An implementation method of appearance consistency of industrial products based on fusion of CMF parameter control

Abstract

The application discloses a kind of industrial product appearance consistency implementation methods of fusion CMF parameter control, comprising: through multidimensional sensor array acquisition material physical characteristics, color space, process execution related parameters, form multi-source CMF parameter matrix;Through multidimensional CMF parameter mapping fusion model generates core parameter set, and then parameter coupling adjustment is carried out through material color process collaborative optimization model, and target parameter set is obtained;Through appearance defect traceability and CMF calibration model positioning and correcting parameter deviation term, input the parameter set after calibration into industrial product appearance consistency verification platform, and complete multidimensional consistency detection through appearance feature comparison library;Finally, according to the detection result dynamic feedback adjustment CMF parameter control process, form closed loop mechanism.The method is through multi-model cooperation and whole-process management and control, realizes CMF parameter cross-dimension fusion and accurate optimization, improves the fine degree and stability of industrial product appearance consistency control.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial product appearance implementation, and in particular to an industrial product appearance consistency implementation method fusing CMF parameter control. BACKGROUND

[0002] In the field of industrial product manufacturing, appearance consistency is a key factor affecting product market competitiveness, brand reputation and user experience, and the coordinated control of material, color and process (CMF) is the core link determining appearance consistency. With the increasing demand for product appearance refinement and personalization in the consumer market, the appearance design of industrial products is becoming increasingly complex, involving an increasing number of CMF parameter dimensions, and the coupling relationship between different parameters is becoming more closely linked. In traditional industrial production, CMF parameters are mostly controlled in a single dimension independently, which is difficult to achieve cross-dimensional coordinated matching, resulting in appearance differences in the batch production process. Against this background, it is urgent to build a technical solution fusing CMF parameter control throughout the whole process, to achieve systematic control of CMF parameters through multi-model collaboration and platform verification, and to solve the industry pain point of difficulty in ensuring industrial product appearance consistency.

[0003] The existing technology has obvious deficiencies in the control of industrial product appearance consistency, mainly in two core aspects: first, there is a lack of deep integration and collaborative optimization mechanism for CMF multi-dimensional parameters, and existing solutions mostly handle material, color and process parameters separately, without fully considering the coupling relationship between parameters, resulting in new appearance deviations after parameter adjustment, and unable to form a precise combination of CMF parameters; second, the correlation between defect tracing and parameter calibration is insufficient, and existing technologies often only detect and repair surface defects, without establishing a direct mapping relationship between defects and CMF parameter deviations, making it difficult to locate the parameter problem that causes appearance inconsistency from the root cause, and lacking a closed-loop parameter feedback adjustment mechanism, which cannot achieve dynamic optimization of CMF parameters in the production process, ultimately affecting the stability and reliability of product appearance consistency. SUMMARY

[0004] In order to overcome the shortcomings and deficiencies of the existing technology, the present application provides an industrial product appearance consistency implementation method fusing CMF parameter control.

[0005] The technical scheme adopted by the present application is a kind of industrial product appearance consistency implementation method fusing CMF parameter control, comprising the following steps: S1, the CMF basic parameter set of industrial product is collected, the CMF basic parameter set includes material physical property parameter, color space parameter, process execution parameter, parameter collection is carried out to the different appearance area of product by multidimensional sensor array, and multi-source CMF parameter matrix is formed;S2, the multi-source CMF parameter matrix is input into multidimensional CMF parameter mapping fusion model, and material, color, process parameter are correlated and mapped across dimensions by model, and the CMF calibration parameter set after fusion is generated;S3, the CMF calibration parameter set is input into material color process collaborative optimization model, and collaborative adjustment is carried out based on the coupling relationship between parameters, and the target CMF parameter set after optimization is output;S4, the target CMF parameter set after optimization is analyzed by appearance defect traceability and CMF calibration model Defect correlation, the parameter deviation item that may cause appearance inconsistency is located and calibrated and corrected;S5, the CMF parameter set after calibration and correction is input into industrial product appearance consistency verification platform, and multidimensional consistency detection is carried out through the appearance feature comparison library built by platform, and detection result data is generated;S6, according to the detection result data, CMF parameter control process is dynamically feedback adjusted, and closed-loop CMF parameter control mechanism is formed, and the implementation of industrial product appearance consistency is completed.

[0006] Further, the expression of the multidimensional CMF parameter mapping fusion model is: Wherein, is the CMF calibration parameter set after fusion, is a mapping fusion weight coefficient, is a material physical property parameter matrix, is a color space parameter vector, is a process execution parameter matrix, is an activation function, is a gradient calculation operator, is a three-dimensional convolution operation operator.

[0007] Further, the expression of the material color process collaborative optimization model is: Wherein, is the target CMF parameter set after optimization, is a process parameter deviation adjustment coefficient, is an initial process execution parameter matrix, is a collaborative optimization coefficient, is a KL divergence weight coefficient, is a tensor product operation, is a principal component analysis operator, is a KL divergence calculation operator, is a reference process execution parameter matrix.

[0008] Further, the expression of appearance defect traceability and CMF calibration model is: wherein, is the calibrated CMF parameter set, is the defect traceability coefficient, is the calibration adjustment coefficient, is the appearance defect feature parameter matrix, is the matrix inverse operator, is the CMF parameter Jacobian matrix, is the long short-term memory network operation operator.

[0009] Further, the expression of appearance consistency verification platform of industrial products is: wherein, is the consistency detection result data, is the CMF parameter vector of the i-th detection area, is the reference CMF parameter vector, is the number of detection areas, is the cosine similarity weight, is the intersection over union weight, is the cosine similarity calculation operator, is the intersection over union calculation operator, is the detected CMF parameter set, is the standard CMF parameter set.

[0010] Further, the expression of feedback adjustment of closed-loop CMF parameter control mechanism is: wherein, is the feedback adjusted CMF parameter set, is the current CMF parameter set, is the feedback adjustment step size, is the gradient of the consistency detection result on the parameter, is the linear rectifier function, is the consistency threshold.

[0011] Further, the S3 comprises the following sub-steps: S31, extracting the calibration characteristic items of the material physical property parameters in the CMF calibration parameter set, including the material surface roughness related parameters, the material hardness related parameters, and the material thermal conductivity related parameters, and determining the value range and distribution law of each characteristic item; S32, based on the coupling relationship between the color space parameters and the material physical property parameters, establishing a color parameter response model for material characteristics, and analyzing the change trend and correlation strength of the color parameters under different material parameters; S33, combining the controllable range of the process execution parameters, integrating the material characteristic items and the color response relationship into a collaborative optimization framework, and adjusting the value combination of the process parameters to form an adaptive matching of material, color, and process parameters; S34, preliminarily evaluating the optimization effect under different parameter combinations, screening out the parameter combinations that meet the basic adaptation requirements, and forming the optimized target CMF parameter set.

[0012] Further, the S4 comprises the following sub-steps: S41, collecting product appearance images through an appearance defect detection device, extracting defect feature parameters including shape feature parameters, position feature parameters, and gray scale feature parameters, and constructing an appearance defect feature parameter matrix; S42, performing correlation analysis on the defect feature parameter matrix and the CMF calibration parameter set, positioning the CMF parameter deviation items with strong correlation with the defects by calculating the correlation coefficients of the defect features and the CMF parameters; S43, based on the correlation analysis result, determining the correction direction of the deviation parameters, combining the output of the appearance defect traceability and the CMF calibration model, and formulating a targeted parameter correction strategy; S44, performing gradient adjustment on the deviation parameters according to the correction strategy, and monitoring the changes of other associated parameters in the adjustment process to avoid the generation of new parameter deviations, and completing the calibration and correction of the CMF parameters.

[0013] Further, the S5 comprises the following sub-steps: S51, converting the calibrated and corrected CMF parameter set into a parameter format recognizable by an industrial product appearance consistency verification platform, and uploading the parameters according to a preset partition rule; S52, calling a built-in appearance feature comparison library of the platform, the comparison library including standard appearance CMF parameter templates and feature matching rules of different specifications of products, and comparing the uploaded parameters with the standard templates dimension by dimension; S53, calculating the matching degree values of each dimension parameter, and comprehensively weighting the matching degree values according to the feature matching rules to obtain an overall consistency score; S54, generating detection result data including parameter matching details, deviation position identification, and consistency level according to the overall consistency score and the dimension matching results, and synchronously storing the detection result data to a platform database.

[0014] The application discloses an industrial product appearance consistency implementation method fusing CMF parameter control, which is realized by different units and comprises the following steps: a multi-source CMF parameter acquisition and preprocessing unit is used for collecting material physical characteristic parameters, color space parameters and process execution parameters of an industrial product through a multi-dimensional sensor array, forming a multi-source CMF parameter matrix and performing format regularization; a multi-dimensional CMF parameter mapping fusion unit is connected with the multi-source CMF parameter acquisition and preprocessing unit, and is internally provided with a multi-dimensional CMF parameter mapping fusion model, so as to perform cross-dimension correlation mapping on the multi-source CMF parameter matrix and output a CMF calibration parameter set; a material color process collaborative optimization unit is connected with the multi-dimensional CMF parameter mapping fusion unit, a material color process collaborative optimization model is used for coupling relationship adjustment on the CMF calibration parameter set, and a target CMF parameter set is generated; an appearance defect tracing and CMF calibration unit is connected with the material color process collaborative optimization unit, and a parameter deviation item is located and calibration correction is completed based on an appearance defect tracing and CMF calibration model; an industrial product appearance consistency verification unit is connected with the appearance defect tracing and CMF calibration unit, an appearance feature comparison library is constructed, multi-dimensional consistency detection is performed, and detection result data are output; and a CMF parameter closed-loop feedback adjustment unit is connected with the industrial product appearance consistency verification unit, the multi-source CMF parameter acquisition and preprocessing unit, the multi-dimensional CMF parameter mapping fusion unit and the material color process collaborative optimization unit respectively, and parameter control processes of the units are dynamically adjusted according to the detection result data, so as to stably output the industrial product appearance consistency.

[0015] Beneficial effects: The application discloses an industrial product appearance consistency implementation method fusing CMF parameter control, multi-dimensional sensor arrays are used for comprehensively collecting material, color and process related parameters, a multi-dimensional CMF parameter mapping fusion model is used for realizing cross-dimension parameter deep fusion, the limitation of separately processing CMF parameters in the traditional technology is broken, the coupling correlation characteristics among parameters are fully considered, a core parameter set is accurately adapted, the problem that a multi-dimensional parameter collaborative optimization mechanism is lacked in the prior art is solved, the core parameters are adjusted by a material color process collaborative optimization model, a direct mapping relationship between defects and parameter deviations is established by combining an appearance defect tracing and CMF calibration model, parameter problems causing appearance inconsistency are located and corrected from the root, and the short board that the correlation between defect tracing and parameter calibration in the prior art is insufficient is made up. Meanwhile, an industrial product appearance consistency verification platform provides multi-dimensional detection support, a closed-loop feedback adjustment mechanism is used for realizing parameter dynamic optimization, product appearance stability in batch production is ensured, the fine level and reliability of industrial product appearance control are greatly improved, and a systematic technical scheme is provided for solving the industrial appearance consistency control problem. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1This is a flowchart illustrating the overall steps of the method of the present invention;

[0017] Figure 2 This is a flowchart of method step S3 of the present invention;

[0018] Figure 3 This is a flowchart of method step S4 of the present invention;

[0019] Figure 4 This is a flowchart of step S5 of the method of the present invention;

[0020] Figure 5 This is a diagram showing the unit composition for implementing the method of the present invention. Detailed Implementation

[0021] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0022] like Figure 1 As shown, a method for achieving consistent appearance of industrial products by integrating CMF parameter control includes the following steps:

[0023] S1. Collect the CMF basic parameter set of industrial products. The CMF basic parameter set includes material physical property parameters, color space parameters, and process execution parameters. Collect parameters from different appearance areas of the product through a multi-dimensional sensor array to form a multi-source CMF parameter matrix.

[0024] Specifically, step S1 involves comprehensive and accurate collection of basic CMF (Color, Material, Finish) parameters for industrial products, providing data support for subsequent parameter fusion and optimization. During implementation, a multi-dimensional sensor array consisting of 32 high-precision sensors is used, including different types such as material physical property detection sensors, color spectrum sensors, and process parameter acquisition sensors. Data is collected in 16 key areas of the product's appearance, including the front, sides, and corners. Material physical property parameters include eight core parameters such as surface roughness, hardness, density, and refractive index, with 30 sets of data collected for each parameter to ensure data stability. Color space parameters select three mainstream color spaces: RGB, Lab, and HSV, collecting six sub-parameters such as brightness, hue, and saturation, with an acquisition accuracy of 0.01 units. Process execution parameters include 12 key process parameters such as injection temperature, spraying pressure, curing time, and grinding speed, collected at a frequency of 10 times per second. By synchronously collecting data from the sensor array, parameters of different regions and types are classified and organized according to region number and parameter type, and a multi-source CMF parameter matrix with a dimension of 16×26 is constructed. The rows represent product appearance areas and the columns represent various CMF parameters, ensuring that all parameters in each region are completely recorded, laying the data foundation for subsequent cross-dimensional parameter mapping and fusion.

[0025] S2, input the multi-source CMF parameter matrix into a multi-dimensional CMF parameter mapping fusion model, and perform cross-dimensional correlation mapping on the material, color and process parameters through the model to generate a fused CMF calibration parameter set;

[0026] Specifically, step S2 realizes cross-dimensional correlation and deep fusion of multi-source CMF parameters through a multi-dimensional CMF parameter mapping fusion model, and solves the information fragmentation problem caused by single parameter independent processing. In implementation, the 16x26 multi-source CMF parameter matrix constructed in step S1 is first subjected to data format standardization processing to ensure that the numerical ranges of the material, color and process parameters are in the same order of magnitude. Three layers of processing logic are set in the model. The first layer performs linear correlation operation on the material physical property parameters and the color space parameters to explore the internal relationship between the two in product appearance presentation. The second layer establishes indirect correlation between the process execution parameters and the material and color parameters through a nonlinear mapping algorithm to analyze the influence law of the process parameters on material performance and color presentation. The third layer uses a feature fusion algorithm to integrate the correlation results of the previous two layers and the original parameter matrix. In the fusion process, the model sets a weight coefficient of 0.35 for the material parameters, a weight coefficient of 0.35 for the color parameters, and a weight coefficient of 0.3 for the process parameters, and generates a 16x12 CMF core parameter set through weighted calculation. The parameter set includes key fusion information such as material-color collaborative features and process-appearance adaptation features, which not only retains the core characteristics of the original parameters, but also eliminates parameter redundancy, providing accurate core data support for subsequent collaborative optimization.

[0027] S3, input the CMF calibration parameter set into a material color process collaborative optimization model, and perform collaborative adjustment based on the coupling relationship between the parameters to output an optimized target CMF parameter set;

[0028] Specifically, step S3 uses a material, color, and process co-optimization model to make targeted adjustments based on the coupling relationship between CMF parameters, generating a target parameter set that meets the product's appearance consistency requirements. During implementation, the coupling relationship analysis of the CMF core parameter set output in step S2 is first performed, identifying 18 key coupling parameter pairs, such as material surface roughness and spraying process pressure, color brightness and curing time, and determining the correlation strength coefficient of each parameter pair, ranging from 0.1 to 0.9. Based on the preset standards for product appearance consistency, the model sets parameter optimization target thresholds, with allowable deviations for material parameters at ±5%, color parameters at ±3%, and process parameters at ±7%. The optimization process employs an iterative adjustment strategy. The first round of adjustments targets strongly coupled parameter pairs with correlation strength coefficients greater than 0.7, using a gradient descent algorithm to adjust process parameters to match material and color parameter requirements. The second round of adjustments targets moderately coupled parameter pairs with correlation strength coefficients between 0.4 and 0.7, simultaneously fine-tuning material and process parameters to ensure stable color presentation. The third round of adjustments targets weakly coupled parameter pairs with correlation strength coefficients less than 0.4, performing local parameter calibration. After three rounds of iterative adjustments, the optimized target CMF parameter set is output. All parameters in this parameter set meet the preset threshold requirements, and the coupling relationship between the parameters reaches the optimal adaptation state, providing parameter assurance for the consistency of product appearance.

[0029] S4. By tracing the source of appearance defects and using the CMF calibration model, a defect correlation analysis is performed on the optimized target CMF parameter set to locate parameter deviation items that may lead to inconsistent appearance and to perform calibration correction.

[0030] Specifically, step S4 utilizes the appearance defect tracing and CMF calibration model to locate parameter deviations in the target CMF parameter set that cause appearance inconsistencies and completes calibration correction, eliminating potential appearance inconsistencies at their source. In implementation, firstly, 1000 standard appearance images and images with appearance defects of similar products are acquired using a high-resolution industrial camera. The shape, location, grayscale, and other feature parameters of the defects are extracted to construct an appearance defect feature library. The target CMF parameter set output in step S3 is then correlated with the defect feature library, calculating the correlation probability between each parameter and various types of defects. Parameters with a correlation probability greater than 0.6 are identified as deviations that may lead to appearance defects. For the located deviations, the model analyzes their deviation types, including systematic deviations and random deviations. Systematic deviations are calibrated by correcting the parameter baseline value, while random deviations are controlled by adjusting the parameter fluctuation range. During calibration, the defect tracing coefficient was set to 0.8, and the calibration adjustment coefficient to 0.2. Based on the correlation probability and magnitude of the deviation items, a differentiated calibration strategy was implemented: deviation items with a correlation probability between 0.6 and 0.7 received a 10% gradient correction; those with a correlation probability between 0.7 and 0.8 received a 15% gradient correction; and those with a correlation probability greater than 0.8 received a 20% gradient correction. Changes in the correlation parameters were monitored in real time during calibration to prevent new deviations. The final output was a calibrated and corrected CMF parameter set, ensuring that all parameters mitigated the risk of appearance defects.

[0031] S5. Input the calibrated and corrected CMF parameter set into the industrial product appearance consistency verification platform, and perform multi-dimensional consistency detection through the appearance feature comparison library built by the platform to generate detection result data.

[0032] Specifically, in step S5, the calibrated CMF parameter set is multi-dimensionally detected through an industrial product appearance consistency verification platform to generate comprehensive test result data. During the implementation process, the platform first constructs an appearance feature comparison library including 10,000 groups of standard product CMF parameters, including standard parameter templates for products of different production batches and different specifications, which are divided into 5 major categories and 20 minor categories according to product types. The calibrated and corrected CMF parameter set is classified and uploaded according to 16 preset appearance regions. The platform calls the standard parameter templates of the corresponding types in the comparison library and performs detection by means of comparison region by region and parameter by parameter. The detection dimensions include three core dimensions: parameter value matching degree, parameter change trend consistency, and parameter coupling relationship adaptation degree. Among them, the parameter value matching degree is calculated using the absolute error, the parameter change trend consistency is calculated using the correlation coefficient, and the parameter coupling relationship adaptation degree is calculated using the deviation rate. The test data for each region are comprehensively weighted according to the weights of 30%, 40%, and 30% to obtain the region consistency score; then the scores of the 16 regions are calculated to obtain the overall consistency score according to the uniform weight. The scoring range is from 0 to 100 points, and a score of 85 or above is judged to be qualified for consistency. After the detection is completed, the platform generates test result data including the parameter matching details of each region, deviation position identification, consistency level, and list of unqualified parameters, and synchronously stores them in the platform database to provide a detailed basis for subsequent feedback and adjustment.

[0033] S6. According to the test result data, dynamically feedback and adjust the CMF parameter control process to form a closed-loop CMF parameter control mechanism, and complete the realization of industrial product appearance consistency.

[0034] Specifically, step S6 constructs a closed-loop CMF parameter control mechanism based on the test result data to achieve dynamic optimization of the parameter control process and ensure the stable realization of consistent appearance of industrial products. During implementation, the test result data output from step S5 is first subjected to hierarchical analysis to extract key deviation items from the list of non-conforming parameters. The frequency and magnitude of deviations for various parameters are statistically analyzed. Parameters with a deviation frequency exceeding 5 times per batch and a deviation magnitude exceeding 10% are identified as key parameters requiring significant adjustment. Based on the analysis results, targeted feedback adjustment strategies are formulated: for material parameter deviations, the sampling frequency and number of sampling points of the sensor are adjusted, increasing the sampling frequency to 12 times per second and the number of sampling points to 40 sets; for color parameter deviations, the color parameter weight coefficient of the multi-dimensional CMF parameter mapping and fusion model is optimized to 0.4; for process parameter deviations, the iteration number of the material, color, and process co-optimization model is adjusted to 4 rounds to expand the adjustment range of strongly coupled parameter pairs. The adjusted parameter control process is applied to the production of the next batch of products, and the process from steps S1 to S5 is repeated for testing and verification. The parameter control strategy is further fine-tuned based on the new test results. Through a closed-loop CMF parameter control mechanism formed by iterative processes of "collection-fusion-optimization-calibration-detection-adjustment", the product appearance consistency pass rate is kept stable at over 98%, continuously improving the control level and stability of industrial product appearance consistency.

[0035] Preferably, the expression for the multi-dimensional CMF parameter mapping fusion model is: ,in, For the fused CMF calibration parameter set, For mapping and fusion weight coefficients, This is a matrix of physical property parameters of the material. A color space parameter vector, For the process execution parameter matrix, For activation function, For gradient computation operators, This is a three-dimensional convolution operator.

[0036] Specifically, the multi-dimensional CMF parameter mapping and fusion model breaks the independent dimensional limitations of material, color, and process parameters, achieving deep correlation and integration of cross-dimensional information, and providing accurate core parameters for subsequent optimization. During implementation, the model first receives the multi-source CMF parameter matrix generated in step S1. This matrix includes the material physical properties, color space, and process execution-related parameters for each appearance area of ​​the product. The model sets three mapping and fusion weight coefficients, each ranging from 0.1 to 0.9, with a sum of 1. Different fusion weights are assigned to material, color, and process parameters through preliminary experimental calibration, ensuring that key parameters dominate the fusion process. The model uses an activation function to perform a non-linear transformation on the product of color space parameters and process execution parameters, strengthening the non-linear correlation between parameters; it uses a gradient calculation operator to solve for the gradient information of material physical property parameters and color space parameters, reflecting the relationship between their rates of change; and it uses a three-dimensional convolution operator to extract and fuse spatial dimensions of the three types of parameters (material, color, and process), integrating the spatial correlation information of multi-dimensional parameters. Finally, through weighted summation, the three types of operation results are integrated to generate the CMF core parameter set. This parameter set not only retains the core characteristics of various original parameters, but also eliminates parameter redundancy and information conflicts. It simplifies the parameter dimension from the high dimension of the multi-source matrix to a reasonable range that is suitable for subsequent optimization, improves the efficiency and accuracy of parameter processing, and lays the foundation for the collaborative optimization of material color process.

[0037] Preferably, the expression for the material, color, and process collaborative optimization model is: ,in, For the optimized target CMF parameter set, This is the process parameter deviation adjustment coefficient. This is the initial process execution parameter matrix. For the collaborative optimization coefficient, These are the KL divergence weighting coefficients. For tensor product operations, Principal component analysis operator, Operator for calculating KL divergence. This is a reference process execution parameter matrix.

[0038] Specifically, the material, color, and process co-optimization model, based on the coupling and correlation characteristics between CMF parameters, makes targeted adjustments to the core parameter set to generate a target parameter set that meets the appearance consistency requirements. During implementation, the model first imports the CMF core parameter set output in step S2, and simultaneously loads the initial process execution parameter matrix and the reference process execution parameter matrix. The reference process execution parameter matrix is ​​obtained based on statistical analysis of process parameters from historical qualified products, including the optimal parameter range for each process step. The model sets a process parameter deviation adjustment coefficient, ranging from 0.3 to 1.2, to adjust the impact of process parameter deviation on the overall optimization result; a co-optimization coefficient, ranging from 0.5 to 1.5, to balance the weight of the tensor product operation result in the optimization; and a KL divergence weight coefficient, ranging from 0.2 to 0.8, to control the constraint strength of the difference between the process execution parameters and the reference parameters on the optimization result. The model first quantifies the initial deviation of the process parameters by calculating the norm of the process execution parameter matrix and the initial process parameter matrix. Then, it integrates the coupling information of material, color, and process parameters through tensor product operation and extracts key coupling features through principal component analysis. At the same time, it calculates the distribution difference between the process parameters and the reference parameters through KL divergence. Finally, it sums the three results by coefficient weight to obtain the optimized target CMF parameter set, ensuring that the material, color, and process parameters form an optimal fit relationship and avoiding appearance deviations caused by adjusting a single parameter.

[0039] Preferably, the expression for the appearance defect tracing and CMF calibration model is: ,in, To calibrate the corrected CMF parameter set, This is the defect tracing coefficient. To calibrate the adjustment coefficient, This is the matrix of characteristic parameters of appearance defects. Find the inverse operator for a matrix. The CMF parameter Jacobian matrix, It is an operator for long short-term memory networks.

[0040] Specifically, the appearance defect tracing and CMF calibration model establishes a mapping relationship between appearance defects and CMF parameter deviations, locating and correcting parameter problems at their root to ensure parameter accuracy. During implementation, the model first acquires the target CMF parameter set output in step S3, and simultaneously imports the appearance defect feature parameter matrix. This matrix is ​​obtained by extracting features such as the shape, location, and grayscale of product appearance defects, including quantitative feature information for various defects. The model sets a defect tracing coefficient, ranging from 0.6 to 0.9, to strengthen the correlation between defect features and parameter deviations; the calibration adjustment coefficient ranges from 0.1 to 0.4 to control the adjustment range during the calibration process and avoid over-correction. The model calculates the Jacobian matrix of CMF parameters to reflect the sensitivity of parameter changes to product appearance, and then performs an inverse operation on the Jacobian matrix to determine the direction and basic magnitude of parameter correction. Simultaneously, it utilizes a Long Short-Term Memory (LSTM) network operator to mine the temporal correlation between appearance defect features and the core CMF parameter set, capturing the dynamic changes in defects caused by parameter deviations. By subtracting the product of defect features and the inverse of the Jacobian matrix (weighted by the traceability coefficient) from the CMF core parameter set, and adding the results of the long short-term memory network operation (weighted by the calibration adjustment coefficient), the calibrated and corrected CMF parameter set is obtained. This enables precise correction of parameter deviations that cause inconsistencies in appearance, while avoiding the generation of new parameter conflicts during the correction process.

[0041] Preferably, the consistency detection expression of the industrial product appearance consistency verification platform is: ,in, For consistency test results data, Let be the CMF parameter vector of the i-th detection region. For reference CMF parameter vector, The number of detection areas. Cosine similarity weights To determine the weights for intersection and union comparisons, To calculate the cosine similarity operator, To compute the intersection-union-comparison operator, For the CMF parameter set obtained from the detection, This is the standard CMF parameter set.

[0042] Specifically, the consistency detection expression of the industrial product appearance consistency verification platform is used to construct a standardized, multi-dimensional detection system and objectively quantify the product appearance consistency level corresponding to the calibrated parameter set. During implementation, the platform first receives the calibrated and corrected CMF parameter set, and divides the product appearance into several detection areas according to preset detection area division rules. Each area corresponds to a set of CMF parameter vectors. The platform's built-in appearance feature comparison library includes a large number of qualified product standard CMF parameter templates of different specifications and batches. Simultaneously, cosine similarity weight and intersection-union ratio (IUU) weight are set, both ranging from 0.3 to 0.7, and their sum is 1. The optimal ratio of these two weights is determined through experimental calibration to balance the detection weights of parameter numerical matching and feature space overlap. During testing, the cosine similarity between the CMF parameter vector of each tested region and the reference CMF parameter vector is first calculated, reflecting the degree of agreement between the parameter values. Then, the intersection-union ratio (IUR) of the detected CMF parameter set and the standard CMF parameter set is calculated, reflecting the overlap range of the parameter feature space. The cosine similarity of all tested regions is averaged and weighted according to the cosine similarity weight. Simultaneously, the IUR is weighted according to its weight, and the sum of these two values ​​yields the consistency test result data. This test result data comprehensively reflects the consistency level of the product appearance in both parameter values ​​and feature space dimensions, providing a quantitative basis for subsequent feedback and adjustments, ensuring the comprehensiveness and objectivity of the test results.

[0043] Preferably, the feedback adjustment expression of the closed-loop CMF parameter control mechanism is: ,in, To provide feedback on the adjusted CMF parameter set, For the current CMF parameter set, To adjust the step size based on feedback, The gradient of the consistency detection result with respect to the parameters. It is a linear rectified function. This is the consistency threshold.

[0044] Specifically, the feedback adjustment expression of the closed-loop CMF parameter control mechanism is used to construct a cyclic optimization system of "detection-adjustment-re-detection" to achieve dynamic optimization of the CMF parameter control process. During implementation, the model first acquires the consistency detection result data output from step S5 and the current CMF parameter set. Simultaneously, it sets the feedback adjustment step size and the consistency threshold. The feedback adjustment step size ranges from 0.05 to 0.2, used to control the magnitude of each parameter adjustment to avoid system instability due to excessive adjustment. The consistency threshold is set according to the product appearance quality requirements, with a specific score from 0 to 100, serving as the standard for judging whether the appearance consistency is qualified. The model calculates the gradient of the consistency detection result with respect to the current CMF parameter set to reflect the rate of influence of parameter changes on the consistency detection result, clarifying the sensitive direction of parameter adjustment. Then, a linear rectification function is used to determine the difference between the current consistency detection result and the threshold. If the detection result is lower than the threshold, the function outputs a non-negative value, triggering parameter adjustment; if the detection result is higher than the threshold, the function outputs 0, stopping the adjustment. The current CMF parameter set is multiplied by the feedback adjustment step size, gradient, and the output of the linear rectified function to obtain the feedback-adjusted CMF parameter set. This parameter set is then fed back to the preceding stages such as parameter acquisition, fusion, and optimization to adjust the parameter processing strategies at each stage. Through this dynamic feedback adjustment mechanism, the CMF parameter control process is continuously optimized, ensuring that the product appearance consistency remains stable above the qualified level, forming a virtuous cycle of closed-loop management.

[0045] Preferred, such as Figure 2 As shown, step S3 includes the following sub-steps: S31, extracting calibration feature items of material physical property parameters from the CMF calibration parameter set, including material surface roughness related parameters, material hardness related parameters, and material thermal conductivity related parameters, and clarifying the value range and distribution law of each feature item; S32, based on the coupling relationship between color space parameters and material physical property parameters, establishing a response model of color parameters to material properties, and analyzing the changing trend and correlation strength of color parameters under different material parameters; S33, combining the controllable range of process execution parameters, integrating material feature items and color response relationships into a collaborative optimization framework, and adjusting the value combination of process parameters to achieve a suitable match between material, color, and process parameters; S34, conducting a preliminary evaluation of the optimization effect under different parameter combinations, selecting parameter combinations that meet the basic adaptation requirements, and forming the optimized target CMF parameter set.

[0046] Specifically, step S3, the material color process co-optimization process, achieves precise adaptation and optimization of CMF core parameters through four sub-steps, ensuring optimal matching of the coupling relationship between parameters. S31 first extracts eight key physical property features of the material from the CMF core parameter set, including surface roughness, hardness, density, and refractive index. Statistical analysis determines the value range of each feature, with surface roughness ranging from 0.02 to 3.5, hardness from 15 to 60, and density from 0.9 to 19.3. Simultaneously, the mean, variance, and extreme values ​​of each feature are calculated to form a material feature distribution report. S32, based on the inherent correlation between color space parameters and material physical property parameters, constructs a response model of color parameters to material properties. Twelve sets of different material parameter values ​​are selected for color parameter testing, recording the changes in brightness, hue, and saturation under each set of material parameters. Correlation analysis quantifies the relationship between the two. The strength and correlation strength coefficient range from 0.1 to 0.9. S33, combining the controllable range of process execution parameters, integrates the defined material characteristics and color response relationships into a pre-defined collaborative optimization framework. The adjustment step size of the process execution parameters is set to 0.05. By iterating through 50 sets of process parameter value combinations, the adaptability of material and color parameters under each combination is analyzed. S34 performs a preliminary evaluation of the optimization effect under different parameter combinations, setting two core evaluation indicators: parameter adaptability and fluctuation stability, each with a weight of 50%. Parameter combinations with adaptability scores higher than 85 and fluctuation stability scores higher than 80 are selected, and finally integrated to form the optimized target CMF parameter set, providing a precise parameter basis for subsequent calibration and correction stages.

[0047] Preferred, such as Figure 3 As shown, step S4 includes the following sub-steps: S41, acquiring product appearance images through appearance defect detection equipment, extracting defect feature parameters, including shape feature parameters, location feature parameters, and grayscale feature parameters, and constructing an appearance defect feature parameter matrix; S42, performing correlation analysis between the defect feature parameter matrix and the CMF calibration parameter set, and locating CMF parameter deviation terms that are strongly correlated with the defects by calculating the correlation coefficient between the defect features and each CMF parameter; S43, determining the correction direction of the deviation parameters based on the correlation analysis results, and formulating a targeted parameter correction strategy by combining the appearance defect tracing and the output of the CMF calibration model; S44, performing gradient adjustment of the deviation parameters according to the correction strategy, while monitoring the changes of other related parameters during the adjustment process to avoid generating new parameter deviations, and completing the calibration and correction of the CMF parameters.

[0048] Specifically, step S4, the appearance defect tracing and CMF calibration process, achieves precise location and correction of parameter deviations through four sub-steps, eliminating potential inconsistencies in appearance at their source. S41 utilizes a high-resolution appearance defect detection device with a resolution of 50 megapixels to acquire product appearance images, covering six directions including front, side, and oblique angles. Twenty images are acquired for each product. Image feature extraction algorithms are used to extract 10 feature parameters of the defects, including shape, location, and grayscale. Shape features include defect area, perimeter, and aspect ratio; location features include defect coordinates and distance from the edge; and grayscale features include the mean and variance of the grayscale value of the defect area. A 20×10 dimension appearance defect feature parameter matrix is ​​constructed. S42 performs correlation analysis between the defect feature parameter matrix and the CMF core parameter set, calculating the Pearson correlation coefficient between the 10 defect features and 26 CMF parameters. The correlation coefficient ranges from -1 to 1, and values ​​with an absolute value greater than 0.6 are considered valid. The CMF parameters corresponding to the correlation coefficients are identified as parameter deviation terms strongly correlated with defects. Based on the correlation analysis results, S43 determines the correction direction for the deviation parameters. Combining the source tracing of appearance defects with the output of the CMF calibration model, calibration strategies are formulated for systematic and random deviations respectively. Systematic deviations are corrected using a benchmark value method, while random deviations are corrected using a fluctuation range adjustment method. S44 adjusts the deviation parameters according to the correction strategy, setting three correction gradients of 10%, 15%, and 20%. The corresponding correction gradient is determined based on the correlation strength coefficient of the deviation terms. During the correction process, the change amplitude of 15 correlated parameters is monitored simultaneously for each parameter adjustment to ensure that the change amplitude of correlated parameters does not exceed 5%, avoiding the generation of new parameter deviations. Finally, the calibration and correction of the CMF parameters are completed.

[0049] Preferred, such as Figure 4 As shown, step S5 includes the following sub-steps: S51, converting the calibrated and corrected CMF parameter set into a parameter format recognizable by the industrial product appearance consistency verification platform, and classifying and uploading the parameters according to preset partitioning rules; S52, calling the platform's built-in appearance feature comparison library, which includes standard appearance CMF parameter templates and feature matching rules for products of different specifications, and comparing the uploaded parameters with the standard templates dimension by dimension; S53, calculating the matching degree values ​​of parameters in each dimension, and performing a comprehensive weighted calculation of the matching degree values ​​in combination with the feature matching rules to obtain an overall consistency score; S54, generating detection result data including parameter matching details, deviation location identifiers, and consistency levels based on the overall consistency score and the matching results of each dimension, and synchronously storing it in the platform database.

[0050] Specifically, step S5, the industrial product appearance consistency verification process, achieves multi-dimensional and high-precision consistency detection through four sub-steps, providing a quantitative basis for feedback and adjustment. S51 classifies the calibrated and corrected CMF parameter set into 16 preset appearance regions, each region corresponding to 12 core CMF parameters. A data format conversion algorithm converts the parameters into a standard format recognizable by the industrial product appearance consistency verification platform, maintaining parameter accuracy to four decimal places during the conversion process. The parameters are uploaded to the platform sequentially according to region number, with an upload rate of 20 parameters per second. S52 calls the platform's built-in appearance feature comparison library, which includes 10,000 sets of standard CMF parameter templates for qualified products of different specifications and production batches, categorized into 5 major categories and 20 subcategories according to product type. The corresponding standard template is matched based on the type of the product being tested, while simultaneously loading 15 preset feature matching rules, including parameter value matching rules, change trend matching rules, and coupling relationship matching rules. S53 calculates each dimension... The matching degree values ​​of the parameters are calculated using the absolute error method, the trend matching degree using the correlation coefficient method, and the coupling relationship matching degree using the deviation rate method. These three matching degree values ​​are weighted at 30%, 40%, and 30% respectively to obtain a consistency score for each region, ranging from 0 to 100. S54 calculates the overall consistency score based on the consistency scores of the 16 regions using the arithmetic mean method. Simultaneously, it statistically analyzes the parameter deviation location, deviation value, and deviation type for each region, generating test result data including parameter matching details, deviation location identifiers, consistency level, and a list of non-compliant parameters. The test result data is categorized by timestamp and stored in the platform database, with a storage capacity supporting the long-term storage of 1 million sets of test data, providing comprehensive and detailed testing basis for subsequent feedback and adjustments.

[0051] like Figure 5As shown, a method for achieving consistent appearance of industrial products by integrating CMF parameter control is implemented through different units, including: a multi-source CMF parameter acquisition and preprocessing unit, used to acquire material physical property parameters, color space parameters, and process execution parameters of industrial products through a multi-dimensional sensor array, forming a multi-source CMF parameter matrix and formatting it; a multi-dimensional CMF parameter mapping and fusion unit, connected to the multi-source CMF parameter acquisition and preprocessing unit, with a built-in multi-dimensional CMF parameter mapping and fusion model, performing cross-dimensional correlation mapping on the multi-source CMF parameter matrix and outputting a CMF calibration parameter set; and a material, color, and process collaborative optimization unit, connected to the multi-dimensional CMF parameter mapping and fusion unit, which couples the CMF calibration parameter set through the material, color, and process collaborative optimization model. The system adjusts the relationship to generate a target CMF parameter set; the appearance defect tracing and CMF calibration unit, connected to the material, color, and process collaborative optimization unit, locates parameter deviation items and completes calibration correction based on the appearance defect tracing and CMF calibration model; the industrial product appearance consistency verification unit, connected to the appearance defect tracing and CMF calibration unit, constructs an appearance feature comparison library and performs multi-dimensional consistency detection, outputting detection result data; the CMF parameter closed-loop feedback adjustment unit, connected to the industrial product appearance consistency verification unit, the multi-source CMF parameter acquisition and preprocessing unit, the multi-dimensional CMF parameter mapping and fusion unit, and the material, color, and process collaborative optimization unit, dynamically adjusts the parameter control process of each unit according to the detection result data, and achieves stable output of industrial product appearance consistency.

[0052] The formulas in this invention can integrate different scalar and vector parameters for calculation. The core lies in eliminating computational conflicts caused by differences in parameter types and dimensions through standardization, dimension adaptation mechanisms, and operator bridging design. Although scalar parameters (such as weighting coefficients and adjustment coefficients) are only single values, they can be applied to vector or matrix-form CMF parameters (such as material property parameter matrices and color space parameter vectors) through weighted operations, achieving quantitative control of multi-dimensional data. For example, scalar weighting coefficients, through multiplication with vector parameters, permeate the influence of single values ​​into multi-dimensional parameter combinations. Simultaneously, the activation functions and gradient calculation operators in the formulas possess dimension transformation capabilities, mapping the results of vector parameter calculations to dimensions suitable for subsequent calculations. For instance, the product of a color space vector and a process parameter matrix can be converted through an activation function into an output consistent with the dimensions of the material parameter matrix, enabling parameters of different structures to have a basis for collaborative computation. In addition, operators such as tensor product and convolution can directly process multi-dimensional parameters. For example, tensor product can be used to integrate material scalar features, color vector parameters, and process matrix parameters into a tensor of a unified dimension. Then, principal component analysis can be used to extract core features, thereby achieving deep fusion calculation of different types of parameters.

[0053] The essence of different scalar and vector parameters is to quantify CMF characteristics. The formula establishes the rationality of the operation between parameters through the correlation of physical meaning and the adaptation of mathematical logic. For example, the defect origination coefficient in scalar form, the defect feature parameter in vector form, and the Jacobian matrix in matrix form can be operated together because all three revolve around the core logic of "correlation between defects and parameter deviations." The scalar coefficient regulates the correlation strength, the vector parameter provides defect feature information, and the matrix parameter reflects the sensitivity of the parameter's influence on appearance. The mathematical operation process is highly consistent with the physical meaning. At the same time, the parameters in the formula have all undergone prior calibration and normalization to ensure that the numerical magnitudes of scalar and vector parameters are in the same operation range, avoiding operational imbalances caused by numerical differences. For example, the scalar weight coefficient can be weighted and summed with the cosine similarity of the vector and the intersection-union ratio of the matrix because the similarity and intersection-union ratio are both mapped to the 0-1 interval, which is adapted to the value range of the scalar coefficient, making the operation results of different types of parameters additive. Ultimately, this achieves the effective integration and accurate calculation of multiple types of CMF parameters in the same formula.

[0054] A method for achieving consistent appearance in industrial products by integrating CMF (Color, Material, and Finish) parameter control is proposed. This method utilizes a multi-dimensional sensor array to comprehensively collect material, color, and process parameters. A dedicated mapping and fusion model establishes deep cross-dimensional parameter correlations, breaking the limitations of traditional single-parameter control and forming a precisely tailored core parameter system. A collaborative optimization model then specifically adjusts the parameter coupling relationships, and a defect tracing and calibration model establishes a direct correlation between appearance defects and parameter deviations, achieving refined control across the entire chain from parameter acquisition to optimization and calibration. Simultaneously, an industrial product appearance consistency verification platform provides multi-dimensional and standardized testing support, coupled with a closed-loop feedback mechanism to dynamically adjust parameters, significantly improving the accuracy and stability of appearance consistency control.

[0055] This method addresses the lack of a multi-dimensional parameter collaborative optimization mechanism for CMF (Color, Material, and Finish). It achieves cross-dimensional fusion and collaborative adjustment of material, color, and process parameters through multi-model linkage, fully considering the coupling and correlation characteristics between parameters, avoiding new deviations caused by adjusting parameters individually, and forming a more adaptable parameter combination. Addressing the insufficient correlation between defect tracing and parameter calibration, it establishes a mapping relationship between defect characteristics and parameter deviations through defect tracing and CMF calibration models. This allows for the root-cause identification and precise correction of parameter causes for inconsistent appearance. Combined with a closed-loop feedback adjustment mechanism, it achieves dynamic optimization of parameters during the production process, completely changing the limitations of traditional technologies that only treat surface defects, and ensuring the stable and consistent appearance of products in mass production.

[0056] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," "link," and "fix" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0057] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various equivalent changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for achieving consistent appearance of industrial products by integrating CMF parameter control, characterized in that, Includes the following steps: S1. Collect the CMF basic parameter set of industrial products. The CMF basic parameter set includes material physical property parameters, color space parameters, and process execution parameters. Collect parameters from different appearance areas of the product through a multi-dimensional sensor array to form a multi-source CMF parameter matrix. S2, input the multi-source CMF parameter matrix into the multi-dimensional CMF parameter mapping fusion model, and perform cross-dimensional correlation mapping of material, color and process parameters through the model to generate the fused CMF calibration parameter set; S3, input the CMF calibration parameter set into the material color process collaborative optimization model, perform collaborative adjustment based on the coupling relationship between parameters, and output the optimized target CMF parameter set; S4. By tracing the source of appearance defects and using the CMF calibration model, a defect correlation analysis is performed on the optimized target CMF parameter set to locate parameter deviation items that may lead to inconsistent appearance and to perform calibration correction. S5. Input the calibrated and corrected CMF parameter set into the industrial product appearance consistency verification platform, and perform multi-dimensional consistency detection through the appearance feature comparison library built by the platform to generate detection result data. S6. Based on the detection result data, the CMF parameter control process is dynamically adjusted to form a closed-loop CMF parameter control mechanism, thereby achieving the consistency of industrial product appearance. The expression for the multi-dimensional CMF parameter mapping fusion model is: ， in, For the fused CMF calibration parameter set, For mapping and fusion weight coefficients, This is a matrix of physical property parameters of the material. A color space parameter vector, For the process execution parameter matrix, For activation function, For gradient computation operators, This is a three-dimensional convolution operator.

2. The method for achieving consistent appearance of industrial products by integrating CMF parameter control according to claim 1, characterized in that, The expression for the material, color, and process collaborative optimization model is: ， in, For the optimized target CMF parameter set, This is the process parameter deviation adjustment coefficient. This is the initial process execution parameter matrix. For the collaborative optimization coefficient, These are the KL divergence weighting coefficients. For tensor product operations, Principal component analysis operator, Operator for calculating KL divergence. This is a reference process execution parameter matrix.

3. The method for achieving consistent appearance of industrial products by integrating CMF parameter control according to claim 2, characterized in that, The expression for the appearance defect tracing and CMF calibration model is as follows: ， in, To calibrate the corrected CMF parameter set, This is the defect tracing coefficient. To calibrate the adjustment coefficient, This is the matrix of characteristic parameters of appearance defects. Find the inverse operator for a matrix. The CMF parameter Jacobian matrix, It is an operator for long short-term memory networks.

4. The method for achieving consistent appearance of industrial products by integrating CMF parameter control according to claim 3, characterized in that, The consistency detection expression of the industrial product appearance consistency verification platform is as follows: ， in, For consistency test results data, Let be the CMF parameter vector of the i-th detection region. For reference CMF parameter vector, The number of detection areas. Cosine similarity weights To determine the weights for intersection and union comparisons, To calculate the cosine similarity operator, To compute the intersection-union-comparison operator, For the CMF parameter set obtained from the detection, This is the standard CMF parameter set.

5. The method for achieving consistent appearance of industrial products by integrating CMF parameter control according to claim 4, characterized in that, The feedback adjustment expression for the closed-loop CMF parameter control mechanism is: ， in, To provide feedback on the adjusted CMF parameter set, For the current CMF parameter set, To adjust the step size based on feedback, The gradient of the consistency detection result with respect to the parameters. It is a linear rectified function. This is the consistency threshold.

6. The method for achieving consistent appearance of industrial products by integrating CMF parameter control according to claim 5, characterized in that, S3 includes the following steps: S31, extract the calibration feature items of material physical property parameters from the CMF calibration parameter set, including material surface roughness related parameters, material hardness related parameters, and material thermal conductivity related parameters, and clarify the value range and distribution law of each feature item; S32. Based on the coupling relationship between color space parameters and material physical property parameters, a response model of color parameters to material properties is established to analyze the changing trend and correlation strength of color parameters under different material parameters. S33, combining the controllable range of process execution parameters, integrates material characteristics and color response relationships into the collaborative optimization framework, and adjusts the combination of process parameter values ​​to achieve a suitable match between material, color, and process parameters; S34. The optimization effect under different parameter combinations is initially evaluated, and parameter combinations that meet the basic adaptation requirements are selected to form the optimized target CMF parameter set.

7. The method for achieving consistent appearance of industrial products by integrating CMF parameter control according to claim 6, characterized in that, S4 includes the following steps: S41. Collect product appearance images through appearance defect detection equipment, extract defect feature parameters, including defect shape feature parameters, location feature parameters, and grayscale feature parameters, and construct an appearance defect feature parameter matrix. S42, perform correlation analysis between the defect feature parameter matrix and the CMF calibration parameter set, and locate the CMF parameter deviation terms that are strongly correlated with the defects by calculating the correlation coefficient between the defect features and each CMF parameter; S43. Based on the correlation analysis results, determine the correction direction of the deviation parameters, and combine the source tracing of appearance defects with the output of the CMF calibration model to formulate targeted parameter correction strategies. S44. Adjust the deviation parameters according to the correction strategy, and monitor the changes of related parameters during the adjustment process to avoid generating new parameter deviations and complete the calibration and correction of CMF parameters.

8. The method for achieving consistent appearance of industrial products by integrating CMF parameter control according to claim 7, characterized in that, S5 includes the following steps: S51 converts the calibrated and corrected CMF parameter set into a parameter format that can be recognized by the industrial product appearance consistency verification platform, and uploads the parameters according to the preset partitioning rules. S52, call the platform's built-in appearance feature comparison library, which includes standard appearance CMF parameter templates and feature matching rules for products of different specifications, and compare the uploaded parameters with the standard templates dimension by dimension; S53, calculate the matching degree values ​​of each dimension parameter, and perform a comprehensive weighted calculation of the matching degree values ​​in combination with feature matching rules to obtain the overall consistency score; S54 generates detection result data, including parameter matching details, deviation location identifiers, and consistency levels, based on the overall consistency score and matching results of each dimension, and stores it synchronously in the platform database.

9. A method for achieving consistent appearance of industrial products by integrating CMF parameter control according to any one of claims 1-8, characterized in that, This method is implemented through different units, including: The multi-source CMF parameter acquisition and preprocessing unit is used to acquire material physical property parameters, color space parameters, and process execution parameters of industrial products through a multi-dimensional sensor array, form a multi-source CMF parameter matrix, and perform formatting. A multi-dimensional CMF parameter mapping and fusion unit is connected to the multi-source CMF parameter acquisition and preprocessing unit. It has a built-in multi-dimensional CMF parameter mapping and fusion model to perform cross-dimensional correlation mapping of the multi-source CMF parameter matrix and output the CMF calibration parameter set. The material, color and process collaborative optimization unit is connected to the multi-dimensional CMF parameter mapping and fusion unit. It adjusts the coupling relationship of the CMF calibration parameter set through the material, color and process collaborative optimization model to generate the target CMF parameter set. The appearance defect tracing and CMF calibration unit is connected to the material color process co-optimization unit. Based on the appearance defect tracing and CMF calibration model, it locates parameter deviation items and completes calibration correction. An industrial product appearance consistency verification unit is connected to the appearance defect tracing and CMF calibration unit, which constructs an appearance feature comparison library and performs multi-dimensional consistency detection, and outputs detection result data. The CMF parameter closed-loop feedback adjustment unit is connected to the industrial product appearance consistency verification unit, the multi-source CMF parameter acquisition and preprocessing unit, the multi-dimensional CMF parameter mapping and fusion unit, and the material, color and process collaborative optimization unit, respectively. It dynamically adjusts the parameter control process of each unit according to the test result data to achieve stable output of industrial product appearance consistency.

Images