Building paint toning method based on particle swarm algorithm

Abstract

This invention relates to a method for color matching of architectural coatings based on particle swarm optimization (PSO) algorithm. The method includes the following steps: using a standard color swatch and a colorimeter, under the same light intensity, light angle, and sampling distance, calibrating the basic color of a display one by one; saving the architectural coating color designed using the calibrated display as a standard color and displaying the standard color on this display; adjusting to obtain multiple different standard colorant ratios; iteratively searching for the optimal standard colorant ratio using the PSO algorithm; and outputting the optimal standard colorant ratio. This invention eliminates the industry's reliance on manual experience in color matching, achieving controllable results for the final product without human intervention; making the experimental process highly directional, quickly obtaining satisfactory results with a small number of iterations and very few samples per iteration; achieving color matching results with much higher accuracy than manual color matching; and enabling large-scale application.

Description

Technical Field

[0001] This invention relates to the field of architectural decoration design, and more specifically to a method for color matching of architectural coatings based on particle swarm optimization algorithm. Background Technology

[0002] In architectural interior design, color scheme design is a crucial task. Simply put, it involves designers considering factors such as the building's style, its environment, and the client's needs to determine the appropriate colors for different parts of the building. A good color scheme can significantly enhance the building's appearance and is a major factor in improving its overall quality.

[0003] There are many types of coatings commonly used in the construction industry. For example, latex paint is widely used in building construction, serving multiple functions such as waterproofing, corrosion prevention, and aesthetics. The aesthetic function involves mixing latex paint into various colors. However, these colors are not created out of thin air; they are first designed on a computer by a designer, and then mixed using standard color masterbatches through various methods. This process inevitably leads to various problems.

[0004] In actual engineering projects, even after designers create color schemes, they often cannot be perfectly applied to real buildings. This isn't due to a lack of execution by the construction team, but rather a technical issue. Specifically, the problem arises as follows:

[0005] First, designers use computers for all their design work, including color scheme design, which is accomplished using various professional software. Therefore, when a designer completes a color scheme, their requirements are essentially the colors displayed on the computer screen they are using—the so-called standard colors. However, due to differences in monitor settings, graphics card settings, and individual physical performance variations between monitors, the same color file will display differently on different computers.

[0006] Secondly, the colors designed on the computer are relatively theoretical values ​​and do not take into account engineering applications; however, in the process of architectural decoration, for large-area painting, architectural paint is mixed and applied. The standard colors designed by the designer do not include the mixing scheme of architectural paint.

[0007] The formulation of architectural coatings involves mixing and stirring various colorants in different proportions to obtain a new color. The set of proportions for each colorant constitutes the formulation of the architectural coating. Therefore, a color displayed on a computer screen is not, and cannot be directly converted into, the proportions of the various colorants. In other words, there is a significant gap between seeing a color and the corresponding architectural coating formula.

[0008] Even worse, the color masterbatch formulation differs from the color design on the computer, introducing significant uncertainty. Specifically:

[0009] First, computer software performs a linear superposition and fitting process of colors during color matching, so designers can have a predetermined direction for adjustment. However, the mixing of color masterbatches is a rather complex process, which involves certain chemical and physical reactions. Therefore, there is actually no predetermined direction for adjustment when mixing paints, which creates great uncertainty in the color mixing results. Often, adding too much color masterbatch will make the color look very strange.

[0010] Secondly, because computer software adjusts colors based on the principle of linear superposition and fitting, when designers adjust color ratios, they can preview the results in real time through the calculation of the graphics card, which is very intuitive. In contrast, the process of mixing paint involves many steps such as sampling, stirring, and curing, which is not intuitive and is relatively time-consuming.

[0011] To address the aforementioned issues, current practices in the application of technology involve specialized color matching technicians adjusting the color of architectural coatings after a designer provides a standard sample color. These technicians rely on their experience, experimenting until they achieve a color that closely approximates the sample color displayed on the computer. This mixing plan is then recorded, and based on this plan, the construction team prepares and mixes the coatings on a large scale before final application.

[0012] The core of existing technology is people, or rather, human experience. Although various color measurement devices, such as colorimeters, are now widely used as aids, these devices can only indicate the color measurement value and still cannot provide the correct color masterbatch ratio.

[0013] At the same time, due to the huge difference between computer software color matching and manual paint mixing, the most common situation in the industry is that color matching technicians, after many days and a lot of experiments around the designed color, are very close to the standard sample color, but still cannot cross the last threshold to mix a satisfactory color. It should be noted that "satisfactory" here means that the naked eye cannot distinguish the difference between the standard sample color displayed on the computer and the mixed architectural paint.

[0014] In summary, the shortcomings of existing technologies are as follows:

[0015] 1. Since the core of existing technologies is based on human experience, this brings great uncertainty to the color matching work of architectural coatings. It is impossible to control the effect of the finished product, the time and number of color matching experiments, or the directional problems in the experiment process. At the same time, sensory experience cannot be transferred. In other words, it is very difficult for a color matching technician with rich experience to pass on his experience to others, which further exacerbates the severity of the problem.

[0016] 2. Because the mixing and blending process of color masterbatch is not a linear superposition process, the process of mixing paint is actually a non-directional trial and error based on human experience to obtain an approximate result. This process is uncontrollable, which means that most of the mixed architectural paint colors still have a certain visual difference from the standard sample color, which cannot be compensated for. In the end, the designer and the client can only make a compromise, and the best design effect in the designer's mind cannot be achieved. Summary of the Invention

[0017] This invention addresses the aforementioned problems by providing a method for color matching of architectural coatings based on particle swarm optimization (PSO) algorithm. Its aim is to eliminate the industry's reliance on manual experience in color matching, achieving controllable results without human intervention. The experimental process for color matching is highly directional, quickly yielding satisfactory results with a small number of iterations and samples per iteration. It produces color matching results with significantly higher accuracy than manual color matching and can be widely adopted.

[0018] To solve the above problems, the technical solution provided by the present invention is as follows:

[0019] A method for color matching of architectural coatings based on particle swarm optimization algorithm includes the following steps:

[0020] S100. Using a standard color chart and a colorimeter, under the same light intensity, light illumination angle, and sampling distance, calibrate the basic colors of the display one by one until the color difference between the display and the corresponding standard color chart when displaying each basic color is less than a manually preset correction color difference threshold; each standard color chart is uniquely coated with a standard color masterbatch; the color of the standard color masterbatch corresponds one-to-one with the basic color of the display;

[0021] S200. Save the color of the architectural paint designed using the calibrated display as a standard color, and display the standard color on the display;

[0022] S300. Adjust to obtain multiple sets of different standard color masterbatch ratios; the standard color masterbatch ratio is the set of proportions of the standard color masterbatch of different colors; then use the particle swarm optimization algorithm to iteratively search for the optimal standard color masterbatch ratio;

[0023] S400. Output the optimal standard color masterbatch ratio as the final output result of this color matching method.

[0024] Preferably, the display has eight basic colors, including red, orange, yellow, green, blue, purple, black, and white;

[0025] The standard color masterbatch comes in eight colors: red, orange, yellow, green, blue, purple, black, and white, each corresponding to one of the eight basic colors of a display.

[0026] Preferably, the standard color masterbatch ratio is expressed by the following formula:

[0027] D = {D red D orange D yellow D green D blue D purple D black D white}

[0028] Where: D is the standard colorant ratio; D red D represents the number of parts of the red standard colorant; orange D represents the number of parts of the standard orange colorant; yellow Yellow represents the proportion of standard color masterbatch; D green The number of parts of the green standard colorant; D blue D represents the number of parts of the standard blue colorant; purple D represents the number of parts of the purple standard colorant; black D represents the number of parts of the standard black colorant; white This refers to the number of parts of the white standard color masterbatch.

[0029] Preferably, S300 specifically includes the following steps:

[0030] S310. Prepare multiple sets of architectural coatings according to different standard color masterbatch ratios; then apply the obtained architectural coatings to different color plate substrates to obtain multiple test color plates; then use the same display and the same colorimeter in S100, and under the same light intensity, light irradiation angle and sampling distance as in S100, obtain the color difference value between each test color plate and the standard sample color displayed on the display.

[0031] S320. Generate a particle swarm; then randomly initialize particles in the feasible solution space; the particle swarm is expressed by the following formula:

[0032] C Di ={C D1 C D2 C D3 ,...C DN}

[0033] Among them, C D Let i be a particle; i is the particle number, and i∈[1,N]; N is the population size of the particle swarm, in units of particles; the particle includes particle velocity and particle position; the particle velocity includes the current velocity and the updated velocity; the particle position includes the current position, the updated position, the current individual optimal position, and the current global optimal position;

[0034] S330. Calculate the particle velocity and particle position for each particle to obtain the optimal particle for this iteration; the optimal particle is the particle with the largest fitness function value in this iteration;

[0035] S340. Based on the value of the fitness function of the optimal particle, perform the following operation:

[0036] If the value of the fitness function of the optimal particle is greater than the artificially preset fitness threshold, then the standard color masterbatch ratio of the optimal particle is calibrated as the optimal standard color masterbatch ratio; then S400 is executed;

[0037] If the value of the fitness function of the optimal particle is not greater than the fitness threshold, then execute S350;

[0038] S350. Determine whether the number of iterations in this iteration is less than the manually preset maximum number of iterations, and then perform the following operations based on the determination result:

[0039] If the number of iterations in this iteration is less than the maximum number of iterations, then execute S360;

[0040] If the number of iterations in this iteration is not less than the maximum number of iterations, then the standard color masterbatch ratio of the optimal particle is calibrated as the optimal standard color masterbatch ratio; then S400 is executed;

[0041] S360. Iteratively correct the particle velocity and particle position of each particle; wherein, during the update, the current individual optimal position and the current global optimal position are updated simultaneously;

[0042] S370. Prepare the corresponding architectural coating according to the updated standard color masterbatch ratio for each particle; then apply the obtained architectural coating to different color plate substrates to obtain the corresponding test color plates; then use the same display and the same colorimeter from S100, and under the same light intensity, light irradiation angle, and sampling distance as in S100, obtain the color difference value between each corresponding test color plate and the standard color displayed on the display.

[0043] S380. Return and execute S330 again.

[0044] Preferably, the iterative correction of the particle velocity in S360 is expressed by the following formula:

[0045] v i_m+1 =ωv i_m +c1r1(p best_m -x i_m )+c2r2(g best_m -x i_m )

[0046] Where: v i_m+1 The updated speed is v; i_m The current speed; x i_m The current position; p best_m The optimal position for the current individual; g best_m Let m be the current global optimal position; m be the current iteration number, and m≥1; c1 be the individual learning factor; c2 be the social learning factor; r1 be a uniformly distributed random number, and r1∈[0,1]; r2 be a uniformly distributed random number, and r2∈[0,1]; ω be the random inertia weight, expressed by the following formula:

[0047] ω=ω min +rand1()*(ω max -ω min )+σ*rand2()

[0048] Where: ω max The maximum value of the random inertia weight is preset manually; ω min The minimum value of the random inertia weight is preset manually; rand1() is a uniformly distributed random number, and rand1()∈[0,1]; rand2() is a normally distributed random number; σ is the standard deviation of the random inertia weight and the expected value of the random inertia weight.

[0049] Preferably, the iterative correction of the particle position in S360 is expressed by the following formula:

[0050] x i_m+1 =x i_m +t*v i_m

[0051] Where: x i_m+1 The updated position is t; the iteration interval is t, which is preset manually.

[0052] Preferably, the fitness function is expressed as follows:

[0053]

[0054] Where: fit jHere, is the fitness function; rand3() is a uniformly distributed random number, and rand3()∈[0,1]; Cost j is the cost function; j is the iteration count.

[0055] Preferably, the cost function is expressed as follows:

[0056]

[0057] Wherein: β1 is the first correction coefficient, used to correct the weight of the luminance value, and is preset manually; β2 is the second correction coefficient, used to correct the weight of the red-green value, and is preset manually; β3 is the third correction coefficient, used to correct the weight of the yellow-blue value, and is preset manually; β4 is the fourth correction coefficient, used to correct the weight of the chroma value, and is preset manually; β5 is the fifth correction coefficient, used to correct the weight of the hue angle, and is preset manually; ΔL j The difference in brightness value for the j-th time is expressed by the following formula:

[0058] ΔL j =L j -L sample

[0059] Where: L j L is the brightness value at the j-th time; sample The brightness value of the standard sample color;

[0060] Δa j The difference between the red and green values ​​in the j-th iteration is expressed by the following formula:

[0061] Δa j =a j -a sample

[0062] Where: a j Let a be the red-green value of the j-th iteration; sample The red-green value of the standard sample color;

[0063] Δb j The difference between the yellow and blue values ​​in the j-th iteration is expressed by the following formula:

[0064] Δb j =b j -b sample

[0065] Where: b j b is the yellow-blue value of the j-th iteration; sample The yellow-blue value of the standard sample color;

[0066] ΔC j The difference in chroma values ​​for the j-th order is expressed by the following formula:

[0067] ΔCj =C j -C sample

[0068] Where: C j C is the chroma value of the j-th order; sample The chroma value of the standard sample color;

[0069] Δh j The hue angle difference value of the j-th order is expressed by the following formula:

[0070] Δh j =h j -h sample

[0071] Where: h j h is the hue angle value of the j-th time. sample The hue angle value of the standard sample color.

[0072] Preferably, the lens of the colorimeter faces a shielding plate; the shielding plate is a flat plate structure with a square acquisition window; the side of the shielding plate facing the lens of the colorimeter is coated with a matte black coating; the geometric center of the acquisition window is on the principal optical axis of the lens of the colorimeter; the acquisition window is perpendicular to the principal optical axis of the lens of the colorimeter.

[0073] The vertical distance from the shielding plate to the colorimeter is a fixed distance;

[0074] When the standard color swatch is used to collect data, it is placed against the side of the shielding plate away from the colorimeter, and its color is exposed through the acquisition window; the standard color swatch covers the entire area of ​​the acquisition window;

[0075] When the colorimeter collects data, the test color swatch is placed against the side of the shielding plate away from the colorimeter, and its color is exposed through the acquisition window; the test color swatch covers the entire area of ​​the acquisition window;

[0076] When the display is collecting data from the colorimeter, it is placed against the side of the shielding plate away from the colorimeter, and the color is exposed through the acquisition window; the display covers the entire area of ​​the acquisition window.

[0077] Compared with the prior art, the present invention has the following advantages:

[0078] 1. Since the color matching method of the present invention does not rely on human experience at all, the entire color matching process is calculated by computer based on the collected objective data, thereby eliminating the industry's reliance on human experience in color matching work and achieving controllable effect of the finished product without human intervention;

[0079] 2. Because the color adjustment method of the present invention utilizes the advantages of particle swarm optimization in optimization problems, the experimental process of color adjustment is very directional. Compared with the random and uncertain number of manual attempts, the method of the present invention can quickly obtain satisfactory results with a small number of iterations and a very small number of samples per iteration.

[0080] 3. Since the color matching method of this invention adopts the particle swarm optimization algorithm, the experimental process of color matching is very directional, thus obtaining color matching results with much higher accuracy than manual color matching; theoretically, as long as the number of iterations is large enough and the accuracy of the colorimeter is high enough, the method of this invention can obtain a color matching scheme that is infinitely close to the standard sample color.

[0081] 4. Since the color matching method of the present invention is free from dependence on human experience, it can be promoted on a large scale, which is more conducive to the development of the industry and enterprises. Attached Figure Description

[0082] Figure 1 This is a schematic diagram of the color adjustment method according to a specific embodiment of the present invention;

[0083] Figure 2 This is a top view schematic diagram of the device arrangement for collecting color difference data of a display according to a specific embodiment of the present invention;

[0084] Figure 3 This is a top view schematic diagram of the device arrangement for collecting color difference data from a standard color chart according to a specific embodiment of the present invention;

[0085] Figure 4 This is a top view schematic diagram of the device arrangement for collecting color difference data from a test color chart according to a specific embodiment of the present invention;

[0086] Figure 5 A computer screenshot of the standard color of a specific embodiment of the present invention;

[0087] Figure 6 This is a composite diagram of nine test color chart photos taken during particle initialization according to a specific embodiment of the present invention.

[0088] Figure 7 This is a schematic diagram of nine test color swatches stitched together during the 28th iteration of a specific embodiment of the present invention;

[0089] Figure 8 This is a photograph of the test color plate corresponding to particle number 9 after the 30th iteration of a specific embodiment of the present invention.

[0090] The components include: 1. Display, 2. Standard color swatch, 3. Test color swatch, 4. Masking plate, 5. Acquisition window, 6. Matte coating, and 7. Colorimeter. Detailed Implementation

[0091] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. After reading the present invention, any modifications of the present invention in various equivalent forms by those skilled in the art will fall within the scope defined by the appended claims.

[0092] It should be noted beforehand that the method of this invention involves a large amount of initial color acquisition work in step S310, which is not a capability that ordinary building decoration companies possess. Therefore, the application scenario of this invention is for paint manufacturers to pre-collect their own color masterbatch products and establish a database; if a color design scheme is subsequently provided by a designer, the paint manufacturer can use the method of this invention to fit the design and provide a direct formula ratio to the designer.

[0093] It should also be noted beforehand that the architectural coating used in this specific embodiment is a commonly used latex paint, and its color masterbatch is produced by the applicant. There are certain differences between color masterbatches from different manufacturers, but this does not affect the application of the present invention. As long as the method of the present invention is implemented step by step, the same result can be obtained.

[0094] like Figure 1 As shown, a method for color matching of architectural coatings based on particle swarm optimization includes the following steps:

[0095] S100. Using standard color swatch 2 and colorimeter 7, under the same light intensity, light irradiation angle, and sampling distance, calibrate the basic colors of display 1 one by one until the color difference between display 1 and the corresponding standard color swatch 2 when displaying each basic color is less than the manually preset correction color difference threshold; each standard color swatch 2 is uniquely coated with a standard color masterbatch; the color of the standard color masterbatch corresponds one-to-one with the basic color of display 1.

[0096] In this specific embodiment, the display 1 has eight basic colors, including red, orange, yellow, green, blue, purple, black, and white.

[0097] There are eight standard color masterbatches: red, orange, yellow, green, blue, purple, black, and white, which correspond to the eight basic colors of display 1.

[0098] like Figures 2-4 As shown in this specific embodiment, in order to minimize environmental interference, the equipment is arranged in the following manner:

[0099] Fix the colorimeter 7 in place, and then point the lens of the colorimeter 7 directly at a shielding plate 4.

[0100] It should be noted that the purpose of the shielding plate 4 is to minimize the interference of ambient light on the data acquisition of the colorimeter 7.

[0101] The shielding plate 4 is a flat plate structure with a square acquisition window 5; the side of the shielding plate 4 facing the lens of the colorimeter 7 is coated with a matte black coating 6.

[0102] It should be noted that since the side of the lens facing the colorimeter 7 is black and non-reflective, the color shown in the acquisition window 5 in the lens of the colorimeter 7 will not be interfered with by reflection, diffuse reflection, or other reasons.

[0103] The geometric center of the acquisition window 5 is on the principal optical axis of the lens of the colorimeter 7; the acquisition window 5 is perpendicular to the principal optical axis of the lens of the colorimeter 7.

[0104] like Figure 3 As shown, when the standard color swatch 2 is used to collect data from the colorimeter 7, it is placed close to the side of the shielding plate 4 away from the colorimeter 7, and the color is exposed through the acquisition window 5; the standard color swatch 2 covers the entire area of ​​the acquisition window 5.

[0105] like Figure 4 As shown, when the test color plate 3 is being collected by the colorimeter 7, it is placed close to the side of the shielding plate 4 away from the colorimeter 7, and the color is exposed through the acquisition window 5; the test color plate 3 covers the entire area of ​​the acquisition window 5.

[0106] like Figure 2 As shown, when the colorimeter 7 collects data, the display 1 is placed close to the side of the shielding plate 4 away from the colorimeter 7, and the color is exposed through the acquisition window 5; the display 1 covers the entire area of ​​the acquisition window 5.

[0107] It should be noted that the reason why the standard color plate 2, the test color plate 3, and the display 1 must be in close contact with the shielding plate 4 when data is being collected is to block light from other directions from passing through the acquisition window 5 and entering the lens of the colorimeter 7.

[0108] It should be further explained that the vertical distance between the shielding plate 4 and the colorimeter 7 is a fixed distance, and the sampling range of the colorimeter 7 is only slightly larger than the range of the sampling window 5, so as to ensure the purity of the sampled colors to the greatest extent.

[0109] It should be noted that the function of S100 is to calibrate the error between the colorimeter 7 and the monitor 1. This step cannot be achieved by the conventional method of parameter calibration on the colorimeter 7. The reason is that when designers are doing color design, their color matching process is not a simple linear adjustment, but a very complex color fitting process. Therefore, it can only correct the error between the basic hue of the monitor 1 and the standard color master, that is, the so-called color difference threshold, within an acceptable range.

[0110] In this specific embodiment, the corrected color difference threshold between the basic color of the display 1 and the standard color masterbatch is set to 0.05.

[0111] It should be further explained that an S100 calibration must be performed every time this method is used for color adjustment. This is because standard color masterbatches produced by different manufacturers have certain differences, and even products from different batches of the same manufacturer may differ. Therefore, calibration is necessary and important.

[0112] S200. Save the color of the architectural latex paint designed using the calibrated display 1 as a standard color, and display the standard color on the display 1.

[0113] like Figure 5 The image shown is a computer screenshot of a pre-designed standard color. It is almost impossible to achieve a completely consistent latex paint combination using only human experience, which is the current state of the industry and a technical problem that needs to be solved.

[0114] It should be noted that designers using a monitor 1 calibrated with S100 must strictly avoid adjusting color display parameters and can only design color schemes in design software.

[0115] S300. Adjust to obtain multiple sets of different standard color masterbatch ratios; the standard color masterbatch ratio is the set of proportions of standard color masterbatches of different colors; then use particle swarm optimization algorithm to iteratively search for the optimal standard color masterbatch ratio.

[0116] In this specific embodiment, the standard color masterbatch ratio is expressed according to formula (1):

[0117] D = {D red D orange D yellow D green D blue D purple D black D white} (1)

[0118] Where: D is the standard colorant ratio; D red D represents the number of parts of the red standard colorant; orange D represents the number of parts of the standard orange colorant; yellow Yellow represents the proportion of standard color masterbatch; D green The number of parts of the green standard colorant; D blue D represents the number of parts of the standard blue colorant; purple D represents the number of parts of the purple standard colorant; black D represents the number of parts of the standard black colorant; white This refers to the number of parts of the white standard color masterbatch.

[0119] In this specific embodiment, each standard color masterbatch is supplied using a titration device, with each portion being 5 ml. The dripped standard color masterbatch is stirred evenly in a stirring flask using an electric shaker. A color-mixing steel ball is also placed in the stirring flask to maximize the uniformity of the standard color masterbatch.

[0120] In this specific embodiment, S300 specifically includes the following steps:

[0121] S310. Prepare multiple sets of latex paint according to different standard color masterbatch ratios; then apply the obtained latex paint to different color plate substrates to obtain multiple test color plates 3; then use the same display 1 and the same colorimeter 7 in S100, and under the same light intensity, light irradiation angle and sampling distance as in S100, obtain the color difference value between each test color plate 3 and the standard sample color displayed on the display 1 one by one.

[0122] It should be noted that the purpose of S310 is to establish an initial color space. This step involves a relatively large amount of work, but once it is established, it can be saved as a database and refined and upgraded in future work to make the color space richer.

[0123] In this specific embodiment, there are 8 colors of standard color masterbatch. In S310, the number of each color is combined from 0 to 4 drops. Excluding the combination of all 0 drops, a total of 390,624 combinations are obtained. Each type of latex paint is sprayed onto a test color plate 3, and then the color difference data is sampled according to the same device and method as in S100.

[0124] It should be noted that the titration, stirring, spraying, and data sampling are all completed by customized automated equipment, reducing the degree of human intervention, improving the accuracy of the mixing ratio, and also freeing up human resources.

[0125] It should be further explained that the combination of 0 to 4 drops of each type is insufficient to meet the color design requirements; therefore, this step is only used to initialize the movement space of the particles. In subsequent iterations, further adjustments will be made based on the position of the particles, so there is no need to combine them too much in this step to save resources.

[0126] It is important to reiterate that the application of this invention's method is intended for paint manufacturing companies, not for designers or design firms. This is because the large-scale production equipment involved, such as titration devices, multi-bottle mixing, and automated spraying, is beyond the capabilities of ordinary companies, and it is unnecessary for routine design work to purchase, install, and maintain such industrial equipment. Therefore, this invention's method is suitable for paint manufacturing companies that have already established an initial color space and stored it as a database. In their daily work, they can receive standard color files from designers, obtain the optimal standard color masterbatch ratio using this invention's method, and then return the optimized color masterbatch to the designers as an additional service.

[0127] S320. Generate a particle swarm; then randomly initialize particles in the feasible solution space; the particle swarm is expressed according to equation (2):

[0128] C Di ={C D1 C D2 C D3 ,...C DN} (2)

[0129] Among them, C D Let i be a particle; i is the particle number, and i∈[1,N]; N is the population size of the particle swarm, in units of particles, representing the number of particles; a particle includes particle velocity and particle position; particle velocity includes current velocity and updated velocity; particle position includes current position, updated position, current individual best position and current global best position.

[0130] In this specific embodiment, the number of particles is set to 9, i.e., N=9.

[0131] like Figure 6 As shown, it is obvious that since the particles generated in the first batch are random, the color of the test color plate 3 made from the latex paint mixed according to the corresponding standard color masterbatch ratio is very different from the standard sample color, which is normal; the next step of this invention is to make the color of these particles gradually approach the standard sample color based on the movement state of these particles until it is within an acceptable range.

[0132] It should be noted that, Figure 6 It is a photograph, and due to lens and lighting conditions, there may be some color differences between it and the actual colors on site. It is for illustrative purposes only.

[0133] It should be noted that the number of particles does not need to be set too high, for two reasons: First, to reduce the amount of computation and increase the computation time for each iteration; second, because each iteration requires readjusting, stirring, spraying, and sampling the standard colorant ratio corresponding to each particle, these processes still take time, so a larger number of particles would lead to a linear increase in workload. On the other hand, although increasing the number of particles may reduce the number of iterations to some extent, this is negligible compared to the time cost of the two considerations mentioned above. Therefore, the final selection of a particle number of 9 is in line with the actual engineering requirements.

[0134] S330. Calculate the particle velocity and particle position of each particle to obtain the optimal particle for this iteration; the optimal particle is the particle with the largest fitness function value in this iteration.

[0135] In this specific embodiment, the fitness function is expressed according to equation (3):

[0136]

[0137] Where: fit j For the fitness function; λ is an infinitesimally small positive real number; Cost j is the cost function; j is the iteration count.

[0138] In this specific embodiment, the cost function is expressed according to equation (4):

[0139]

[0140] Wherein: β1 is the first correction coefficient, used to correct the weight of the brightness value, and is preset manually; β2 is the second correction coefficient, used to correct the weight of the red-green value, and is preset manually; β3 is the third correction coefficient, used to correct the weight of the yellow-blue value, and is preset manually; β4 is the fourth correction coefficient, used to correct the weight of the chroma value, and is preset manually; β5 is the fifth correction coefficient, used to correct the weight of the hue angle, and is preset manually.

[0141] It should be noted that the functions of β1, β2, β3, β4, and β5 are to dynamically adjust the weight of each color difference parameter for different devices and testing environments, so as to make the visual effect better match the machine display effect.

[0142] In this specific embodiment, β1 = 1.65, β2 = 1.94, β3 = 1.89, β4 = 1.06, and β5 = 1.25.

[0143] It should be noted that the above values ​​are only valid in this specific embodiment. Each time this method is used for color adjustment, it needs to be dynamically adjusted according to the equipment, environment, and design requirements.

[0144] ΔLj The difference in brightness value for the j-th time is expressed by equation (5):

[0145] ΔL j =L j -L sample (5)

[0146] Where: L j L is the brightness value at the j-th time; sample The brightness value of the standard sample color.

[0147] Δa j The difference between the red and green values ​​in the j-th iteration is expressed by equation (6):

[0148] Δa j =a j -a sample (6)

[0149] Where: a j Let a be the red-green value of the j-th iteration; sample The red and green values ​​of the standard sample color.

[0150] Δb j The difference between the yellow and blue values ​​in the j-th iteration is expressed by equation (7):

[0151] Δb j =b j -b sample (7)

[0152] Where: b j b is the yellow-blue value of the j-th iteration; sample The yellow-blue value of the standard sample color.

[0153] ΔC j The difference in chroma value for the j-th order is expressed by equation (8):

[0154] ΔC j =C j -C sample (8)

[0155] Where: C j C is the chroma value of the j-th order; sample The chroma value of the standard sample color.

[0156] Δh j The hue angle difference value of the j-th order is expressed according to equation (9):

[0157] Δh j =h j -h sample (9)

[0158] Where: h j h is the hue angle value of the j-th time.sample The hue angle value of the standard sample color.

[0159] In this specific embodiment, L was measured. sample =38.46, a sample =5.51, b sample = -31.89, C sample =31.84, h sample =276.5.

[0160] S340. Based on the value of the fitness function of the optimal particle, perform the following operations:

[0161] If the fitness function value of the optimal particle is greater than the preset fitness threshold, then the standard color masterbatch ratio of the optimal particle is calibrated as the best standard color masterbatch ratio; then S400 is executed.

[0162] If the value of the fitness function of the optimal particle is not greater than the fitness threshold, then execute S350.

[0163] S350. Determine whether the number of iterations in this iteration is less than the manually preset maximum number of iterations, and then perform the following operations based on the determination result:

[0164] If the number of iterations in this iteration is less than the maximum number of iterations, then execute S360.

[0165] If the number of iterations in this iteration is not less than the maximum number of iterations, then the standard color masterbatch ratio of the optimal particle is calibrated as the best standard color masterbatch ratio; then execute S400.

[0166] S360. Iteratively correct the particle velocity and particle position of each particle; when updating, simultaneously update the current individual optimal position and the current global optimal position.

[0167] In this specific embodiment, the iterative correction of the particle velocity is expressed by equation (10):

[0168] v i_m+1 =ωv i_m +c1r1(p best_m -x i_m )+c2r2(g best_m -x i_m (10)

[0169] Where: v i_m+1 For the updated speed; for v i_m Current speed; x i_m This is the current position; p best_m This represents the optimal position for the current individual; g best_mω is the current global optimal position; m is the current iteration number, and m≥1; c1 is the individual learning factor, used to characterize the influence of the optimal position passed by the particle itself on the particle behavior; c2 is the social learning factor, used to characterize the influence of the global optimal position on the particle behavior; r1 is a uniformly distributed random number, and r1∈[0,1]; r2 is a uniformly distributed random number, and r2∈[0,1]; ω is the random inertia weight, expressed by equation (11):

[0170] ω=ω min +rand1()*(ω max -ω min )+σ*rand2() (11)

[0171] Where: ω max The maximum value of the random inertia weight is preset manually; ω min σ is the minimum value of the random inertia weight, which is preset manually; rand1() is a uniformly distributed random number, and rand1()∈[0,1]; rand2() is a normally distributed random number; σ is the standard deviation of the random inertia weight and the expected value of the random inertia weight.

[0172] In this specific embodiment, the iterative correction of the particle position is expressed by equation (12):

[0173] x i_m+1 =x i_m +t*v i_m (12)

[0174] Where: x i_m+1 The position is the updated position; t is the iteration interval, which is preset manually.

[0175] S370. Prepare the corresponding latex paint according to the updated standard color masterbatch ratio for each particle; then apply the obtained latex paint to different color swatches to obtain the corresponding test color swatches 3.

[0176] It should be noted that the updated standard color masterbatch ratio for each particle is not limited to 0-4 drops, but rather the number of each standard color masterbatch is gradually increased, that is, the mixing precision is gradually improved, and the color of the corresponding test color plate 3 gradually approaches the standard sample color.

[0177] Then, using the same display 1 and the same colorimeter 7 in S100, and under the same light intensity, light illumination angle, and sampling distance as S100, the color difference value between each corresponding test color plate 3 and the standard color displayed on the display 1 is obtained one by one.

[0178] S380. Return and execute S330 again.

[0179] Table 1 shows the results of the first 7 iterations. It is clear that each particle is converging toward the standard color during each iteration.

[0180]

[0181] Table 1. Particle Swarm Optimization Iteration Data Acquisition Table (First 6 Iterations)

[0182] Table 2 shows the results of iterations 28 to 30. It is clear that all 9 particles converge around the standard color.

[0183] Table 2. Particle Swarm Optimization Iteration Data Acquisition Table (Iterations 28-30)

[0184]

[0185] like Figure 7 As shown, after the 28th iteration, the colors of the 9 test color swatches 3 are very close to the standard color, but there are still subtle differences that can be detected by the naked eye.

[0186] It should be noted that, Figure 7 It is a photograph, and due to lens and lighting conditions, there are some color differences between it and the actual colors on site. Even under human observation, subtle but visible differences can still be seen between it and the standard sample color and the test color charts 1-9. Therefore... Figure 7 For illustrative purposes only.

[0187] S400 outputs the optimal standard color masterbatch ratio as the final output of this color matching method.

[0188] like Figure 8 As shown, by the 30th iteration, no further iterations are needed because the human eye can no longer distinguish the differences between the colors and the standard sample. At this point, the iteration process can be manually interrupted, and any particle can be randomly selected.

[0189] It should be noted that, Figure 8 The color swatch 30_#9 is a photograph, compared to a standard sample color taken as a computer screenshot. Figure 5 While the two appear to have some color differences in computer files, they are indistinguishable to the naked eye in person. Figure 7 For illustrative purposes only.

[0190] By saving the optimal standard colorant ratio corresponding to the selected particles, it can be used directly in the mixing of architectural latex paints without the need for manual experience.

[0191] In the above detailed description, various features are combined together in a single embodiment to simplify this disclosure. This approach to disclosure should not be construed as reflecting an intention that embodiments of the claimed subject matter require more features than are explicitly stated in each claim. Rather, as reflected in the appended claims, the invention is presented with fewer features than all of the features of the single disclosed embodiment. Therefore, the appended claims are hereby explicitly incorporated into the detailed description, wherein each claim stands alone as a preferred embodiment of the invention.

[0192] The disclosed embodiments have been described above to enable any person skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be applied to other embodiments without departing from the spirit and scope of this disclosure. Therefore, this disclosure is not limited to the embodiments given herein, but is consistent with the broadest scope of the principles and novel features disclosed in this application.

[0193] The foregoing description includes examples of one or more embodiments. It is certainly impossible to describe all possible combinations of components or methods in order to describe the above embodiments, but those skilled in the art will recognize that further combinations and arrangements of the various embodiments are possible. Therefore, the embodiments described herein are intended to cover all such changes, modifications, and variations that fall within the scope of the appended claims. Furthermore, the term "comprising" as used in the specification or claims is interpreted in a manner similar to the term "including," as interpreted when used as a conjunction in the claims. Additionally, the use of any term "or" in the specification of the claims is intended to mean "non-exclusive or."

[0194] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for color matching of architectural coatings based on particle swarm optimization algorithm, characterized in that: Includes the following steps: S100. Using a standard color plate (2) and a colorimeter (7), under the same light intensity, light irradiation angle, and sampling distance, calibrate the basic colors of the display (1) one by one until the color difference between the display (1) and the corresponding standard color plate (2) when displaying each basic color is less than the manually preset correction color difference threshold; each standard color plate (2) is uniquely coated with a standard color masterbatch; the color of the standard color masterbatch corresponds one-to-one with the basic color of the display (1); S200. Save the color of the architectural paint designed using the calibrated display (1) as a standard color and display the standard color on the display (1); S300. Adjust to obtain multiple sets of different standard colorant ratios; the standard colorant ratio is a set of proportions of the standard colorant for different colors; then use the particle swarm optimization algorithm to iteratively search for the optimal standard colorant ratio; wherein: The S300 specifically includes the following steps: S310. Prepare multiple sets of architectural coatings according to different standard color masterbatch ratios; then apply the obtained architectural coatings to different color plate substrates to obtain multiple test color plates (3); then use the same display (1) and the same colorimeter (7) in S100, and under the same light intensity, light irradiation angle and sampling distance as in S100, obtain the color difference value between each test color plate (3) and the standard sample color displayed on the display (1) one by one; S320. Generate a particle swarm; then randomly initialize particles in the feasible solution space; the particle swarm is expressed by the following formula: C Di ={C D1 ,C D2 ,C D3 ,...C DN } Among them, C D Let i be a particle; i is the particle number, and i∈[1,N]; N is the population size of the particle swarm, in units of particles; the particle includes particle velocity and particle position; the particle velocity includes the current velocity and the updated velocity; the particle position includes the current position, the updated position, the current individual optimal position, and the current global optimal position; S330. Calculate the particle velocity and particle position for each particle to obtain the optimal particle for this iteration; the optimal particle is the particle with the largest fitness function value in this iteration; The fitness function is expressed as follows: Where: fit j Here, is the fitness function; rand3() is a uniformly distributed random number, and rand3()∈[0,1]; Cost j Here, j is the cost function; j is the iteration count. The cost function is expressed as follows: Wherein: β1 is the first correction coefficient, used to correct the weight of the luminance value, and is preset manually; β2 is the second correction coefficient, used to correct the weight of the red-green value, and is preset manually; β3 is the third correction coefficient, used to correct the weight of the yellow-blue value, and is preset manually; β4 is the fourth correction coefficient, used to correct the weight of the chroma value, and is preset manually; β5 is the fifth correction coefficient, used to correct the weight of the hue angle, and is preset manually; ΔL j The difference in brightness value for the j-th time is expressed by the following formula: ΔL j =L j -L sample Where: L j L is the brightness value at the j-th time; sample The brightness value of the standard sample color; Δa j The difference between the red and green values ​​in the j-th iteration is expressed by the following formula: Yes j = yes j -in sample Where: a j Let a be the red-green value of the j-th iteration; sample The red-green value of the standard sample color; Δb j The difference between the yellow and blue values ​​in the j-th iteration is expressed by the following formula: Δb j =b j -b sample Where: b j b is the yellow-blue value of the j-th iteration; sample The yellow-blue value of the standard sample color; ΔC j The difference in chroma values ​​for the j-th order is expressed by the following formula: ΔC j =C j -C sample Where: C j C is the chroma value of the j-th order; sample The chroma value of the standard sample color; Δh j The hue angle difference value of the j-th time is expressed by the following formula: Δh j =h j -h sample Where: h j h is the hue angle value of the j-th time. sample The hue angle value of the standard sample color; S400. Output the optimal standard color masterbatch ratio as the final output result of this color matching method.

2. The method for color matching of architectural coatings based on particle swarm optimization algorithm according to claim 1, characterized in that: The display (1) has eight basic colors, including red, orange, yellow, green, blue, purple, black, and white; The standard color masterbatch includes red, orange, yellow, green, blue, purple, black and white, which correspond to the eight basic colors of the display (1).

3. The method for color matching of architectural coatings based on particle swarm optimization algorithm according to claim 2, characterized in that: The standard colorant ratio is expressed by the following formula: D={D red ,D orange ,D yellow ,D green ,D blue ,D purple ,D black ,D white } Where: D is the standard colorant ratio; D red D represents the number of parts of the red standard colorant; orange D represents the number of parts of the standard orange colorant; yellow Yellow represents the proportion of standard color masterbatch; D green The number of parts of the green standard colorant; D blue D represents the number of parts of the standard blue colorant; purple D represents the number of parts of the purple standard colorant; black D represents the number of parts of the standard black colorant; white This refers to the number of parts of the white standard color masterbatch.

4. The method for color matching of architectural coatings based on particle swarm optimization algorithm according to claim 3, characterized in that: The S300 also specifically includes the following steps: S340. Based on the value of the fitness function of the optimal particle, perform the following operation: If the value of the fitness function of the optimal particle is greater than the artificially preset fitness threshold, then the standard color masterbatch ratio of the optimal particle is calibrated as the optimal standard color masterbatch ratio; then S400 is executed; If the value of the fitness function of the optimal particle is not greater than the fitness threshold, then execute S350; S350. Determine whether the number of iterations in this iteration is less than the manually preset maximum number of iterations, and then perform the following operations based on the determination result: If the number of iterations in this iteration is less than the maximum number of iterations, then execute S360; If the number of iterations in this iteration is not less than the maximum number of iterations, then the standard color masterbatch ratio of the optimal particle is calibrated as the optimal standard color masterbatch ratio; then S400 is executed; S360. Iteratively correct the particle velocity and particle position of each particle; wherein, during the update, the current individual optimal position and the current global optimal position are updated simultaneously; S370. Prepare the corresponding building coating according to the updated standard color masterbatch ratio of each particle; then apply the obtained building coating to different color plate base plates to obtain the corresponding test color plate (3); then use the same display (1) and the same colorimeter (7) in S100, and under the same light intensity, light irradiation angle and sampling distance as in S100, obtain the color difference value between each corresponding test color plate (3) and the standard sample color displayed on the display (1) one by one; S380. Return and execute S330 again.

5. The method for color matching of architectural coatings based on particle swarm optimization algorithm according to claim 4, characterized in that: The iterative correction of the particle velocity in S360 is expressed by the following formula: v i_m+1 =ωv i_m +c1r1(p best_m -x i_m )+c2r2(g best_m -x i_m ) Where: v i_m+1 v is the updated speed; i_m x represents the current speed; i_m The current position; p best_m The optimal position for the current individual; g best_m Let m be the current global optimal position; m be the current iteration number, and m≥1; c1 be the individual learning factor; c2 be the social learning factor; r1 be a uniformly distributed random number, and r1∈[0,1]; r2 be a uniformly distributed random number, and r2∈[0,1]; ω be the random inertia weight, expressed by the following formula: oh = oh min +rand1()*(ω max -oh min )+σ*rand2() Where: ω max The maximum value of the random inertia weight is preset manually; ω min The minimum value of the random inertia weight is preset manually; rand1() is a uniformly distributed random number, and rand1()∈[0,1]; rand2() is a normally distributed random number; σ is the standard deviation of the random inertia weight and the expected value of the random inertia weight.

6. The method for color matching of architectural coatings based on particle swarm optimization algorithm according to claim 5, characterized in that: In S360, the iterative correction of the particle position is expressed by the following formula: x i_m+1 =x i_m +t*v i_m Where: x i_m+1 The updated position is t; the iteration interval is t, which is preset manually.

7. The method for color matching of architectural coatings based on particle swarm optimization algorithm according to claim 6, characterized in that: The lens of the colorimeter (7) faces a shielding plate (4); the shielding plate (4) is a flat plate structure with a square acquisition window (5); the side of the shielding plate (4) facing the lens of the colorimeter (7) is coated with a matte black coating (6); the geometric center of the acquisition window (5) is on the principal optical axis of the lens of the colorimeter (7); the acquisition window (5) is perpendicular to the principal optical axis of the lens of the colorimeter (7); The vertical distance between the shielding plate (4) and the colorimeter (7) is a fixed distance; When the standard color swatch (2) is used to collect data by the colorimeter (7), it is placed close to the side of the shielding plate (4) away from the colorimeter (7) and the color is exposed through the acquisition window (5); the standard color swatch (2) covers the entire area of ​​the acquisition window (5); When the test color plate (3) is used to collect data by the colorimeter (7), it is placed close to the side of the shielding plate (4) away from the colorimeter (7) and the color is exposed through the acquisition window (5); the test color plate (3) covers the entire area of ​​the acquisition window (5); When the display (1) is collecting data from the colorimeter (7), it is placed close to the side of the shielding plate (4) away from the colorimeter (7) and the color is exposed through the acquisition window (5); the display (1) covers the entire area of ​​the acquisition window (5).

Images