Multi-parameter integrated fiberboard quality assessment method and system
Through the multi-parameter integrated fiberboard quality evaluation method, the problem of inaccurate fiberboard quality evaluation is solved, the full extraction and quantification of fiberboard quality characteristics is realized, the evaluation accuracy and scientific nature of production decisions are improved, and the product quality stability is ensured. .
Patent Information
- Application Number
- CN202410833362.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-26
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-06-26
AI Technical Summary
The quality characteristics of the fiberboard are not fully extracted and quantified during the quality evaluation process of existing fiberboards, resulting in inaccurate quality evaluation, poor scientific production decisions, and unstable product quality.
The multi-parameter integrated fiberboard quality evaluation method is adopted, and the quality evaluation coefficient and derivative evaluation are combined by determining the evaluation quality matrix, training the minimum distance classifier, collecting image data, performing feature extraction and integration, index quantization and hierarchical empowerment calculation, and derivative evaluation. coefficient, determine the quality evaluation results.
The full extraction and quantification of the quality characteristics of fiberboard are achieved, the accuracy of quality evaluation and the scientific nature of production decisions are improved, and the stability of product quality is ensured.
Smart Images

Figure CN118608002B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of quality assessment, and in particular to a multi-parameter integrated fiberboard quality assessment method and system. Background Art
[0002] Fiberboard is a man-made board made of wood or other plant fibers after physical and chemical treatment, mixed with adhesives and hot-pressed. Fiberboard has the characteristics of high strength, durability and easy processing, and is widely used in furniture manufacturing, building decoration, packaging materials and other fields.
[0003] At present, the existing fiberboard quality assessment process does not fully extract and quantify the quality characteristics of the fiberboard, resulting in inaccurate quality assessment and failure to accurately reflect the quality level of the fiberboard. As a result, production decisions are not made based on accurate quality assessment results, causing quality problems in the production process, affecting product quality and production efficiency. At the same time, there is no effective monitoring and control of the quality characteristics of the fiberboard, resulting in product quality fluctuations and failure to maintain consistency and stability. Therefore, a method is needed to solve the above problems.
[0004] In summary, the prior art has technical problems in that the quality characteristics of the fiberboard are not fully extracted and quantified during the fiberboard quality assessment process, resulting in poor accuracy of quality assessment, scientific nature of production decisions, and stability of product quality. Summary of the invention
[0005] The purpose of this application is to provide a multi-parameter integrated fiberboard quality assessment method and system to solve the technical problems in the prior art that the quality characteristics of the fiberboard are not fully extracted and quantified during the fiberboard quality assessment process, resulting in poor accuracy of quality assessment, scientific nature of production decisions, and stability of product quality.
[0006] In view of the above problems, the present application provides a multi-parameter integrated fiberboard quality assessment method and system.
[0007] In a first aspect, the present application provides a multi-parameter integrated fiberboard quality assessment method, which is implemented through a multi-parameter integrated fiberboard quality assessment system, wherein the method includes: determining an assessment quality matrix of the fiberboard, wherein the assessment quality matrix has quality indicators as matrix rows and vector features as matrix columns, and wherein there is a one-to-many mapping relationship between the quality indicators and the vector features; training a minimum distance classifier and constructing a quality detection unit, wherein the minimum distance classifier can perform self-target setting, and the quality detection unit includes multiple integrated detectors; collecting image data of the fiberboard, combining with the quality detection unit, performing feature extraction and integration guided by the assessment quality matrix, and determining an integrated feature distribution, wherein the image data includes multiple perspectives and internal cross-sections; based on the integrated feature distribution, performing indicator quantification and hierarchical weighted calculation to determine a quality assessment coefficient; based on the integrated feature distribution, performing derivative statistical analysis to determine a derived assessment coefficient; based on the integrated feature distribution, the quality assessment coefficient and the derived assessment coefficient, determining a quality assessment result.
[0008] In a second aspect, the present application also provides a multi-parameter integrated fiberboard quality assessment system for executing the multi-parameter integrated fiberboard quality assessment method as described in the first aspect, wherein the system comprises: an assessment quality matrix determination module, the assessment quality matrix determination module is used to determine the assessment quality matrix of the fiberboard, wherein the assessment quality matrix has quality indicators as matrix rows and vector features as matrix columns, and wherein a one-to-many mapping relationship exists between the quality indicators and the vector features; a quality detection unit construction module, the quality detection unit construction module is used to train a minimum distance classifier and construct a quality detection unit, the minimum distance classifier can perform self-target setting, and the quality detection unit comprises multiple integrated detectors; an integrated feature distribution determination module, the integrated feature distribution determination module A feature distribution determination module is used to collect image data of the fiberboard, and in combination with the quality detection unit, performs feature extraction and integration guided by the assessment quality matrix to determine the integrated feature distribution, wherein the image data includes multiple perspectives and internal cross-sections; a quality assessment coefficient determination module is used to perform indicator quantification and hierarchical weighted calculation based on the integrated feature distribution to determine the quality assessment coefficient; a derivative assessment coefficient determination module is used to perform derivative statistical analysis based on the integrated feature distribution to determine the derivative assessment coefficient; a quality assessment result determination module is used to determine the quality assessment result based on the integrated feature distribution, the quality assessment coefficient and the derivative assessment coefficient.
[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0010] By determining an evaluation quality matrix of fiberboard, wherein the evaluation quality matrix has quality indicators as matrix rows and vector features as matrix columns, wherein there is a one-to-many mapping relationship between the quality indicators and the vector features; training a minimum distance classifier and constructing a quality detection unit, wherein the minimum distance classifier can perform self-target setting, and the quality detection unit includes multiple integrated detectors; collecting image data of the fiberboard, combining with the quality detection unit, extracting and integrating features guided by the evaluation quality matrix, and determining the integrated feature distribution, wherein the image data includes multiple perspectives and internal cross-sections; based on the integrated feature distribution, performing indicator quantification and hierarchical weighted calculation, and determining the quality assessment coefficient; based on the integrated feature distribution, performing derivative statistical analysis, and determining the derivative assessment coefficient; based on the integrated feature distribution, the quality assessment coefficient and the derived assessment coefficient, determining the quality assessment result, thereby achieving the technical goal of fully extracting and quantifying the quality features of the fiberboard, and achieving the technical effect of improving the accuracy of quality assessment, the scientific nature of production decisions, and the stability of product quality.
[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0012] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.
[0013] Figure 1 This is a schematic flow chart of the multi-parameter integrated fiberboard quality assessment method of this application.
[0014] Figure 2 This is a schematic diagram of the structure of the multi-parameter integrated fiberboard quality assessment system for this application.
[0015] Description of reference numerals:
[0016] An evaluation quality matrix determination module 11, a quality detection unit construction module 12, an integrated feature distribution determination module 13, a quality evaluation coefficient determination module 14, a derived evaluation coefficient determination module 15, and a quality evaluation result determination module 16. DETAILED DESCRIPTION
[0017] This application provides a multi-parameter integrated fiberboard quality assessment method and system to solve the technical problem in the prior art that the quality characteristics of the fiberboard are not fully extracted and quantified during the fiberboard quality assessment process, resulting in poor accuracy of quality assessment, scientificity of production decisions, and stability of product quality. The technical goal of fully extracting and quantifying the quality characteristics of the fiberboard is achieved, achieving the technical effect of improving the accuracy of quality assessment, the scientificity of production decisions, and the stability of product quality.
[0018] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.
[0019] Embodiment 1
[0020] Please refer to the attached Figure 1 The present application provides a multi-parameter integrated fiberboard quality assessment method, wherein the method is applied to a multi-parameter integrated fiberboard quality assessment system, and the method specifically comprises the following steps:
[0021] Step 1: Determine an evaluation quality matrix of the fiberboard, wherein the evaluation quality matrix has quality indicators as matrix rows and vector features as matrix columns, and wherein there is a one-to-many mapping relationship between the quality indicators and the vector features.
[0022] Specifically, the quality indicators are determined, including tensile strength, compressive strength, density, moisture content, dimensional stability, formaldehyde emission, etc. Vector features refer to variables that can represent the quality characteristics of fiberboard. Vector features are selected from raw material characteristics, production process parameters, finished product test results, etc. A large amount of fiberboard sample data is collected, including various quality indicators and corresponding vector features. Through data analysis methods such as regression analysis, principal component analysis, cluster analysis, etc., a one-to-many mapping relationship between quality indicators and vector features is established. According to the mapping relationship, an evaluation quality matrix is constructed, in which the matrix rows represent different quality indicators and the matrix columns represent different vector features.
[0023] Step 2: Train a minimum distance classifier and construct a quality detection unit. The minimum distance classifier can perform self-targeting. The quality detection unit includes multiple integrated detectors.
[0024] Specifically, collect a large amount of fiberboard sample data, including the quality indicator measurement value and the corresponding category label of each sample. For example, qualified or unqualified. Select a distance measurement method, such as Euclidean distance, Manhattan distance, etc. Calculate the center point of each category in the feature space, that is, the mean vector. Train a minimum distance classifier to identify the center point of each category. Design an integrated detector. For example, a sensor, measuring instrument, or other device that can collect fiberboard quality-related data to ensure that the integrated detector can cover all vector features that are important for fiberboard quality assessment.
[0025] Step three: Collect image data of the fiberboard, combine with the quality detection unit, perform feature extraction and integration guided by the evaluation quality matrix, and determine the integrated feature distribution, wherein the image data includes multiple perspectives and internal cross-sections.
[0026] Specifically, an image acquisition device, such as a camera or scanner, is used to collect image data of the fiberboard from multiple viewpoints and internal cross-sections. The multi-view images provide a comprehensive view of the external structure of the fiberboard, while the internal cross-section images reveal the arrangement and structure of the internal fibers. The acquired image data is combined with a quality inspection unit to utilize the integrated detectors and analysis tools in the inspection unit. The quality inspection unit includes sensors and measuring devices that can provide physical and chemical property information complementary to the image data. Using the evaluation quality matrix as a guide, features corresponding to the quality indicators in the matrix are extracted from the image data. Feature extraction includes image processing such as texture analysis, color recognition, shape detection, and defect identification. The extracted features are integrated to construct an integrated feature distribution, which reflects the quality characteristics of the fiberboard.
[0027] Step 4: Based on the integrated feature distribution, perform indicator quantification and hierarchical weighting calculation to determine the quality assessment coefficient.
[0028] Specifically, based on the integrated feature distribution, indicator quantification and hierarchical weighted calculation are performed. Indicator quantification refers to converting each feature in the integrated feature distribution into a quantifiable value for comparison and analysis. The quantification process involves image processing, statistical analysis, machine learning, etc. to ensure the accuracy and consistency of the features. The quantified features better reflect the quality characteristics of the fiberboard, such as strength, density, uniformity, etc. Hierarchical weighted calculation refers to assigning weights to each quantitative indicator based on the importance or influence of the feature. The determination of weights can be based on expert knowledge, historical data analysis, principal component analysis, etc. Hierarchical weighted calculation is performed based on the interrelationships and dependencies between different features. Determine the quality assessment coefficient. The quality assessment coefficient is a comprehensive indicator that comprehensively considers all quantitative indicators and hierarchical weightings, reflecting the overall quality level of the fiberboard.
[0029] Step 5: Based on the integrated feature distribution, perform derivative statistical analysis to determine the derivative evaluation coefficient.
[0030] Specifically, statistical analysis is performed on the data in the integrated feature distribution to discover hidden patterns or associations between the data. Statistical methods include correlation analysis, regression analysis, cluster analysis, principal component analysis, etc. Based on the results of the statistical analysis, the derived evaluation coefficients are determined to obtain the impact on the quality of the fiberboard. The derived evaluation coefficients include coefficient of variation, confidence interval, prediction error, etc., which provide an assessment of the stability and predictability of fiberboard quality, help predict the quality characteristics that fiberboard may exhibit in the future, and thus guide production process optimization and risk assessment.
[0031] Step six: Determine a quality assessment result based on the integrated feature distribution, the quality assessment coefficient and the derived assessment coefficient.
[0032] Specifically, based on the integrated feature distribution, the quality assessment coefficient and the derived assessment coefficient are combined to form a more comprehensive quality assessment result. The quality assessment result reflects the current quality status of the fiberboard and also provides a prediction of future quality performance and early warning of potential problems. The quality assessment result can be used for decision support, such as optimization of production processes, setting of quality control standards, market positioning, etc.
[0033] The multi-parameter integrated fiberboard quality assessment method is applied to a multi-parameter integrated fiberboard quality assessment system, which can achieve the technical goal of fully extracting and quantifying the quality characteristics of the fiberboard, and achieve the technical effect of improving the accuracy of quality assessment, the scientific nature of production decisions, and the stability of product quality.
[0034] Furthermore, the present application also includes:
[0035] The neighborhood fiber distance is taken as the minimum distance target; based on the minimum distance target, the spatial neighborhood fiber feature position is located in combination with the minimum distance classifier to determine the feature extraction point; the feature extraction points are traversed to extract fiber features and the spatial fiber structure distribution is determined by combination; the spatial fiber structure distribution is added to the integrated feature distribution.
[0036] Specifically, the neighborhood fiber distance refers to the relative distance between fibers in the same spatial region. The neighborhood fiber distance helps the classifier identify the distribution characteristics of fibers in the fiberboard. The neighborhood fiber distance is calculated by the distance measurement method to determine the neighborhood fiber distance. According to the minimum neighborhood fiber distance, the minimum distance classifier is used to classify the fiberboard samples. The position of each fiber feature in space is determined by the decision boundary of the classifier. After determining the position of the spatial neighborhood fiber feature, each feature extraction point is visited in turn to extract the fiber feature of each point. Fiber features include fiber length, diameter, orientation, density, morphology, etc. The fiber features extracted from each feature extraction point are combined to construct a comprehensive spatial fiber structure distribution map to obtain the structure and arrangement of the fibers inside the fiberboard. The determined spatial fiber structure distribution is added to the integrated feature distribution for the final fiberboard quality assessment.
[0037] Through refined feature extraction and structural analysis, it helps to improve product quality and reduce production costs.
[0038] Furthermore, the present application also includes:
[0039] If there are multiple minimum distance targets, integrated detector matching and parallel distribution are performed to determine a temporary detection architecture; a mapping between the minimum distance classifier and the temporary detection architecture is established, and feature positioning and extraction based on the minimum distance target are performed in parallel to determine parallel detection results; the parallel detection results are integrated and added to the integrated feature distribution, and the temporary detection architecture is dissociated.
[0040] Specifically, for each minimum distance target, a corresponding integrated detector is designed and matched, and the integrated detector can work simultaneously or in parallel to improve the detection efficiency. The parallel distributed detector can detect multiple areas or multiple quality indicators of the fiberboard at the same time, thereby speeding up the detection speed. According to the configuration and distribution of the integrated detectors, a temporary detection architecture is established to adapt to the detection needs of multiple minimum distance targets. The temporary detection architecture is adjusted according to different detection tasks to achieve maximum flexibility and adaptability.
[0041] Then, the minimum distance classifier is mapped to the temporary detection architecture to ensure that each classifier can work with the corresponding detector. Feature location and extraction are performed in parallel, and fiber features are identified and extracted based on multiple minimum distance targets at the same time, which improves the comprehensiveness and accuracy of detection. Through parallel processing, multiple detection results are output simultaneously, and each result corresponds to a minimum distance target.
[0042] Next, the results of the parallel inspections are combined into an integrated feature distribution to provide a comprehensive view for the final fiberboard quality assessment. The integrated feature distribution helps improve the accuracy and reliability of the quality assessment. After the inspection is completed, the temporary inspection architecture can be disassembled to facilitate future reconfiguration and optimization.
[0043] The modular and detachable design makes the detection system more flexible and can quickly adapt to new detection tasks or different production requirements.
[0044] Furthermore, the present application also includes:
[0045] The evaluation quality matrix is traversed to determine a standardized matrix, which is a reference for quality evaluation; based on the mapping relationship between the quality index and the vector feature, combined with the standardized matrix, a feature weighting and assembly calculation are performed on the integrated feature distribution to determine the indicator quality coefficient, and the indicator quality coefficient corresponds to the quality index one by one; the indicator quality coefficient is traversed, and a secondary weighting and assembly calculation is performed based on the indicator importance to determine the one-dimensional quality evaluation coefficient.
[0046] Specifically, the evaluation quality matrix is accessed in turn, that is, the mapping relationship between each quality indicator and the vector feature is reviewed. A standardized matrix is determined, which contains the normalized or standardized quality indicator values as a reference standard for quality evaluation. The standardization process can eliminate the dimensional influence between different indicators, so that different indicators can be compared and integrated.
[0047] Then, based on the mapping relationship between quality indicators and vector features, combined with the standardized matrix, each feature in the integrated feature distribution is weighted. Feature weighting can be determined by expert knowledge, statistical analysis or machine learning techniques, and the weight reflects the degree of influence of each feature on the quality of the fiberboard. The overall calculation is to synthesize the weighted features to determine the quality coefficient of each quality indicator, reflecting the contribution to the overall quality.
[0048] Next, after determining the quality coefficient of each quality indicator, the quality coefficients are visited in turn. Secondary weighting is performed based on the indicator importance, that is, weights are applied to each quality coefficient again to reflect the importance of different indicators in the overall quality assessment.
[0049] Next, the overall calculation after secondary weighting will obtain a one-dimensional quality assessment coefficient, which integrates the final evaluation results of all quality indicators and characteristics.
[0050] By achieving a comprehensive, objective and accurate assessment of fiberboard quality, it provides strong support for production management and quality control.
[0051] Furthermore, the present application also includes:
[0052] Based on the composition ratio of the fiberboard, the first quality data is determined; the process flow of the fiberboard is interacted with to perform process gain evaluation and determine the second gain data; the first quality data and the second gain data are fitted to determine the two-dimensional quality assessment coefficient; the one-dimensional quality assessment coefficient and the two-dimensional quality assessment coefficient are averaged to determine the quality assessment coefficient.
[0053] Specifically, the first quality data is determined based on the composition ratio, which refers to the proportion of various raw materials in the fiberboard (such as wood pulp, adhesives, additives, etc.). By analyzing the composition ratio, the impact on the quality of the fiberboard is determined, and the first quality data is obtained, which reflects the contribution of different components to the performance of the fiberboard.
[0054] Then, the interactive process flow is evaluated for process gain, which includes the manufacturing process of fiberboard, such as pressing, drying, cutting, etc. Each step in the process flow is evaluated to determine the gain on the quality of the final product, thereby obtaining the second gain data. The process gain evaluation is performed based on production efficiency, energy consumption, equipment status, etc.
[0055] Next, the first quality data and the second gain data are fitted to determine the two-dimensional quality evaluation coefficient, and the first quality data and the second gain data are fitted. The fitting method can be linear regression, polynomial regression or machine learning model. Through fitting, the two-dimensional quality evaluation coefficient is obtained, which comprehensively considers the influence of the component ratio and the process flow on the quality.
[0056] Next, the one-dimensional quality assessment coefficient and the two-dimensional quality assessment coefficient are averaged to obtain a comprehensive quality assessment coefficient, because the raw materials, process flow and final product performance are taken into account at the same time, thereby obtaining a more comprehensive fiberboard quality assessment.
[0057] By more accurately evaluating the quality of fiberboard, a scientific basis can be provided for production decisions, thereby improving product quality and meeting market demand.
[0058] Furthermore, the present application also includes:
[0059] Determine a derived evaluation target and determine a target evaluation characteristic, wherein the derived evaluation target includes at least one item, and the target evaluation characteristic is used to determine an evaluation decision method; based on the derived evaluation target, traverse the integrated feature distribution to extract derived associated features; traverse the derived associated features, combine the target evaluation characteristics, and evaluate and determine the derived evaluation coefficient.
[0060] Specifically, the derived evaluation objectives and target evaluation characteristics are determined. Derived evaluation objectives refer to indirect indicators derived from basic quality indicators, obtained through the long-term performance, environmental impact, cost-effectiveness, etc. of the product. Target evaluation characteristics refer to the methods and criteria used to determine the derived evaluation objectives, for example, using statistical models, machine learning algorithms, or expert systems.
[0061] Then, the integrated feature distribution is accessed in turn to extract derived related features, and based on the derived evaluation target, the integrated feature distribution is traversed to extract features related to the derived evaluation target. The derived related features include product durability, recyclability, production cost, etc.
[0062] Next, the extracted derived associated features are visited sequentially and evaluated in combination with the target evaluation features, including quantification, weighting and comprehensive calculation of each feature to determine the contribution to the overall quality and the derived evaluation coefficient, which is an indicator reflecting the degree of influence of the derived target on the fiberboard quality.
[0063] By ensuring that the quality assessment of fiberboard considers not only current performance indicators but also long-term and environmental factors, it provides more comprehensive information for production and management.
[0064] Furthermore, the present application also includes:
[0065] Traverse the quality assessment results to locate the quality defects of the fiberboard; trace the causes of the quality defects, locate the process points and make optimization decisions in combination with the process flow, and determine the optimization strategy; based on the optimization strategy, reversely optimize the production of the fiberboard.
[0066] Specifically, the quality assessment results are accessed in turn, and the quality coefficient of each indicator is analyzed to identify where the quality is insufficient. Locating quality defects means determining that the quality indicators do not meet the predetermined standards or expected values, including insufficient strength, deformation, delamination, uneven density, etc.
[0067] Then, we conduct an in-depth analysis of each quality defect identified to find out the root cause of the defect, including raw material quality, component ratio, production equipment performance, operator skills, environmental conditions, etc. Based on the results of the cause tracing, combined with the fiberboard process flow, we determine the process point that causes the defect. We locate the process point and make optimization decisions to reduce or eliminate the occurrence of quality defects.
[0068] Next, based on the process point positioning and optimization decision, specific optimization measures are determined, including adjusting the raw material ratio, improving equipment maintenance procedures, strengthening operator training, improving the production environment, etc. The determined optimization strategy is applied to the production process to actually optimize the production of fiberboard. By implementing the optimization strategy, product quality can be improved, defect rate can be reduced, and overall production efficiency can be improved.
[0069] This improves product quality by ensuring that the fiberboard production process is more stable and efficient.
[0070] In summary, the multi-parameter integrated fiberboard quality assessment method provided in this application has the following technical effects:
[0071] By determining an evaluation quality matrix of fiberboard, wherein the evaluation quality matrix has quality indicators as matrix rows and vector features as matrix columns, wherein there is a one-to-many mapping relationship between the quality indicators and the vector features; training a minimum distance classifier and constructing a quality detection unit, wherein the minimum distance classifier can perform self-target setting, and the quality detection unit includes multiple integrated detectors; collecting image data of the fiberboard, combining with the quality detection unit, extracting and integrating features guided by the evaluation quality matrix, and determining the integrated feature distribution, wherein the image data includes multiple perspectives and internal cross-sections; based on the integrated feature distribution, performing indicator quantification and hierarchical weighted calculation, and determining the quality assessment coefficient; based on the integrated feature distribution, performing derivative statistical analysis, and determining the derivative assessment coefficient; based on the integrated feature distribution, the quality assessment coefficient and the derived assessment coefficient, determining the quality assessment result, thereby achieving the technical goal of fully extracting and quantifying the quality features of the fiberboard, and achieving the technical effect of improving the accuracy of quality assessment, the scientific nature of production decisions, and the stability of product quality.
[0072] Embodiment 2
[0073] Based on the multi-parameter integrated fiberboard quality assessment method in the aforementioned embodiment, the present application also provides a multi-parameter integrated fiberboard quality assessment system, see the attached Figure 2 , the system comprising:
[0074] An evaluation quality matrix determination module 11, the evaluation quality matrix determination module 11 is used to determine the evaluation quality matrix of the fiberboard, wherein the evaluation quality matrix has quality indicators as matrix rows and vector features as matrix columns, wherein the quality indicators and the vector features have a one-to-many mapping relationship; a quality detection unit construction module 12, the quality detection unit construction module 12 is used to train a minimum distance classifier and construct a quality detection unit, the minimum distance classifier can perform self-target setting, and the quality detection unit includes multiple integrated detectors; an integrated feature distribution determination module 13, the integrated feature distribution determination module 13 is used to collect image data of the fiberboard, combined with the quality detection unit , feature extraction and integration are performed guided by the evaluation quality matrix to determine the integrated feature distribution, and the image data includes multiple perspectives and internal cross-sections; a quality assessment coefficient determination module 14, the quality assessment coefficient determination module 14 is used to perform indicator quantification and hierarchical weighting calculation based on the integrated feature distribution to determine the quality assessment coefficient; a derivative assessment coefficient determination module 15, the derivative assessment coefficient determination module 15 is used to perform derivative statistical analysis based on the integrated feature distribution to determine the derivative assessment coefficient; a quality assessment result determination module 16, the quality assessment result determination module 16 is used to determine the quality assessment result based on the integrated feature distribution, the quality assessment coefficient and the derivative assessment coefficient.
[0075] Furthermore, the system is also used for:
[0076] The neighborhood fiber distance is taken as the minimum distance target; based on the minimum distance target, the spatial neighborhood fiber feature position is located in combination with the minimum distance classifier to determine the feature extraction point; the feature extraction points are traversed to extract fiber features and the spatial fiber structure distribution is determined by combination; the spatial fiber structure distribution is added to the integrated feature distribution.
[0077] Furthermore, the system is also used for:
[0078] If there are multiple minimum distance targets, integrated detector matching and parallel distribution are performed to determine a temporary detection architecture; a mapping between the minimum distance classifier and the temporary detection architecture is established, and feature positioning and extraction based on the minimum distance target are performed in parallel to determine parallel detection results; the parallel detection results are integrated and added to the integrated feature distribution, and the temporary detection architecture is dissociated.
[0079] Furthermore, the system is also used for:
[0080] The evaluation quality matrix is traversed to determine a standardized matrix, which is a reference for quality evaluation; based on the mapping relationship between the quality index and the vector feature, combined with the standardized matrix, a feature weighting and assembly calculation are performed on the integrated feature distribution to determine the indicator quality coefficient, and the indicator quality coefficient corresponds to the quality index one by one; the indicator quality coefficient is traversed, and a secondary weighting and assembly calculation is performed based on the indicator importance to determine the one-dimensional quality evaluation coefficient.
[0081] Furthermore, the system is also used for:
[0082] Based on the composition ratio of the fiberboard, the first quality data is determined; the process flow of the fiberboard is interacted with to perform process gain evaluation and determine the second gain data; the first quality data and the second gain data are fitted to determine the two-dimensional quality assessment coefficient; the one-dimensional quality assessment coefficient and the two-dimensional quality assessment coefficient are averaged to determine the quality assessment coefficient.
[0083] Furthermore, the system is also used for:
[0084] Determine a derived evaluation target and determine a target evaluation characteristic, wherein the derived evaluation target includes at least one item, and the target evaluation characteristic is used to determine an evaluation decision method; based on the derived evaluation target, traverse the integrated feature distribution to extract derived associated features; traverse the derived associated features, combine the target evaluation characteristics, and evaluate and determine the derived evaluation coefficient.
[0085] Furthermore, the system is also used for:
[0086] Traverse the quality assessment results to locate the quality defects of the fiberboard; trace the causes of the quality defects, locate the process points and make optimization decisions in combination with the process flow, and determine the optimization strategy; based on the optimization strategy, reversely optimize the production of the fiberboard.
[0087] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The multi-parameter integrated fiberboard quality assessment method and specific examples in the aforementioned embodiment 1 are also applicable to the multi-parameter integrated fiberboard quality assessment system of this embodiment. Through the aforementioned detailed description of the multi-parameter integrated fiberboard quality assessment method, those skilled in the art can clearly understand the multi-parameter integrated fiberboard quality assessment system of this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.
[0088] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
[0089] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.
Claims
1. A multi-parameter integrated fiberboard quality assessment method, characterized in that: The method comprises: Determine an evaluation quality matrix of the fiberboard, wherein the evaluation quality matrix has quality indicators as matrix rows and vector features as matrix columns, wherein there is a one-to-many mapping relationship between the quality indicators and the vector features; Training a minimum distance classifier and constructing a quality detection unit, wherein the minimum distance classifier can perform self-target setting, and the quality detection unit includes a plurality of integrated detectors; Collecting image data of the fiberboard, combining with the quality detection unit, performing feature extraction and integration guided by the evaluation quality matrix, and determining integrated feature distribution, wherein the image data includes multiple viewing angles and internal cross-sections; Based on the integrated feature distribution, index quantification and hierarchical weighting calculation are performed to determine the quality assessment coefficient; Based on the integrated feature distribution, a derivative statistical analysis is performed to determine a derivative evaluation coefficient; Determining a quality assessment result based on the integrated feature distribution, the quality assessment coefficient and the derived assessment coefficient; The feature extraction and integration comprises: The neighborhood fiber distance is taken as the minimum distance target; Based on the minimum distance target, the minimum distance classifier is combined to locate the spatial neighborhood fiber feature position and determine the feature extraction point; Traversing the feature extraction points, extracting fiber features, and combining to determine spatial fiber structure distribution; adding the spatial fiber structure distribution into the integrated feature distribution; The feature extraction and integration comprises: If there are multiple minimum distance targets, perform integrated detector matching and parallel distribution to determine a temporary detection architecture; Establishing a mapping between the minimum distance classifier and the temporary detection architecture, performing feature location and extraction based on the minimum distance target in parallel, and determining parallel detection results; The parallel detection results are integrated and added into the integrated feature distribution, and the temporary detection architecture is dissociated.
2. The multi-parameter integrated fiberboard quality assessment method according to claim 1, characterized in that: The indicator quantification and hierarchical weighting calculation include: Traversing the evaluation quality matrix to determine a standardized matrix, wherein the standardized matrix is a reference for quality evaluation; Based on the mapping relationship between the quality index and the vector feature, in combination with the standardized matrix, the integrated feature distribution is weighted and the overall calculation is performed to determine the index quality coefficient, and the index quality coefficient corresponds to the quality index one by one; The indicator quality coefficients are traversed, and secondary weighting and overall calculation are performed based on the indicator importance to determine a one-dimensional quality assessment coefficient.
3. The multi-parameter integrated fiberboard quality assessment method according to claim 2, characterized in that: The determining of the quality assessment coefficient comprises: Determining first quality data based on the composition ratio of the fiberboard; Interacting the process flow of the fiberboard, performing process gain evaluation, and determining second gain data; Fitting the first quality data and the second gain data to determine a two-dimensional quality assessment coefficient; The one-dimensional quality assessment coefficient and the two-dimensional quality assessment coefficient are averaged to determine the quality assessment coefficient.
4. The multi-parameter integrated fiberboard quality assessment method according to claim 1, characterized in that: Conduct derivative statistical analysis to determine the derivative evaluation coefficients including: Determining a derived evaluation target and determining a target evaluation characteristic, wherein the derived evaluation target includes at least one item, and the target evaluation characteristic is used to determine an evaluation decision method; Based on the derived evaluation target, traverse the integrated feature distribution to extract derived associated features; The derived associated features are traversed, combined with the target evaluation characteristics, and the derived evaluation coefficient is evaluated and determined.
5. The multi-parameter integrated fiberboard quality assessment method according to claim 1, characterized in that: The method further comprises: Traversing the quality assessment results to locate quality defects of the fiberboard; Trace the causes of the quality defects, locate the process points and make optimization decisions based on the process flow, and determine the optimization strategy; Based on the optimization strategy, the production optimization of the fiberboard is performed in reverse.
6. Multi-parameter integrated fiberboard quality assessment system, characterized in that: Steps for implementing the multi-parameter integrated fiberboard quality assessment method according to any one of claims 1 to 5, the multi-parameter integrated fiberboard quality assessment system comprising: An evaluation quality matrix determination module, wherein the evaluation quality matrix determination module is used to determine an evaluation quality matrix of the fiberboard, wherein the evaluation quality matrix has quality indicators as matrix rows and vector features as matrix columns, wherein there is a one-to-many mapping relationship between the quality indicators and the vector features; A quality detection unit construction module, the quality detection unit construction module is used to train a minimum distance classifier to construct a quality detection unit, the minimum distance classifier can perform self-target setting, and the quality detection unit includes a plurality of integrated detectors; An integrated feature distribution determination module, the integrated feature distribution determination module is used to collect image data of the fiberboard, combine with the quality detection unit, perform feature extraction and integration guided by the evaluation quality matrix, and determine the integrated feature distribution, the image data includes multiple viewing angles and internal cross-sections; A quality assessment coefficient determination module, the quality assessment coefficient determination module is used to perform indicator quantification and hierarchical weighting calculation based on the integrated feature distribution to determine the quality assessment coefficient; A derivative evaluation coefficient determination module, the derivative evaluation coefficient determination module is used to perform derivative statistical analysis based on the integrated feature distribution to determine the derivative evaluation coefficient; A quality assessment result determination module is used to determine a quality assessment result based on the integrated feature distribution, the quality assessment coefficient and the derived assessment coefficient.
