Method and system for determining material proportion formula of all colors of full-body ceramic tile
By acquiring tiles images for pixel-level analysis and clustering, combining genetic algorithms and digital twin technology to calculate the toner ratio, the problems of high cost and poor management of color materials in the color matching process of whole tiles are solved, and accurate feeding and reducing waste are achieved. Improves production efficiency and product consistency.
Patent Information
- Application Number
- CN202410918149.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-10
- Publication Date
- 2024-11-08
Smart Images

Figure BDA0004935781480000031 
Figure BDA0004935781480000041 
Figure BDA0004935781480000081
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of ceramic tile manufacturing technology, in particular to a method and a system for determining a formula of material proportions of various colors of a ceramic tile. Background Art
[0002] Full-body ceramic tiles are a type of ceramic tile widely used in modern architecture and decoration. Their notable feature is that the color and texture of the entire tile are completely consistent from the surface to the inside, and they have excellent wear resistance and anti-slip properties. Even if the surface of this tile is worn, it will not show a different color or texture from the inside. In addition, the surface of full-body ceramic tiles has been specially treated, which is easy to clean and maintain, has high anti-fouling ability, and can imitate the texture and color of various natural stones, with a high decorative effect.
[0003] It is very important to determine the proportion of each color of full-body ceramic tiles, because it directly affects the color consistency, material performance, cost control, production stability, environmental protection and safety of the ceramic tiles. Accurate formula ensures that the color of the ceramic tiles is consistent from the surface to the inside, ensuring the excellent aesthetics and physical properties, such as wear resistance and anti-slip. At the same time, a reasonable formula can optimize production costs, improve production efficiency and market competitiveness, ensure that products meet environmental standards, reduce harmful substances emissions, and meet the needs of different markets and customers. Through scientific formula design and strict production control, full-body ceramic tiles with excellent performance, beautiful appearance and strong market competitiveness can be produced.
[0004] In existing research, a Chinese patent with application number CN202110343291.4 proposes a method and device for determining the material ratio formula of each color of full-body ceramic tiles. Multiple full-body ceramic tile formula images are obtained and color separation is performed to extract and mark different color areas and areas. According to these areas and areas, the workstation fabric diagram corresponding to each design is allocated, and the material cutting ratio is set for cutting. Subsequently, the fabric parameters are set, the fabric is transmitted to the weighing equipment, the weight data of each workstation is detected, and the material ratio formula corresponding to each design is calculated. The method also involves an image acquisition method, including using image design software to design images or using a camera to acquire images, and setting the fabric diagram and cutting ratio by computer or custom. The color separation process uses a color recognition module and a controller to number different color areas and set workstations. When setting the fabric parameters, fabric parameters X, Y and Z are included, and the material cutting loss rate (0.5% to 5.0%) is considered. The Chinese patent with application number CN202110341926.7 proposes a method for calculating the material ratio of each color of multi-design full-body ceramic tiles. Similar to the above patent, first obtain an image of the material formula of the full-body ceramic tile, and use image design software or camera equipment to design and obtain it. Then perform color separation on the image, and extract and mark the areas and areas of different colors. According to the extracted color areas and areas, design the color matching cutting ratio and set the laying rules, complete the laying operations of each color area in turn, and set the cutting weight as the laying parameter. Next, the cutting weight and total weight of each color area are detected by the weighing system, and the material ratio of each color area is calculated. Finally, design the material ratio of multiple full-body ceramic tiles based on the calculation results, and calculate the weight of each color area and the weight of the added colorant. In contrast, the method for determining the material ratio formula of each color of full-body ceramic tiles focuses on the distribution of the material diagram and weighing detection of specific workstations to ensure accurate material cutting and weight monitoring at each workstation; the method for calculating the material ratio of each color of multi-design full-body ceramic tiles pays more attention to the overall color matching material ratio design and the setting of the material parameters, and realizes the multi-design color matching calculation by setting the material rules. The Chinese patent with application number CN202210191868.9 proposes a new type of refined material laying method for rock slab full-body ceramic tiles, which includes preparing design drawings, preparing powders of different colors, laying out the main hopper, line hopper and scraping hopper, performing reciprocating material laying, and pressing and forming the material cakes. This method sets up hoppers for monochrome materials in partitions and performs reciprocating material laying, which realizes the precise patterning of full-body ceramic tiles, so that the texture of the body material and the main texture of the surface glaze layer are naturally vertically aligned, with a deviation of less than 2mm. At the same time, the reciprocating cloth can make the texture at the same position overlap two colors, forming transition colors, marginal effects and dark details, and the dark and light colors correspond one to one, enhancing the overall beauty of the tile. The calculation method for the color matching material ratio in the above methods is complex and costly to implement, and the degree of automation needs to be improved.
[0005] In the existing color matching process of full-body tiles, there may be multiple options for the color matching scheme and proportion of the same color. The mixing and blending costs of colorants are high, resulting in unnecessary material loss, increasing production costs, and being unfavorable for the management of different colorants.
[0006] Therefore, a method and system for determining the proportion formula of each color material of a whole body ceramic tile are proposed. Summary of the invention
[0007] The object of the present invention is to provide a method and system for determining the material ratio formula of each color of a full-body ceramic tile, obtaining a surface image of a standard full-body ceramic tile sample and a plurality of cross-sectional images at different depth levels to form a first standard image set; performing pixel-level color analysis on the set to determine the color information of each pixel point and generate a comprehensive color distribution map; performing cluster analysis on the distribution map using a K-means algorithm to determine the main color tone and its ratio to obtain K main color tones and K main color tone ratios; according to the color of the toner, using a color decomposition technique, obtaining a color matching scheme corresponding to each main color tone, including at least one toner color selection scheme; calculating the ratio of each toner color using a color matching algorithm to match the main color tone, and generating a toner color ratio formula of each main color tone; dividing the distribution map into a plurality of grids, determining the ratio of the main color tone in each grid, and obtaining the main color tone ratio data of each grid; calculating the specific feeding amount of the toner color in each grid in combination with the K main color tone ratios according to the main color tone ratio data of the grid and the toner color ratio formula, and obtaining an overall feeding scheme; designing and implementing an automatic control system for managing a plurality of independent hoppers, each hopper being equipped with a control valve to ensure the accuracy of feeding.
[0008] To achieve the above object, the present invention provides the following technical solutions:
[0009] A method for determining a formula for the proportion of materials of each color of a full-body tile comprises:
[0010] Acquire a surface image of a standard full-body ceramic tile sample and a plurality of cross-sectional images at different depth levels to obtain a first standard image set;
[0011] Performing pixel-level color analysis on the first standard image set to determine the color information of each pixel and generate a comprehensive color distribution map;
[0012] Performing cluster analysis on the comprehensive color distribution map using a clustering algorithm to determine the main colors and their proportions in the comprehensive color distribution map, thereby obtaining K main colors and K main color proportions;
[0013] According to the color of the toner, using the color decomposition technology, a color matching scheme corresponding to each of the main colors is obtained, wherein the color matching scheme includes at least one toner color selection scheme;
[0014] According to the toner color selection scheme of each main color tone, a color matching algorithm is used to calculate the ratio of each toner color to match the main color tone, and a toner color ratio formula corresponding to each main color tone is obtained;
[0015] Divide the comprehensive color distribution map into a plurality of grids, each grid representing a small area, determine the proportion of each main color tone in each grid, and obtain main color tone ratio data of each grid;
[0016] According to the main color ratio data and the color ratio formula of the toner in each grid, and in combination with K main color ratios, the specific feeding amount of each toner color in each grid is calculated to obtain an overall feeding plan;
[0017] Design and implement an automated control system for managing a number of independent hoppers, each of which is equipped with a control valve.
[0018] Furthermore, the surface image and the cross-sectional images at multiple different depth levels of the standard full-body ceramic tile sample are obtained specifically as follows:
[0019] The standard full-body ceramic tile sample is placed in a shadowless light box and photographed using a high-resolution camera; a multi-angle diffuse light source is arranged in the shadowless light box.
[0020] Furthermore, before performing pixel-level color analysis on the first standard image set, cropping and white balance correction are performed on the first standard image set.
[0021] Furthermore, the clustering algorithm is used to perform cluster analysis on the comprehensive color distribution map to determine the main color tones and their proportions in the comprehensive color distribution map, and the K main color tones and K main color tones proportions are obtained as follows:
[0022] Using K-means algorithm to perform cluster analysis on the comprehensive color distribution map;
[0023] K is the cluster number parameter specified in the K-means algorithm.
[0024] Further, according to the toner color selection scheme of each main color tone, the color matching algorithm is used to calculate the ratio of each toner color to match the main color tone, and the toner color ratio formula corresponding to each main color tone is obtained, which includes:
[0025] The proportion of the toner colors is adjusted using a genetic algorithm, and the specific steps include:
[0026] First, a plurality of toner color ratio combinations are initialized as a population, the color difference of each toner color ratio combination is evaluated by a fitness function, and the combination with the highest fitness is selected as a parent generation; crossover and mutation operations are performed on the selected parent generation to generate a new toner ratio configuration, and the fitness evaluation, selection, crossover and mutation steps are repeated until a preset number of iterations or a color difference threshold is reached, and finally the toner ratio configuration whose fitness exceeds the preset threshold is output as the toner color ratio formula;
[0027] The calculation formula of the fitness is:
[0028]
[0029] Wherein, j represents the index of the toner color ratio combination, f j represents the fitness of the jth toner color ratio combination, ΔE j represents the color difference of the jth toner color ratio combination;
[0030]
[0031] in, and is the CIELAB value of the color generated by the jth combination of the toner color ratios; and is the CIELAB value of the target color.
[0032] Furthermore, according to the toner color selection scheme of each main color tone, a color matching algorithm is used to calculate the ratio of each toner color to match the main color tone, and obtaining the toner color ratio formula corresponding to each main color tone also includes:
[0033] Using digital twin technology to verify all the toner color ratio formulas includes the following steps:
[0034] Using CAD software to create a geometric model of the full-body ceramic tile, collecting production data of the full-body ceramic tile and establishing a physical model and a behavior model simulating the production process, and creating a digital twin model;
[0035] The toner color ratio formula is input into the digital twin model, initial conditions and boundary conditions are set, and the toner color ratio formula is simulated in the simulation software to obtain a simulated color; image processing and analysis technology is used to evaluate the effect of the simulated color, the color difference between the simulated color and the corresponding main color tone is calculated, and the toner color ratio formula whose color difference is greater than a preset threshold is filtered out.
[0036] Furthermore, according to the main color ratio data and the color powder color ratio formula of each grid, and in combination with K main color ratios, the specific feeding amount of each color powder color in each grid is calculated, and the overall feeding scheme is obtained as follows:
[0037] The toner color ratio formula includes at least one toner color selection scheme and its corresponding ratio;
[0038] According to the main color ratio data of each grid and the color ratio formula of the toner, and in combination with K main color ratios, the specific feeding amount of each toner in each grid is calculated;
[0039] Sum up the toner feeding amounts in all grids and calculate the total feeding amount of each toner;
[0040] The optimization is performed using a mixed integer linear programming algorithm. The specific steps include:
[0041] Define the toner usage variable and the toner feed amount variable;
[0042] Minimize the number of toner types used;
[0043] Establish constraints, including main color requirement satisfaction constraints and feed quantity constraints;
[0044] Using an optimization solver to solve the problem and obtain an optimal solution; the optimal solution includes an optimal toner color selection scheme for each main color tone and its corresponding proportion;
[0045] According to the optimal solution, the overall feeding plan is generated, including the specific feeding amount T of each color powder. i and toner using variable Y i , where i represents the index of the color powder, Y i A binary variable indicating whether toner i is used; and obtaining the toner delivery type according to the toner usage variable.
[0046] Furthermore, the hopper is specifically:
[0047] The number of hoppers is equal to the number of toner feeding types in the overall feeding scheme. The i-th hopper is based on Y i , T i The material loss rate is equipped with the toner delivery type i and can perform reciprocating motion, and the material loss rate is 0.3% to 2.0%.
[0048] A system for determining the proportion formula of each color of a full-body tile, comprising:
[0049] An image acquisition module is used to acquire a surface image of a standard full-body ceramic tile sample and a plurality of cross-sectional images at different depth levels to obtain a first standard image set;
[0050] A color analysis module, used to perform pixel-level color analysis on the first standard image set, determine the color information of each pixel, and generate a comprehensive color distribution map;
[0051] A cluster analysis module, used for performing cluster analysis on the comprehensive color distribution map using a clustering algorithm, determining the main colors and their proportions in the comprehensive color distribution map, and obtaining K main colors and K main color proportions;
[0052] A color scheme generating module, used for obtaining a color scheme corresponding to each of the main colors according to the color of the toner, wherein the color scheme includes at least one toner color selection scheme;
[0053] A color matching module, for calculating the ratio of each toner color using a color matching algorithm according to the toner color selection scheme of each main color tone to match the main color tone, and obtaining a toner color ratio formula corresponding to each main color tone;
[0054] A grid division module is used to divide the comprehensive color distribution map into a plurality of grids, each grid representing a small area, determine the proportion of each main color tone in each grid, and obtain the main color tone ratio data of each grid;
[0055] A feeding calculation module is used to calculate the specific feeding amount of each toner color in each grid according to the main color ratio data and the toner color ratio formula of each grid, and in combination with K main color ratios, to obtain an overall feeding plan;
[0056] The automation control module is used to design and implement an automation control system for managing multiple independent hoppers, each of which is equipped with a control valve.
[0057] Furthermore, the surface image and the cross-sectional images at multiple different depth levels of the standard full-body ceramic tile sample are obtained specifically as follows:
[0058] The standard full-body ceramic tile sample is placed in a shadowless light box and photographed using a high-resolution camera; a multi-angle diffuse light source is arranged in the shadowless light box.
[0059] Compared with the prior art, the present invention has the following beneficial effects:
[0060] 1. The present invention uses a color matching algorithm to calculate the ratio of each color powder according to the color selection scheme of each main color tone, and uses a genetic algorithm to adjust these ratios. First, the application of the genetic algorithm ensures the precise adjustment of the color powder ratio, so that the final formula can accurately match the main color tone and reduce color difference. Secondly, the method improves the efficiency of formula design by iterative calculation through an automated genetic algorithm. In addition, multiple iterations and fitness evaluation processes ensure the color consistency and quality stability of each batch of products. This method is highly adaptable and can flexibly adjust the color powder formula. It is suitable for the production needs of various types and colors of full-body tiles. At the same time, through standardized algorithms and steps, it ensures the controllability of the production process and the repeatability of the results. The genetic algorithm accurately regulates the color powder ratio, which helps to reduce the material loss rate, reduce production costs and material waste, and thus significantly optimize the color powder management.
[0061] 2. The present invention verifies all toner color ratio formulas using digital twin technology, thereby achieving the beneficial effects of accurate verification and optimization of formula design. Digital twin technology combines geometric models, physical models and behavioral models, and can simulate the real production process in a simulation environment, accurately predict the color reaction of each toner under high temperature conditions, so that the color scheme can take these changes into account during the design phase, thereby more accurately matching the color requirements of the final product. Virtual simulation reduces the number of actual experiments and the use of related materials, reduces costs and improves production efficiency. By discovering and filtering formulas with color differences greater than the preset threshold in advance, the color consistency and stability of each batch of products in the production process are guaranteed. This method significantly optimizes toner management, effectively reduces material waste and rework rates caused by color difference problems, ensures efficient use of toner and accuracy of inventory management, and further reduces production costs.
[0062] 3. The present invention combines the main color ratio data of each grid and the color powder ratio formula, and uses a mixed integer linear programming algorithm to optimize the use and feeding amount of the color powder, so as to achieve accurate color matching and efficient resource utilization. Through the application of automation and optimization algorithms, the production efficiency and degree of automation are improved, and the production constraints are comprehensively considered to ensure the balance between product quality and production process. In addition, with the goal of minimizing the number of color powders, the management and procurement process is simplified, and production costs and material waste are reduced. Reducing the number of color powders makes the use of hoppers more efficient, reduces the frequency of cleaning and maintenance of hoppers, and reduces equipment loss and related maintenance costs. This method of determining the ratio formula of each color helps to optimize color powder management, improve the utilization rate of color powder, and reduce waste. BRIEF DESCRIPTION OF THE DRAWINGS
[0063] Figure 1 A flow chart of a method for determining a formula for the proportion of materials used in each color of a full-body ceramic tile provided by an embodiment of the present invention;
[0064] Figure 2 A schematic diagram of a toner color selection scheme for main color tone 1 provided in an embodiment of the present invention;
[0065] Figure 3 A schematic diagram of a toner sample provided by an embodiment of the present invention;
[0066] Figure 4 A schematic diagram of grid division provided by an embodiment of the present invention;
[0067] Figure 5 A system structure diagram for determining a formula for material proportions of various colors of a full-body ceramic tile provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0068] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0069] Embodiment 1:
[0070] Factory A specializes in producing high-quality square full-body tiles and has accumulated mature products in the market. In order to further optimize the production process, save costs, and ensure the consistency and high quality of product colors, Factory A adopts a method to determine the proportion of materials for each color of full-body tiles, including:
[0071] See also Figure 1 In S10, a surface image of a standard full-body ceramic tile sample and a plurality of cross-sectional images at different depth levels are acquired to obtain a first standard image set;
[0072] Furthermore, the surface image and the cross-sectional images at multiple different depth levels of the standard full-body ceramic tile sample are obtained specifically as follows:
[0073] The standard full-body ceramic tile sample is placed in a shadowless light box and photographed using a high-resolution camera; a multi-angle diffuse light source is arranged in the shadowless light box;
[0074] As an embodiment of the present invention, a high-resolution camera is used to obtain a complete surface view of the standard solid ceramic tile sample; the standard solid ceramic tile sample is then cut into a number of thin slices of uniform thickness, and cross-sectional images are taken piece by piece after cutting to reflect the internal color distribution; finally, the surface image and all the cross-sectional images are aligned to ensure the spatial consistency of each image.
[0075] By placing the standard full-body tile sample in a shadowless light box and shooting with a high-resolution camera, the influence of ambient light on the image can be eliminated, ensuring uniform lighting and high quality of the captured image. At the same time, the multi-angle diffuse light source arranged in the shadowless light box can effectively avoid shadows and reflections, making the color information of each pixel in the image more accurate and true, thereby improving the accuracy of subsequent color analysis and formula calculation, and ensuring the reliability and consistency of the formula results.
[0076] Furthermore, before performing pixel-level color analysis on the first standard image set, cropping and white balance correction are performed on the first standard image set.
[0077] Image cropping can remove irrelevant parts of the image edge to ensure that the analysis area is focused on the tile sample. White balance correction can correct the color deviation in the image so that it reflects the true color of a standard full-body tile sample under standard lighting conditions. This can improve the accuracy of color analysis and ensure that the extracted color information truly reflects the actual color of the tile, thereby providing a more reliable data basis for subsequent color ratio formula calculations.
[0078] In some optional implementations, the following steps are also included: including a standard color card when shooting, and using image processing software to perform color calibration on the captured image to ensure color accuracy; performing denoising on the first standard image set to improve image quality and the accuracy of color analysis; selecting several representative areas in the image after processing, and performing color comparison to ensure that the color of the processed image is consistent with the actual tile color. By comparing the original image and the processed image, it is ensured that no color deviation is introduced in the preprocessing step.
[0079] See also Figure 1 S20, performing pixel-level color analysis on the first standard image set, determining color information of each pixel, and generating a comprehensive color distribution map;
[0080] A feasible implementation method is provided, including: converting the first standard image set from RGB color space to Lab color space, and using the perceptual uniformity of Lab color space for more accurate color analysis and color difference calculation. Traversing each pixel of each image to extract its color information; integrating the color information of the surface image and each layer of cross-section image into a two-dimensional plane to generate the final comprehensive color distribution map. The specific steps are as follows:
[0081] The color information of each layer is superimposed according to the same weight. Since it is a whole tile, the weight of each layer is the same, and the color data of each layer can be directly averaged to obtain the final color value of each pixel.
[0082] For example, for a pixel P, if its color values in the surface layer, middle layer, and bottom layer are C1, C2, and C3 respectively, then its final color value C can be expressed as:
[0083]
[0084] See also Figure 1 In S30, cluster analysis is performed on the comprehensive color distribution map using a clustering algorithm to determine the main colors and their proportions in the comprehensive color distribution map, thereby obtaining K main colors and K main color proportions;
[0085] Furthermore, the clustering algorithm is used to perform cluster analysis on the comprehensive color distribution map to determine the main color tones and their proportions in the comprehensive color distribution map, and the K main color tones and K main color tones proportions are obtained as follows:
[0086] Using K-means algorithm to perform cluster analysis on the comprehensive color distribution map;
[0087] K is the cluster number parameter specified in the K-means algorithm.
[0088] Through cluster analysis, a large amount of color data in the image can be grouped and summarized into a limited number of main colors (main colors). The proportion of each main color in the image can reflect the overall color distribution of the tile and provide a basis for subsequent color formula calculations. At the same time, by adjusting the K value, it can flexibly respond to color distribution requirements of different complexities, with high adaptability and practicality.
[0089] According to the complexity of tile colors and actual needs, the clustering number parameter K in the K-means algorithm is determined. For example, K=5 can be selected, indicating that the color data is expected to be divided into 5 main color groups (main colors).
[0090] K-means algorithm is used to perform cluster analysis on color data.
[0091] Initial step: randomly select K initial cluster centers (dominant colors).
[0092] Iteration steps: For each pixel, calculate its distance to each cluster center and assign it to the cluster center with the closest distance. Then recalculate the color value of each cluster center as the mean color of all pixels in the cluster.
[0093] Repeat the above process until the cluster center no longer changes significantly or the maximum number of iterations is reached.
[0094] After clustering is completed, each cluster represents a main color. The ratio of the number of pixels in each cluster to the total number of pixels is calculated to obtain the ratio of each main color in the image. The clustering results are shown in Table 1.
[0095] Table 1. K-means clustering result data table
[0096] Cluster number Main color tone (CIELAB value) Number of pixels Total number of pixels Main color ratio (%) 1 (50,70,60) 200,000 1,000,000 20% 2 (90,-10,90) 250,000 1,000,000 25% 3 (40,50,-80) 150,000 1,000,000 15% 4 (70,-50,50) 200,000 1,000,000 20% 5 (95,0,0) 200,000 1,000,000 20%
[0097] See also Figure 1 In S40, according to the toner color, a color decomposition technology is used to obtain a color scheme corresponding to each of the main colors, wherein the color scheme includes at least one toner color selection scheme, such as Figure 2 As shown, the figure shows three different toner color selection schemes for generating the color of the main tone 1. Each scheme achieves the color matching of the main tone 1 through a different toner combination.
[0098] Furthermore, the color powder is an existing inorganic pigment powder used for coloring the whole ceramic tile, and is mixed to prepare the main color tone that meets the design requirements. The schematic diagram of the color powder sample provided by the embodiment of the present invention is as follows: Figure 3 shown.
[0099] See also Figure 1 S50, according to the toner color selection scheme of each main color tone, using a color matching algorithm to calculate the ratio of each toner color to match the main color tone, and obtaining a toner color ratio formula corresponding to each main color tone;
[0100] Further, according to the toner color selection scheme of each main color tone, the color matching algorithm is used to calculate the ratio of each toner color to match the main color tone, and the toner color ratio formula corresponding to each main color tone is obtained, which includes:
[0101] The proportion of the toner colors is adjusted using a genetic algorithm, and the specific steps include:
[0102] First, a plurality of toner color ratio combinations are initialized as a population, the color difference of each toner color ratio combination is evaluated by a fitness function, and the combination with the highest fitness is selected as a parent generation; crossover and mutation operations are performed on the selected parent generation to generate a new toner ratio configuration, and the fitness evaluation, selection, crossover and mutation steps are repeated until a preset number of iterations or a color difference threshold is reached, and finally the toner ratio configuration whose fitness exceeds the preset threshold is output as the toner color ratio formula;
[0103] like Figure 2As shown, there are three toners in the toner color selection scheme 3 corresponding to the main color 1: A, E and F. For the initial population, several groups of toner color ratio combinations are randomly generated. For example, 5 groups of ratio combinations are generated as the initial population (the population size is 5): P1 = [0.2, 0.5, 0.3]; P2 = [0.4, 0.4, 0.2]; P3 = [0.3, 0.3, 0.4]; P4 = [0.25, 0.5, 0.25]; P5 = [0.35, 0.3, 0.35];
[0104] Among them, P i represents the i-th group of initial toner color ratio combination;
[0105] The calculation formula of the fitness is:
[0106]
[0107] Wherein, j represents the index of the toner color ratio combination, f j represents the fitness of the jth toner color ratio combination, ΔE j represents the color difference of the jth toner color ratio combination;
[0108]
[0109] in, and is the CIELAB value of the color generated by the jth combination of the toner color ratios; and is the CIELAB value of the target color.
[0110] In this embodiment, toner A, toner E and toner F are used to synthesize the main color 1. and This is the CIELAB value of primary color 1.
[0111] By using color matching algorithms and genetic algorithms, the ratio of each color powder is accurately calculated to ensure that the final formula can accurately match each main color and reduce color difference problems. Automated iterative calculations, crossover and mutation operations improve the efficiency of formula design. Multiple iterations and fitness evaluations ensure color consistency and stable quality of each batch of products, reducing rework and scrap rates. This method is highly adaptable and can flexibly adjust the color powder formula. Accurate adjustment of the color powder ratio reduces the material loss rate and color powder waste, significantly reducing production costs. Through standardized algorithms and steps, the controllability of the production process and the repeatability of the results are guaranteed, and the reliability and predictability of production are improved.
[0112] See also Figure 1 In S60, the comprehensive color distribution map is divided into a plurality of grids, each grid representing a small area, such as Figure 4 As shown, the proportion of each main color tone in each grid is determined to obtain main color tone ratio data of each grid;
[0113] Furthermore, according to the toner color selection scheme of each main color tone, a color matching algorithm is used to calculate the ratio of each toner color to match the main color tone, and obtaining the toner color ratio formula corresponding to each main color tone also includes:
[0114] Using digital twin technology to verify all the toner color ratio formulas includes the following steps:
[0115] Using CAD software to create a geometric model of the full-body ceramic tile, collecting production data of the full-body ceramic tile and establishing a physical model and a behavior model simulating the production process, and creating a digital twin model;
[0116] The toner color ratio formula is input into the digital twin model, initial conditions and boundary conditions are set, and the toner color ratio formula is simulated in the simulation software to obtain a simulated color; image processing and analysis technology is used to evaluate the effect of the simulated color, the color difference between the simulated color and the corresponding main color tone is calculated, and the toner color ratio formula whose color difference is greater than a preset threshold is filtered out.
[0117] In this embodiment, the method further includes: adjusting the toner ratio according to the simulation result to obtain an optimized toner ratio, and repeating the simulation and verification process.
[0118] Table 2. Toner ratio formula optimization verification table
[0119]
[0120] As can be seen from Table 2, after optimizing the toner ratio, the color difference between the simulated color CIELAB value and the target CIELAB value is used to determine whether the formula has passed the verification. The preset threshold is 4. Finally, the toner color selection schemes that fail are filtered out.
[0121] In the production process of full-body ceramics, high-temperature sintering has a key impact on color. Digital twin technology and genetic algorithms can simulate the chemical and physical changes under these high-temperature conditions in the design stage, accurately predict the performance of toner in actual production, optimize the toner ratio, and ensure color consistency and quality stability. The toner color ratio formula is input into the model, and the color reaction of the toner under high temperature conditions is simulated in the simulation software, so that the color scheme can take these changes into account in the design stage, thereby more accurately matching the color requirements of the final product. Virtual simulation reduces the number of actual experiments and the use of related materials, reduces costs and improves production efficiency. At the same time, by discovering and filtering formulas with color differences greater than the preset threshold in advance, the color consistency and stability of each batch of products in the production process are guaranteed, effectively reducing material waste and rework rate caused by color difference problems, and significantly optimizing toner management and production efficiency.
[0122] See also Figure 1 S70, according to the main color ratio data and the color ratio formula of the toner of each grid, and in combination with K main color ratios, calculate the specific feeding amount of each toner color in each grid to obtain an overall feeding plan;
[0123] See also Figure 1 In the S80, an automated control system is designed and implemented to manage multiple independent hoppers, each of which is equipped with a control valve.
[0124] Furthermore, according to the main color ratio data and the color powder color ratio formula of each grid, and in combination with K main color ratios, the specific feeding amount of each color powder color in each grid is calculated, and the overall feeding scheme is obtained as follows:
[0125] The toner color ratio formula includes at least one toner color selection scheme and its corresponding ratio;
[0126] According to the main color ratio data of each grid and the color ratio formula of the toner, and in combination with K main color ratios, the specific feeding amount of each toner in each grid is calculated;
[0127] Sum up the toner feeding amounts in all grids and calculate the total feeding amount of each toner;
[0128] The optimization is performed using a mixed integer linear programming algorithm. The specific steps include:
[0129] Define the toner usage variable and the toner feed amount variable;
[0130] Minimize the number of toner types used;
[0131] Among them, the objective function is:
[0132]
[0133] Where m represents the total number of existing toners; i represents the index of the toner, and Y i A binary variable indicating whether toner i is used; and obtaining the toner delivery type according to the toner usage variable.
[0134] Establish constraints, including main color requirement satisfaction constraints and feed quantity constraints;
[0135] Using an optimization solver to solve the problem and obtain an optimal solution; the optimal solution includes an optimal toner color selection scheme for each main color tone and its corresponding proportion;
[0136] According to the optimal solution, the overall feeding plan is generated, including the specific feeding amount T of each color powder. i and toner using variable Y i , where i represents the index of the color powder, Y i A binary variable indicating whether toner i is used.
[0137] The present invention calculates the main color ratio data and the color ratio formula of each grid, and combines the ratio of multiple main colors to optimize the specific feeding amount of the color powder using a mixed integer linear programming algorithm, thereby achieving higher resource utilization efficiency and production accuracy. The method ensures the accuracy of the color ratio by accurately calculating the specific feeding amount of each color powder color in each grid, improves the consistency and stability of the product color, and reduces the color difference problem. Using a mixed integer linear programming algorithm, the use of color powder types is minimized, the resource utilization of the color powder is optimized, and the production cost is reduced. The automated optimization algorithm and solution process improve the efficiency of the feeding scheme design and the overall production efficiency. By establishing the main color demand satisfaction constraint and the feeding amount constraint, the strict control of various indicators and requirements in the production process is ensured, and the balance between product quality and production process is guaranteed.
[0138] Reducing the number of toner types simplifies toner and inventory management, and reduces the complexity of warehousing and management. By reducing the frequency and quantity of purchases, procurement and storage costs can be significantly reduced. During the production process, the fewer the number of toner types, the shorter the time to switch toners and adjust equipment, thereby improving production efficiency and capacity. Simplifying the toner mixing process can reduce mixing errors and ensure the consistency and stability of product color. Reducing the number of toner types also helps to reduce scrap and waste, better control material use, and improve resource utilization efficiency.
[0139] Furthermore, the hopper is specifically:
[0140] The number of hoppers is equal to the number of toner feeding types in the overall feeding scheme. The i-th hopper is based on Y i 、Ti The material loss rate is equipped with the toner delivery type i and can perform reciprocating motion, and the material loss rate is 0.3% to 2.0%.
[0141] By designing each hopper to correspond to a color powder, and setting the number of hoppers according to the overall feeding plan, considering the material loss rate of 0.3% to 2.0%, accurate control of feeding, reduction of cross contamination, optimization of production efficiency, reduction of material waste, reduction of equipment wear and improvement of automation level are achieved. The production efficiency and product quality are significantly improved, the production cost and maintenance cost are reduced, the production process is optimized, and the overall economic benefits are improved.
[0142] Embodiment 2:
[0143] Factory B is a full-body tile manufacturer. In the existing production process, the use of color powder is not accurate, resulting in a certain degree of waste. At the same time, the management of color powder is chaotic. In order to solve these problems, Factory B applied a system to determine the proportion of each color of full-body tiles, including:
[0144] An image acquisition module is used to acquire a surface image of a standard full-body ceramic tile sample and a plurality of cross-sectional images at different depth levels to obtain a first standard image set;
[0145] A color analysis module, used to perform pixel-level color analysis on the first standard image set, determine the color information of each pixel, and generate a comprehensive color distribution map;
[0146] A cluster analysis module, used for performing cluster analysis on the comprehensive color distribution map using a clustering algorithm, determining the main colors and their proportions in the comprehensive color distribution map, and obtaining K main colors and K main color proportions;
[0147] A color scheme generating module, used for obtaining a color scheme corresponding to each of the main colors according to the color of the toner, wherein the color scheme includes at least one toner color selection scheme;
[0148] A color matching module, for calculating the ratio of each toner color using a color matching algorithm according to the toner color selection scheme of each main color tone to match the main color tone, and obtaining a toner color ratio formula corresponding to each main color tone;
[0149] A grid division module is used to divide the comprehensive color distribution map into a plurality of grids, each grid representing a small area, determine the proportion of each main color tone in each grid, and obtain the main color tone ratio data of each grid;
[0150] A feeding calculation module is used to calculate the specific feeding amount of each toner color in each grid according to the main color ratio data and the toner color ratio formula of each grid, and in combination with K main color ratios, to obtain an overall feeding plan;
[0151] The automation control module is used to design and implement an automation control system for managing multiple independent hoppers, each of which is equipped with a control valve.
[0152] In order to reduce waste and improve production efficiency, a new method for determining color matching ratio is proposed, which aims to use as few basic colors as possible and realize automatic control.
[0153] Furthermore, the surface image and the cross-sectional images at multiple different depth levels of the standard full-body ceramic tile sample are obtained specifically as follows:
[0154] The standard full-body ceramic tile sample is placed in a shadowless light box and photographed using a high-resolution camera; a multi-angle diffuse light source is arranged in the shadowless light box.
[0155] Table 3. Distribution of toner feed amount
[0156]
[0157] In this embodiment, the comprehensive color distribution diagram is divided into 4 grids, and the specific feeding amount of each toner and the total feeding amount are shown in Table 3, and a total of 6 toners need to be fed. Among them, toner A is loaded in hopper 1, toner B is loaded in hopper 2, toner C is loaded in hopper 3, toner D is loaded in hopper 4, toner E is loaded in hopper 5, and toner F is loaded in hopper 6.
[0158] Since the material loss rate is 0.3% to 2.0%, the actual total loading amount of each hopper is calculated according to the upper limit of the material loss rate of 2.0%:
[0159] The actual total loading amount of hopper 1 = 4.5 × (1 + 0.02) = 4.59 kg;
[0160] The actual total loading amount of hopper 2 = 5.5 × (1 + 0.02) = 5.61 kg;
[0161] The actual total loading amount of hopper 3 = 4.6 × (1 + 0.02) = 4.692 kg;
[0162] The actual total loading amount of hopper 4 = 4.6 × (1 + 0.02) = 4.692 kg;
[0163] The actual total loading amount of hopper 5 = 2.8 × (1 + 0.02) = 2.856 kg;
[0164] The actual total loading amount of hopper 6 = 2.0 × (1 + 0.02) = 2.04 kg;
[0165] The automation control module includes a central control unit (CCU), a hopper unit, a control valve, a sensor and a human-machine interface (HMI). After the feeding formula is input through the human-machine interface, the CCU allocates the amount of toner to each hopper according to the calculated loading amount and sends a feeding instruction. The control valve accurately feeds the toner according to the instruction, and the sensor monitors the feeding amount in real time and feeds back to the CCU to ensure that the feeding process is accurate. The CCU continuously monitors the status and feeding progress of each hopper, and automatically adjusts the control valve to deal with feeding abnormalities. After the feeding is completed, the system generates a report to record the actual feeding amount and adjustment details. The precise feeding of toner and the stability of the production process are achieved, effectively reducing waste and improving production efficiency.
[0166] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A method for determining the proportion of materials used in each color of a full-body tile, characterized in that: include: Acquire a surface image of a standard full-body ceramic tile sample and a plurality of cross-sectional images at different depth levels to obtain a first standard image set; Performing pixel-level color analysis on the first standard image set to determine the color information of each pixel and generate a comprehensive color distribution map; Performing cluster analysis on the comprehensive color distribution map using a clustering algorithm to determine the main colors and their proportions in the comprehensive color distribution map, thereby obtaining K main colors and K main color proportions; According to the color of the toner, using the color decomposition technology, a color matching scheme corresponding to each of the main colors is obtained, wherein the color matching scheme includes at least one toner color selection scheme; According to the toner color selection scheme of each main color tone, a color matching algorithm is used to calculate the ratio of each toner color to match the main color tone, and a toner color ratio formula corresponding to each main color tone is obtained; Divide the comprehensive color distribution map into a plurality of grids, each grid representing a small area, determine the proportion of each main color tone in each grid, and obtain main color tone ratio data of each grid; According to the main color ratio data and the color ratio formula of the toner in each grid, and in combination with K main color ratios, the specific feeding amount of each toner color in each grid is calculated to obtain an overall feeding plan; Design and implement an automated control system for managing a number of independent hoppers, each of which is equipped with a control valve.
2. A method for determining the proportion of materials of each color of a full-body tile according to claim 1, characterized in that: The surface image and multiple cross-sectional images at different depth levels of the standard full-body ceramic tile sample are obtained as follows: The standard full-body ceramic tile sample is placed in a shadowless light box and photographed using a high-resolution camera; a multi-angle diffuse light source is arranged in the shadowless light box.
3. A method for determining the proportion of materials of each color of a full-body tile according to claim 1, characterized in that: The first standard image set is cropped and white balance corrected before pixel-level color analysis is performed on the first standard image set.
4. A method for determining the proportion of materials of each color of a full-body ceramic tile according to claim 1, characterized in that: The clustering algorithm is used to perform cluster analysis on the comprehensive color distribution map to determine the main color tones and their proportions in the comprehensive color distribution map, and the K main color tones and K main color tones proportions are obtained as follows: Using K-means algorithm to perform cluster analysis on the comprehensive color distribution map; K is the cluster number parameter specified in the K-means algorithm.
5. A method for determining the proportion of materials of each color of a full-body tile according to claim 1, characterized in that: According to the toner color selection scheme of each main color tone, the color matching algorithm is used to calculate the ratio of each toner color to match the main color tone, and the toner color ratio formula corresponding to each main color tone is obtained, which includes: The proportion of the toner colors is adjusted using a genetic algorithm, and the specific steps include: First, a plurality of toner color ratio combinations are initialized as a population, the color difference of each toner color ratio combination is evaluated by a fitness function, and the combination with the highest fitness is selected as a parent generation; crossover and mutation operations are performed on the selected parent generation to generate a new toner ratio configuration, and the fitness evaluation, selection, crossover and mutation steps are repeated until a preset number of iterations or a color difference threshold is reached, and finally the toner ratio configuration whose fitness exceeds the preset threshold is output as the toner color ratio formula; The calculation formula of the fitness is: Wherein, j represents the index of the toner color ratio combination, f j represents the fitness of the jth toner color ratio combination, ΔE j represents the color difference of the jth toner color ratio combination; in, and is the CIELAB value of the color generated by the jth combination of the toner color ratios; and is the CIELAB value of the target color.
6. A method for determining the proportion of materials of each color of a full-body tile according to claim 1, characterized in that: According to the toner color selection scheme of each main color tone, the ratio of each toner color is calculated by using a color matching algorithm to match the main color tone, and the toner color ratio formula corresponding to each main color tone is obtained, which also includes: Using digital twin technology to verify all the toner color ratio formulas includes the following steps: Using CAD software to create a geometric model of the full-body ceramic tile, collecting production data of the full-body ceramic tile and establishing a physical model and a behavior model simulating the production process, and creating a digital twin model; The toner color ratio formula is input into the digital twin model, initial conditions and boundary conditions are set, and the toner color ratio formula is simulated in the simulation software to obtain a simulated color; image processing and analysis technology is used to evaluate the effect of the simulated color, the color difference between the simulated color and the corresponding main color tone is calculated, and the toner color ratio formula whose color difference is greater than a preset threshold is filtered out.
7. A method for determining the proportion of materials of each color of a full-body ceramic tile according to claim 1, characterized in that: According to the main color ratio data and the color powder color ratio formula of each grid, and combined with K main color ratios, the specific feeding amount of each color powder color in each grid is calculated, and the overall feeding scheme is obtained as follows: The toner color ratio formula includes at least one toner color selection scheme and its corresponding ratio; According to the main color ratio data of each grid and the color ratio formula of the toner, and in combination with K main color ratios, the specific feeding amount of each toner in each grid is calculated; Sum up the toner feeding amounts in all grids and calculate the total feeding amount of each toner; The optimization is performed using a mixed integer linear programming algorithm. The specific steps include: Define the toner usage variable and the toner feed amount variable; Minimize the number of toner types used; Establish constraints, including main color requirement satisfaction constraints and feed quantity constraints; Using an optimization solver to solve the problem and obtain an optimal solution; the optimal solution includes an optimal toner color selection scheme for each main color tone and its corresponding proportion; According to the optimal solution, the overall feeding plan is generated, including the specific feeding amount T of each color powder. i and toner using variable Y i , where i represents the index of the color powder, Y i A binary variable indicating whether toner i is used; and obtaining the toner delivery type according to the toner usage variable.
8. A method for determining the proportion of materials of each color of a full-body ceramic tile according to claim 7, characterized in that: The hopper is specifically: The number of hoppers is equal to the number of toner feeding types in the overall feeding scheme. The i-th hopper is based on Y i , T i The material loss rate is equipped with the toner delivery type i and can perform reciprocating motion, and the material loss rate is 0.3% to 2.0%.
9. A system for determining the proportion of materials used in each color of a full-body tile, characterized in that: include: An image acquisition module is used to acquire a surface image of a standard full-body ceramic tile sample and a plurality of cross-sectional images at different depth levels to obtain a first standard image set; A color analysis module, used to perform pixel-level color analysis on the first standard image set, determine the color information of each pixel, and generate a comprehensive color distribution map; A cluster analysis module, used for performing cluster analysis on the comprehensive color distribution map using a clustering algorithm, determining the main colors and their proportions in the comprehensive color distribution map, and obtaining K main colors and K main color proportions; A color scheme generating module, used for obtaining a color scheme corresponding to each of the main colors according to the color of the toner, wherein the color scheme includes at least one toner color selection scheme; A color matching module, for calculating the ratio of each toner color using a color matching algorithm according to the toner color selection scheme of each main color tone to match the main color tone, and obtaining a toner color ratio formula corresponding to each main color tone; A grid division module is used to divide the comprehensive color distribution map into a plurality of grids, each grid representing a small area, determine the proportion of each main color tone in each grid, and obtain the main color tone ratio data of each grid; A feeding calculation module is used to calculate the specific feeding amount of each toner color in each grid according to the main color ratio data and the toner color ratio formula of each grid, and in combination with K main color ratios, to obtain an overall feeding plan; The automation control module is used to design and implement an automation control system for managing multiple independent hoppers, each of which is equipped with a control valve.
10. A system for determining the proportion of materials used in each color of a full-body ceramic tile according to claim 9, characterized in that: The surface image and multiple cross-sectional images at different depth levels of the standard full-body ceramic tile sample are obtained as follows: The standard full-body ceramic tile sample is placed in a shadowless light box and photographed using a high-resolution camera; a multi-angle diffuse light source is arranged in the shadowless light box.
Citation Information
Cited By
Automatic gelatin feeding system and method
CN119327334A
Gelatin Automatic Feeding System and Feeding Method
CN119327334B
Joint control lamp effect adaptation method and device, equipment and medium
CN119383799A
Joint control lighting effect adaptation method, device, equipment, and medium
CN119383799B
Laboratory color matching intelligent management method and system
CN121982117A
