed BLSTM-CRFs model comprises the following steps:
building a BLSTM-CRFs model based on a TensorFlow tool; marking the historical customer comment data based on a BIO standard mode to obtain a data set; randomly disturbing the sequence of the data set, training the data set for preset times, and selecting a model with the highest accuracy as a BLSTM-CRFs model which is constructed in advance;
the utility value obtaining method comprises the following steps:
Figure BDA0003793594860000171
wherein:
u(H n ) Indicates evaluation level H n The utility value of (c).
2. Application examples. In order to prove the creativity and the technical value of the technical scheme of the invention, the part is the application example of the technical scheme of the claims on specific products or related technologies.
According to the method, the Lp norm of the occurrence frequency of each gray scale of the product image is obtained through the image enhancement module, and then the histogram equalization processing is carried out on the product image based on the Lp norm of the occurrence frequency of each gray scale, so that the enhancement processing of the product image is realized, the contrast of the product image can be effectively improved, the overall or local characteristics of the product image are effectively improved, the darker part in the product image becomes brighter, more product image details are highlighted, the layering sense of the product image is increased, and the imaging effect of the product image is effectively improved; meanwhile, the quality evaluation module is used for matching the evaluation of the user on the target product with the quality index of the target product to obtain the specific opinion under the evaluation index, so that the quality problem of the target product can be traced, and the quality evaluation result of the target product is accurate. The enterprise can improve the design of target products, adjust the sales strategy and service and the like in a targeted manner through the evaluation result, so that the profit of the enterprise is improved.
It should be noted that the embodiments of the present invention can be realized by hardware, software, or a combination of software and hardware. The hardware portions may be implemented using dedicated logic; the software portions may be stored in a memory and executed by a suitable instruction execution system, such as a microprocessor or specially designed hardware. Those skilled in the art will appreciate that the apparatus and methods described above may be implemented using computer executable instructions and/or embodied in processor control code, such code being provided on a carrier medium such as a disk, CD-or DVD-ROM, programmable memory such as read only memory (firmware), or a data carrier such as an optical or electronic signal carrier, for example. The apparatus of the present invention and its modules may be implemented by hardware circuits such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, or software executed by various types of processors, or a combination of hardware circuits and software, e.g., firmware.
3. Evidence of the relevant effects of the examples. The embodiment of the invention achieves some positive effects in the process of research and development or use, and has great advantages compared with the prior art, and the following contents are described by combining data, diagrams and the like in the test process.
According to the method, the Lp norm of the occurrence frequency of each gray scale of the product image is obtained through the image enhancement module, and then the histogram equalization processing is carried out on the product image based on the Lp norm of the occurrence frequency of each gray scale, so that the enhancement processing of the product image is realized, the contrast of the product image can be effectively improved, the overall or local characteristics of the product image are effectively improved, the darker part in the product image becomes brighter, more product image details are highlighted, the layering sense of the product image is increased, and the imaging effect of the product image is effectively improved; meanwhile, the quality evaluation module is used for matching the evaluation of the user on the target product with the quality index of the target product to obtain the specific opinion under the evaluation index, so that the quality problem of the target product can be traced, and the quality evaluation result of the target product is accurate. The enterprise can pertinently improve the design of target products, adjust sales strategies and services and the like through the evaluation result, and therefore profits of the enterprise are improved.
The above description is only for the purpose of illustrating the present invention and the appended claims are not to be construed as limiting the scope of the invention, which is intended to cover all modifications, equivalents and improvements that are within the spirit and scope of the invention as defined by the appended claims.

Claims (10)

1. An image-data-based quality discrimination ranking management system, characterized by comprising:
the system comprises a product image acquisition module, a main control module, an image enhancement module, an image feature extraction module, a product traceability module, a quality identification module, a quality evaluation module, a grade determination module and a display module;
the product image acquisition module is connected with the main control module and is used for acquiring product image data through the camera equipment;
the main control module is connected with the product image acquisition module, the image enhancement module, the image feature extraction module, the product traceability module, the quality identification module, the quality evaluation module, the grade determination module and the display module and is used for controlling the normal work of each module;
the image enhancement module is connected with the main control module and is used for enhancing the acquired image through an image enhancement program;
the image feature extraction module is connected with the main control module and used for extracting the image features of the product through an extraction program;
the product source tracing module is connected with the main control module and is used for tracing the source of the product through a source tracing program;
the quality identification module is connected with the main control module and is used for identifying the product quality through an identification program;
the quality evaluation module is connected with the main control module and used for evaluating the product quality through an evaluation program;
the grade determining module is connected with the main control module and used for determining the quality grade of the product;
and the display module is connected with the main control module and is used for displaying the product image, the product source tracing result, the quality identification result, the quality evaluation result and the product grade.
2. The method for managing quality discrimination ranking based on image data according to claim 1, characterized in that the method for managing quality discrimination ranking based on image data comprises the steps of:
the method comprises the following steps that firstly, product image data are collected through a product image collecting module by utilizing camera equipment;
secondly, the main control module utilizes an image enhancement program to enhance the collected image through an image enhancement module;
extracting the image characteristics of the product by using an extraction program through an image characteristic extraction module; tracing the source of the product by using a tracing program through a product tracing module; identifying the product quality by using an identification program through a quality identification module;
evaluating the product quality by utilizing an evaluation program through a quality evaluation module; determining the quality grade of the product through a grade determining module;
and fifthly, displaying the product image, the product traceability result, the quality identification result, the quality evaluation result and the product grade through a display module.
3. The system for quality discrimination hierarchical management based on image data according to claim 1, wherein the image enhancement module enhancement method is as follows:
(1) Collecting a product image, and denoising the product image; counting the number of pixel points corresponding to each gray scale in the product image, thereby determining the occurrence frequency of each gray scale; solving the Lp norm of the times of occurrence of each gray scale, thereby obtaining the Lp norm of the times of occurrence of each gray scale;
(2) In the process of performing histogram equalization processing on a product image, when the number of times of occurrence of a certain gray scale increases, the rate of increase decreases with the increase of the number of times of occurrence of the gray scale, and histogram equalization is performed on the product image based on the Lp norm of the number of times of occurrence of each gray scale so as to realize enhancement processing on the product image;
the method for denoising the product image comprises the following steps:
estimating the noise intensity of the product image according to the pixel value of the pixel in the product image;
carrying out primary denoising treatment on the obtained product image to be processed to obtain a primary denoised product image;
calculating residual quantity of a central pixel corresponding to each unit region on the product image to be processed, wherein the residual quantity is an absolute value of a difference value between a gray value of the product image to be processed and a gray value of the preliminary denoising product image;
calculating a weight matrix corresponding to each unit region by using the residual quantity, and performing non-local mean value calculation on the product image to be processed according to the weight matrix to realize the denoising processing of the product image to be processed, wherein the process of calculating the weight matrix comprises the following steps:
selecting any unit area on the product image to be processed, and determining the associated area of any unit area on the product image to be processed;
calculating a weight value corresponding to each of the associated unit regions according to the distance value between the associated unit region and the residual amount to obtain the weight matrix, wherein the weight value corresponding to any associated unit region in the associated region is calculated by using the following formula:
w(n，m)＝e (-(d(n，m)+residuals(n，m))/h) ，
wherein w (n, m) is a weight value corresponding to any associated unit region (n, m), d (n, m) is a distance value between the associated unit region and the unit region, residual (n, m) is a residual amount corresponding to a central pixel of the unit region, and h is a preset control coefficient,
calculating a distance value between said any associated unit area (n, m) and said any unit area (i, j) using the following formula:
Figure FDA0003793594850000031
Figure FDA0003793594850000032
where 2r +1 is the side length of any one of the associated unit areas, and T, k and T are intermediate values.
The Lp norm is obtained by the following method:
storing Lp norms corresponding to integers smaller than a first threshold in advance;
acquiring the Lp norm of the occurrence times of each gray scale of the product image by searching the Lp norm corresponding to the stored integer;
the storage process comprises: storing Lp norms corresponding to integers every interval of a second threshold number;
and the first threshold value is correspondingly set according to the resolution of the product image.
4. The system for quality-discrimination hierarchical management based on image data according to claim 3, wherein the obtaining of the Lp-norm of the number of times each gray scale occurs to obtain the Lp-norm of the number of times each gray scale occurs comprises: when the occurrence frequency of the gray scale is greater than a first threshold value, the Lp norm value of the occurrence frequency of the gray scale is the corresponding Lp norm value when the occurrence frequency of the gray scale is the first threshold value;
the value range of p of the Lp norm is that p is more than or equal to 0 and less than 1;
carrying out histogram equalization on pixel points in the product image through the following formula:
Figure FDA0003793594850000041
wherein I (x, y) is a brightness value of a pixel point at a (x, y) position in the product image, j is an index value of a gray scale in the product image, M is a total number of gray scales in the product image, H × j is an Lp norm value of the number of times of occurrence of the gray scale j in the product image, and ψ (I (x, y)) is a brightness value of the pixel point at the (x, y) position after histogram equalization processing;
the product image is in a YUV format, and the brightness value of the pixel point is the Y component of the pixel point.
5. The system for quality-discrimination hierarchical management based on image data according to claim 1, wherein the quality-evaluation module evaluates as follows:
1) Constructing a product database; acquiring target product data and client comment data of a user on a target product; storing the acquired data into a product database;
2) Preprocessing the client comment data to obtain historical client comment data; acquiring an initial evaluation index based on the target product data; acquiring an index seed word based on the historical client comment data and the initial evaluation index; obtaining a word vector based on the historical customer comment data;
3) Obtaining an evaluation index based on the word vector and the index seed word; acquiring evaluation data based on the word vector and a BLSTM-CRFs model constructed in advance; matching the evaluation index with the evaluation data, and scoring the matched evaluation data;
4) Processing the scoring result based on an evidence reasoning method to obtain evaluation grade reliability distribution; converting the evaluation grade reliability distribution into a utility value; obtaining the evaluation of the quality of the target product based on the utility value;
the evidence-based reasoning method for processing the scoring result comprises the following steps:
presetting the evaluation grade of the evaluation index; converting the scoring result into the reliability of the evaluation grade of the corresponding evaluation index; giving a weight to the evaluation index, and acquiring basic probability distribution of the evaluation index based on the reliability and the weight;
based on an evidence reasoning method, combining the basic probability distribution and fusing the evaluation data to obtain evaluation grade reliability distribution;
based on an evidence reasoning method, combining the basic probability distribution and fusing the evaluation data to obtain the evaluation grade reliability distribution specifically comprises the following steps:
and selecting ER algorithm in an evidence reasoning method to fuse the evaluation data, wherein a fusion formula is as follows:
m n，I(k+1) ＝K I(k+1) (m n，I(k) m n，k+1 +m n，I(k) m H，k+1 +m H，I(k) m n，k+ 1)n＝1，2，…，N
m H，I(k+1) ＝K I(k+1) m H，I(k) m H，k+1
Figure FDA0003793594850000051
wherein:
m n，I(k) means that the first k indexes are fused and then distributed toStage H n The probability of (d);
m H，I(k) representing the probability of the first k indexes being distributed to the complete set H after fusion;
n represents the number of evaluation levels, L represents the number of evaluation indexes,
k represents a normalization coefficient;
m obtained n，I(L) Reconversion to a confidence distribution on the evaluation level:
Figure FDA0003793594850000052
wherein:
β n indicates that L indexes are evaluated to the grade H after being synthesized n The confidence level of the user;
the final obtained evaluation grade reliability distribution is as follows:
S(y)＝{(H n ，β n )n＝1，2，...N}
wherein:
H n the evaluation scale is shown.
6. The system for quality discrimination ranking management based on image data of claim 5 wherein said preprocessing of said customer comment data comprises:
deleting repeated customer comments, customer comments which are not filled with effective contents and customer comments of which the character length is lower than a preset value;
converting the traditional form of the characters in the client comment data into a simplified form;
and performing sentence segmentation, word segmentation and stop word removal on the client comment data.
7. The system for quality discrimination hierarchical management based on image data according to claim 5, wherein the method for obtaining the index seed word comprises:
performing part-of-speech tagging on the historical client comment data; extracting high-frequency nouns and verbs, and filtering out nouns and verbs with frequency lower than a preset value and single words irrelevant to quality attributes; and matching the finally obtained words with the initial evaluation indexes to obtain index seed words corresponding to the initial evaluation indexes.
8. The system for quality discrimination hierarchical management based on image data according to claim 5, wherein the method for obtaining the word vector comprises:
and processing the historical client comment data based on a pre-trained word2vec model to obtain a word vector with a preset dimension.
9. The system for quality-discrimination ranking management based on image data according to claim 5, wherein the method for obtaining the evaluation index includes:
vectorizing the index seed words based on the word vectors to obtain the correlation degrees of all words and words in the historical client comment data and the vectorized index seed words, selecting the first a words with the maximum correlation degrees to expand the index seed words, and removing repeated words to obtain expanded seed words;
processing the extended seed words based on the KNN to obtain target product quality attribute words;
and calculating the frequency of the target product quality attribute words appearing in the historical customer comment data, and filtering out initial evaluation indexes corresponding to the target product quality attribute words with the frequency lower than a preset value to obtain the evaluation indexes.
10. The system for quality-discrimination ranking management based on image data according to claim 5, wherein the method for obtaining the evaluation data includes:
processing the historical client comment data based on a BLSTM-CRFs model and the word vector which are constructed in advance, extracting degree words and emotion words corresponding to the target product quality attribute words, and storing the degree words and the emotion words in a text form to obtain evaluation data;
the method for acquiring the pre-constructed BLSTM-CRFs model comprises the following steps:
building a BLSTM-CRFs model based on a TensorFlow tool; marking the historical customer comment data based on a BIO standard mode to obtain a data set; randomly disordering the sequence of the data set, training the data set for preset times, and selecting a model with the highest accuracy as a pre-constructed BLSTM-CRFs model;
the utility value obtaining method comprises the following steps:
Figure FDA0003793594850000071
wherein:
u(H n ) Indicates evaluation level H n The utility value of (c).
CN202210967037.6A 2022-08-11 2022-08-11 Quality identification grading management system and method based on image data Withdrawn CN115392662A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210967037.6A CN115392662A (en) 2022-08-11 2022-08-11 Quality identification grading management system and method based on image data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210967037.6A CN115392662A (en) 2022-08-11 2022-08-11 Quality identification grading management system and method based on image data

Publications (1)

Publication Number Publication Date
CN115392662A true CN115392662A (en) 2022-11-25

Family

ID=84118186

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210967037.6A Withdrawn CN115392662A (en) 2022-08-11 2022-08-11 Quality identification grading management system and method based on image data

Country Status (1)

Country Link
CN (1) CN115392662A (en)

Similar Documents

Publication Publication Date Title
CN109255506B (en) Internet financial user loan overdue prediction method based on big data
CN110704730A (en) Product data pushing method and system based on big data and computer equipment
CN110751557A (en) Abnormal fund transaction behavior analysis method and system based on sequence model
US20190378180A1 (en) Method and system for generating and using vehicle pricing models
CN112990386B (en) User value clustering method and device, computer equipment and storage medium
CN111523678A (en) Service processing method, device, equipment and storage medium
CN110782349A (en) Model training method and system
CN113239227A (en) Image data structuring method and device, electronic equipment and computer readable medium
WO2021174699A1 (en) User screening method, apparatus and device, and storage medium
CN114118816A (en) Risk assessment method, device and equipment and computer storage medium
CN113590945B (en) Book recommendation method and device based on user borrowing behavior-interest prediction
CN112950347A (en) Resource data processing optimization method and device, storage medium and terminal
CN111784053A (en) Transaction risk detection method, device and readable storage medium
CN110851708A (en) Negative sample extraction method and device, computer equipment and storage medium
CN115392662A (en) Quality identification grading management system and method based on image data
CN114066018A (en) Power distribution station construction material demand prediction method based on support vector
CN113240513A (en) Method for determining user credit line and related device
CN116758576B (en) Marketing information identification method, device, equipment and medium based on machine learning
Burger et al. Deriving correlation matrices for missing financial time-series data
Gaikwad et al. Unveiling Market Dynamics through Machine Learning: Strategic Insights and Analysis
CN115187387B (en) Identification method and equipment for risk merchant
Dwarakanath et al. Optimal Stopping with Gaussian Processes
CN113515867A (en) Model training method, business processing method, device and equipment
CN118071482A (en) Method for constructing retail credit risk prediction model and consumer credit business Scorebetad model
CN116933130A (en) Enterprise industry classification method, system, equipment and medium based on big data

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
WW01 Invention patent application withdrawn after publication

Application publication date: 20221125

WW01 Invention patent application withdrawn after publication