Color mixing device based on AI and operation method thereof

Abstract

According to one embodiment of the present disclosure, an operation method of a color mixing device includes: determining an area including a color value of a target color from one or more areas included in a color space based on the color value of the target color; generating one or more samples based on the color value of the target color; selecting a deep learning model for the determined region from a plurality of pre-learned AI models; and determining, from among the samples, a sample that constitutes the optimal mixing conditions of the color value of the target color using the selected AI model. The plurality of AI models are learned based on a learning-specific data set in which a mixing ratio of each of the one or more reference colors is arbitrarily set.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of Korean Patent Application No. 10-2023-0028019 filed on March 2, 2023, in the Korean Intellectual Property Office, the entire contents of which are incorporated herein by reference. Technical Field

[0003] Embodiments of the present disclosure relate to an artificial intelligence (AI)-based color mixing device and an operating method thereof. Background Art

[0004] Computerized color matching (CCM) is a technology used to determine the color of paint during the color mixing process. CCM measures the reflectance of the target color and calculates the mixing conditions, including the mixing ratio of each reference color (e.g., red, green, and blue). If the color created based on the calculated mixing conditions differs from the target color, the color created by the worker must be corrected, necessitating the calculation of precise mixing conditions. Summary of the Invention

[0005] When using CCM to determine color combinations, there may be a large color difference between the color combined by the CCM and the target color. In order to reduce the color difference between the combined color and the target color, workers are required to perform the additional task of correcting the combined color.

[0006] Therefore, the time required to perform color mixing increases due to the additional tasks required after mixing colors using CCM.

[0007] Furthermore, whether the target color can be achieved may depend on the experience and skills of the workers.

[0008] Technical challenges of the embodiments of the present disclosure are not limited to the aforementioned problems, and any other technical problems not mentioned herein will be clearly understood by those skilled in the art to which the present disclosure pertains from the following description.

[0009] According to one embodiment of the present disclosure, a method for operating a color mixing device includes: determining, based on the color value of the target color, an area including the color value of the target color from one or more areas included in a color space; generating one or more samples based on the color value of the target color; selecting an artificial intelligence (AI) model for the determined area from a plurality of pre-learned AI models; and determining, using the selected AI model, a sample of optimal mixing conditions for forming the color value of the target color from the samples. The plurality of AI models are learned based on a learning-specific dataset, wherein a mixing ratio of each of the one or more reference colors is arbitrarily set.

[0010] In one embodiment, the generating of the sample may include generating the sample based on a primary mixing condition in which the reference colors are arbitrarily mixed within a ratio from a minimum mixing ratio to a maximum mixing ratio of each reference color constituting the determined area.

[0011] In one embodiment, generating the sample may include generating the sample based on a secondary mixing condition, wherein the primary mixing condition is arbitrarily changed within a preset range.

[0012] In one embodiment, the determining may include: determining a sample having a color difference between a color value of the target color and a color value of the sample smaller than a preset value as the optimal mixing condition.

[0013] In one embodiment, the determining may include determining a sample that minimizes a color difference value between a color value of the target color and a color value of the sample as the optimal mixing condition.

[0014] According to one embodiment of the present disclosure, a color mixing device includes: an area determination device that determines an area including the color value of a target color from one or more areas included in a color space based on the color value of the target color; a sample generation device that generates one or more samples based on the color value of the target color; a model selection device that selects an AI model for the determined area from a plurality of pre-learned AI models; and a determination device that determines, from the samples, a sample of optimal mixing conditions for forming the color value of the target color using the selected AI model. The plurality of AI models are learned based on a dedicated learning dataset, wherein the mixing ratio of each of the one or more reference colors is arbitrarily set.

[0015] In one embodiment, the sample generating device may generate samples based on a primary mixing condition in which the reference colors are arbitrarily mixed within a ratio from a minimum mixing ratio to a maximum mixing ratio of each reference color constituting the determined area.

[0016] In one embodiment, the sample generating device may generate a sample based on a secondary mixing condition, wherein the primary mixing condition is arbitrarily changed within a preset range.

[0017] In one embodiment, the determining device may determine a sample having a color difference between a color value of the target color and a color value of the sample that is less than a preset value as the optimal mixing condition.

[0018] In one embodiment, the determining device may determine the sample in which the color difference between the color value of the target color and the color value of the sample is minimized as the optimal mixing condition.

[0019] The color mixture determination device and the operating method thereof according to various embodiments disclosed in the present disclosure may divide the color space into a plurality of regions and may construct an AI model optimized for a specific color region.

[0020] The color mixture determination apparatus and the operating method thereof according to various embodiments disclosed in the present disclosure may select an optimized AI model among a plurality of AI models based on a color value of a target color.

[0021] Therefore, the time of the color mixing process can be shortened and the accuracy of the mixing can be improved.

[0022] The technical effects of the color mixture determination device and the operating method thereof according to the embodiments of the present disclosure are not limited to the above-mentioned effects, and those skilled in the art can clearly understand other effects not mentioned based on the embodiments of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 is a block diagram of a color mixing device according to one embodiment of the present disclosure.

[0024] Figure 2 FIG. 4 shows a color space based on the Lab color system according to one embodiment of the present disclosure.

[0025] Figure 3 A table showing region-specific reference colors according to one embodiment of the present disclosure is shown.

[0026] Figure 4 3 is a flowchart illustrating a deep learning model learning method according to one embodiment of the present disclosure.

[0027] Figure 5 1 is a flowchart illustrating an operating method of a color mixing device according to an embodiment of the present disclosure.

[0028] For the description of the drawings, the same or similar components will be marked with the same or similar reference numerals. DETAILED DESCRIPTION

[0029] The embodiments of the present disclosure will be described below with reference to the accompanying drawings. However, it should be understood by those skilled in the art that various modifications, equivalents and / or substitutions may be made to the various embodiments described herein without departing from the scope and spirit of the present disclosure.

[0030] The embodiments of the present disclosure and the terms used herein are not intended to limit the technical features described in the present disclosure to specific embodiments, and it should be understood that these embodiments and terms include modifications, equivalents, or alternatives to the corresponding embodiments described herein. In the description of the drawings, similar or related components may be marked with similar reference signs / numbers. Unless the context otherwise indicates, the singular form of a noun corresponding to an item may include one or more items.

[0031] In the present disclosure, expressions such as “A or B,” “at least one of A and B,” “at least one of A or B,” “A, B or C,” “at least one of A, B and C,” and “at least one of A, B or C” may include any or all combinations of one or more of the relevant listed items. Terms such as “first,” “second,” “A,” “B,” “(a),” or “(b)” may be used to simply distinguish the corresponding component from other components, but do not limit the corresponding component in other aspects (such as importance or order) unless otherwise explicitly stated.

[0032] In this specification, when a component (e.g., a first component) is referred to as being “coupled / connected” or “connected / connected to” another component (e.g., a second component), whether or not accompanied by the term “operatively” or “communicatively,” this may mean that one component can be connected to the other component directly (e.g., by wire or wirelessly) or indirectly (e.g., through a third component).

[0033] The methods according to various embodiments disclosed in this specification may be included in a computer program product. The computer program product may be traded as a product between a seller and a buyer. The computer program product may be distributed in the form of a machine-readable storage medium (e.g., a read-only compact disc (CD-ROM)), or may be distributed through an app store, directly between two user devices, or online (e.g., downloaded or uploaded). In the case of online distribution, at least a portion of the computer program product may be at least temporarily stored in a machine-readable storage medium, such as in the memory of a manufacturer server, an app store server, or a relay server, or may be temporarily generated.

[0034] According to the embodiments disclosed in this specification, each component (e.g., a module or a program) in the above-mentioned components may include a single entity or multiple entities, and some of the multiple objects may be arranged separately on other components. According to the embodiments disclosed in this specification, one or more components in the above-mentioned components or operations may be omitted, or one or more other components or operations may be added. Alternatively or in addition, multiple components (e.g., modules or programs) may be integrated into one component. In this case, the integrated component may perform one or more functions of each component in the multiple components in the same or similar manner as the corresponding components in the multiple components before integration. According to the embodiments disclosed in this specification, the operations performed by modules, programs or other components may be performed by sequential methods, parallel methods, repetitive methods or heuristic methods. Alternatively, at least one or more operations may be performed in other orders, or at least one or more operations may be omitted, or one or more operations may be added.

[0035] Figure 1 is a block diagram of a color mixing device 10 according to one embodiment of the present disclosure.

[0036] refer to Figure 1 , the color mixing device 10 may include a sensor 120, a memory 140 and a processor 160, or any combination thereof. According to one embodiment, Figure 1 The color mixing device 10 shown may also include Figure 1 At least one component other than those shown (eg, a display, an input device, or an output device).

[0037] In one embodiment, the sensor 120 may be an optical sensor. In one embodiment, the sensor 120 may detect the reflectivity of a color. For example, the sensor 120 may represent the color value of the detected color as a Lab color value based on the reflectivity of the detected color.

[0038] In one embodiment, memory 140 may include volatile memory and / or non-volatile memory.

[0039] In one embodiment, the memory 140 may store data used by at least one component of the color mixing device 10 (e.g., the processor 160). For example, the data may include software (or instructions related thereto), input data, or output data. In one embodiment, when the processor 160 executes the instructions, the instructions may cause the color mixing device 10 to perform the operations defined by the instructions.

[0040] In one embodiment, the memory 140 may include one or more software (eg, region determination means 141, sample generation means 143, model learning means 145, model selection means 147, determination means 149, or any combination thereof).

[0041] In one embodiment, processor 160 may include a central processing unit, an application processor, a graphics processing unit, a neural processing unit (NPU), an image signal processor, a sensor hub processor, or a communication processor.

[0042] In one embodiment, the processor 160 can execute software (e.g., the area determination device 141, the sample generation device 143, the model learning device 145, the model selection device 147, the determination device 149, or any combination thereof) to control at least another component (e.g., a hardware or software component) of the color mixing device 10 connected to the processor 160, and can process and calculate various types of data.

[0043] In one embodiment, the database 15 may include data related to color values ​​according to color codes, color mixing data related to previously performed color mixing processes, and data related to minimum and maximum mixing ratios for each reference color.

[0044] A method is described below in which the color mixing device 10 determines the optimal mixing conditions of the target color through the area determination device 141, the sample generation device 143, the model learning device 145, the model selection device 147, the determination device 149 or any combination thereof.

[0045] AI model learning methods

[0046] The model learning device 145 may receive a dedicated learning dataset. The model learning device 145 may receive the dedicated learning dataset from the database 15. Here, the dedicated learning dataset may be data obtained during a previously executed coloring process. The dedicated learning dataset may include data related to a color code of a specific color, a color value of a specific color, a mixing condition for mixing specific colors, or any combination thereof.

[0047] The model learning device 145 can identify a learning-specific data set based on specific standards. The model learning device 145 can identify a learning-specific data set according to each area in the color space. Here, the color space can be a space that expresses a color system in three dimensions, and can be a set of colors based on an RGB (red, green, blue) color system or a Lab color system. The colors included in the Lab color system can be represented by a combination of brightness, a red-green series (a channel), and a yellow-blue series (b channel). The colors in the color system can be represented by a combination of reference colors. The reference colors can be primary colors. The reference colors in the Lab color system may include black and white representing brightness, red and green representing "a channel", and yellow and blue representing "b channel". Multiple areas can be composed based on reference colors (for example, black or white, red or green, yellow or blue). For example, when the color space is divided into eight areas, the first area of ​​the eight areas can be composed based on white, red and yellow, and white, red and yellow can be used as reference colors to express the colors included in the first area.

[0048] The following assumes that the color is a color of the Lab color system and the number of regions in the color space is 8. However, this is merely an illustrative premise, and the present disclosure is not limited to the Lab color system. For example, the present disclosure can be applied to various other color systems (e.g., the RGB color system), and the number of regions in the color space can also be divided into multiple regions.

[0049] Model learning device 145 may learn one or more AI models. Model learning device 145 may learn an AI model for each region. Model learning device 145 may use a dedicated learning dataset to learn one or more AI models. Model learning device 145 may use a dedicated learning dataset identified for each region to learn an AI model for each region. Here, the AI ​​model for each region may be an AI model optimized for that region. AI models may include machine learning models and deep learning models.

[0050] In one embodiment, model learning device 145 may use a dedicated learning dataset based on supervised learning to learn one or more AI models. For example, model learning device 145 may learn an AI model so that, when the color values ​​of a specific color included in the dedicated learning dataset are input into the AI ​​model, the model model outputs a mixing condition for mixing the specific color. Here, the mixing condition may include all information regarding the mixing ratio and mixing amount of each reference color, so as to mix the specific color using the reference colors.

[0051] [Table 1]

[0052] AI models <![CDATA[R 2 (Coefficient of Determination)]]> Prediction accuracy (color difference value) Linear regression 0.782 14.05 Lasso regression 0.755 14.86 K Nearest Neighbors 0.869 9.83 Decision Tree 0.931 6.98 Support Vector Regression 0.836 11.43 Ada Boost 0.832 12.86 Gradient Boosting Regression 0.938 7.16 XG Boost Returns 0.980 3.79 Random Forest Regression 0.969 4.45 Extra Tree Regression 0.979 3.55 Neural Network Regression 0.997 1.45 8-combination NN regression 0.999 0.78

[0053] Table 1 may exemplify various AI models. Among them, the determination coefficient may be an indicator for determining the applicability of the model, and the prediction accuracy may be an indicator indicating the degree of similarity between the color value of the color determined by using the AI ​​model and the color value of the target color.

[0054] Referring to Table 1, the AI ​​model trained by the model learning device 145 may be a deep learning model that implements neural network (NN) regression. The combination of deep learning models performed for each of the eight regions may be referred to as "8-combination NN regression."

[0055] Referring to Table 1, the present disclosure can determine the mixed sample whose color value is closest to the target color by using 8-combination NN regression. When using 8-combination NN regression, it can be identified that the color value of the target color can be most accurately predicted due to the high coefficient of determination, high model fitness, and low color difference value compared to other AI models.

[0056] Hereinafter, it is assumed that the AI ​​model is a deep learning model consisting of three layers. The number of nodes in each of the three layers is 64, 64, or 40, respectively. However, this is for ease of explanation only, and the present disclosure is not limited to deep learning models and can also be applied to various other AI models.

[0057] Methods for determining optimal mixing conditions

[0058] Region determining means 141 may determine one of one or more regions based on the color value of the target color. Region determining means 141 may determine a region including the color value of the target color from one or more regions included in the Lab color space based on the color value of the target color. For example, when the color values ​​of the target color are represented as white, red, and yellow, region determining means 141 may determine the first region using white, red, and yellow as reference colors from the plurality of regions as the region including the target color.

[0059] In order to determine the area including the color value of the target color, the area determination device 141 can search for the color value of the target color. The area determination device 141 can search for the color value of the target color based on the data stored in the database 15. The area determination device 141 can search for the color value of the target color based on the color code of the target color. The area determination device 141 can search for the color value of the target color by comparing the color code of the target color with the data stored in the database 15. Here, the color value can be a value for specifying a specific color. For example, the color value in the Lab color system can be a coordinate in the Lab color space. The color value in the Lab color system can be a coordinate based on the x-axis (a channel), y-axis (b channel) and z-axis (brightness) in the Lab color space. In one embodiment, the color code can represent information indicating the target color in a color distinction classification standard that is different from Lab.

[0060] The sample generating device 143 may generate one or more samples based on the color value of the target color, wherein the sample may be a color combined based on one of a plurality of mixing conditions for forming the color value of the target color.

[0061] In one embodiment, the sample generating device 143 may generate one or more samples using a screening technique. The sample generating device 143 may generate one or more samples using a primary screening technique and / or a secondary screening technique. The primary screening technique may be a technique for generating a sample based on macro-mixing conditions for mixing a target color. The secondary screening technique may be a technique for generating a sample in which the macro-mixing conditions are arbitrarily varied within a preset range.

[0062] The sample generating device 143 may generate one or more samples using a primary screening technique. The sample generating device 143 may generate one or more samples based on a macro mixing condition. The sample generating device 143 may generate one or more samples based on a primary mixing condition. The sample generating device 143 may generate a sample by randomly combining one or more reference colors within a ratio from a minimum mixing ratio to a maximum mixing ratio of each reference color. The sample generating device 143 may generate one or more samples based on a primary mixing condition, wherein the reference colors are randomly mixed within a ratio from a minimum mixing ratio to a maximum mixing ratio of each reference color. Here, the minimum mixing ratio and the maximum mixing ratio may be set based on statistical data of the mixing ratios of the reference colors in a previously executed coloring process. For example, when the mixing ratio of white as a reference color in a previously executed coloring process is an average of 30%, the minimum mixing ratio of white may be set to 10% and the maximum mixing ratio may be set to 50%.

[0063] The sample generating device 143 may generate samples using a secondary screening technique. The sample generating device 143 may use a primary screening technique to derive macro-mixing conditions, and may use the secondary screening technique to generate samples based on mixing conditions within a preset range of the macro-mixing conditions. The sample generating device 143 may generate one or more samples based on the secondary mixing conditions, wherein the primary mixing conditions may be arbitrarily varied within a preset range. Here, the preset range may be set by a setter. For example, when the maximum mixing ratio of white as a reference color is 50% and the mixing ratio of white included in the macro-mixing conditions is 50%, the sample generating device 143 may generate samples in which the maximum mixing ratio of white varies between 10% and 30%.

[0064] The sample generating device 143 can quickly generate a sample having a color value close to the target color based on macro-mixing conditions by using a primary screening technology and / or a secondary screening technology.

[0065] The model selection device 147 may select at least one deep learning model from a plurality of pre-trained deep learning models. The model selection device 147 may select at least one deep learning model from a plurality of pre-trained deep learning models based on the color value of the target color.

[0066] The determination device 149 may determine the optimal mixing conditions for color values ​​constituting the target color using the deep learning model selected by the model selection device 147. In one embodiment, the determination device 149 may determine a sample corresponding to the optimal mixing condition from a plurality of samples using the deep learning model.

[0067] In order to determine the optimal mixing condition between the samples, the determination device 149 may use the color difference between the color value of the sample and the color value of the target color. The color difference may be expressed as Formula 1.

[0068] [Formula 1]

[0069]

[0070] Here, ΔE can represent a color difference value. "L," "a," and "b" can represent coordinate components in the Lab color space. Subscript 1 can represent the coordinate component of the sample color value. Subscript 2 can represent the coordinate component of the target color value.

[0071] In one embodiment, the determining device 149 may determine a sample having a color difference value less than a preset value from among the multiple samples as the optimal mixing condition.

[0072] In one embodiment, the determining device 149 may determine the sample having the smallest color difference value among the plurality of samples as the optimal mixing condition.

[0073] Figure 2A color space 200 based on the Lab color system according to one embodiment of the present disclosure is shown.

[0074] refer to Figure 2 , the color space 200 may be a spherical space. The color space 200 may include reference lines 210, 212, and 214. The reference lines 210, 212, and 214 may be lines connecting reference color coordinates. The color space 200 may be divided into eight regions by the reference lines 210, 212, and 214. The color space 200 may be divided into eight regions, which are bounded by the reference lines 210, 212, and 214. For example, a portion of the color space 200, which is based on the origin 202 and is bounded by the line segment of the first reference line 210 toward white (L=100), the line segment of the second reference line 212 toward red (+a), and the line segment of the third reference line 214 toward yellow (+b), may be referred to as a "first region." Furthermore, a portion of color space 200, based on origin 202 and bounded by a first reference line 210 oriented toward white (L=100), a second reference line 212 oriented toward red (+a), and a third reference line 214 oriented toward blue (-b), may be referred to as a "second region." The reference color for the first region may be white, red, yellow, or any combination thereof, and the reference color for the second region may be white, red, blue, or any combination thereof.

[0075] Figure 3 A table 300 of region-specific reference colors according to one embodiment of the present disclosure is shown.

[0076] refer to Figure 3 The color space 200 can be divided into one or more regions. For example, the color space 200 can be divided into eight regions by reference lines 210, 212, and 214. Each of the eight regions has a reference color that represents the colors included in each region.

[0077] Referring to table 300, the reference color of the first region may be white, red, yellow, or any combination thereof. The colors included in the first region may be represented by color values ​​of white, red, yellow, or any combination thereof. Here, the color values ​​may be coordinates in color space 200. For example, the color value of white may be (0, 0, 100); the color value of red may be (a, 0, 0); and the color value of yellow may be (0, b, 0). When a target color is included in the first region, the color value of the target color may be represented based on the color values ​​of white, red, and yellow.

[0078] Figure 4 3 is a flowchart illustrating a deep learning model learning method according to one embodiment of the present disclosure.

[0079] refer to Figure 4In operation 400, the model learning device 145 may identify a dedicated learning dataset based on a specific standard. The model learning device 145 may identify a dedicated learning dataset according to each region in the color space.

[0080] In operation 402, the model learning device 145 may learn one or more deep learning models. The model learning device 145 may learn a deep learning model for each region. The model learning device 145 may use a learning-specific dataset to learn the one or more deep learning models. The model learning device 145 may use a learning-specific dataset identified for each region to learn a deep learning model for each region.

[0081] In one embodiment, the model learning device 145 may learn one or more deep learning models by using a dedicated learning dataset based on supervised learning.

[0082] Figure 5 is a flowchart illustrating an operating method of the color mixing device 10 according to one embodiment of the present disclosure.

[0083] refer to Figure 5 In operation 500, the region determining device 141 may determine one of the one or more regions based on the color value of the target color. The region determining device 141 may determine a region including the color value of the target color from one or more regions included in the Lab color space based on the color value of the target color.

[0084] To determine an area including the color value of the target color, the area determination device 141 may search for the color value of the target color. The area determination device 141 may search for the color value of the target color based on data stored in the database 15. The area determination device 141 may search for the color value of the target color based on a color code of the target color. The area determination device 141 may search for the color value of the target color by comparing the color code of the target color with the data stored in the database 15.

[0085] In operation 502 , the sample generating device 143 may generate one or more samples based on the color value of the target color.

[0086] In one embodiment, the sample generating device 143 may generate one or more samples using a screening technique. The sample generating device 143 may generate one or more samples using a primary screening technique and / or a secondary screening technique.

[0087] The sample generating device 143 may generate one or more samples using a primary screening technique. The sample generating device 143 may generate one or more samples based on macro-mixing conditions. The sample generating device 143 may generate a sample by randomly combining one or more reference colors within a ratio from a minimum mixing ratio to a maximum mixing ratio of each reference color.

[0088] The sample generating device 143 may generate samples using a secondary screening technique. The sample generating device 143 may derive macro-mixing conditions using a primary screening technique and may generate samples using a secondary screening technique based on mixing conditions within a preset range of the macro-mixing conditions.

[0089] In operation 504, the model selection device 147 may select at least one deep learning model from a plurality of pre-trained deep learning models. The model selection device 147 may select at least one deep learning model from a plurality of pre-trained deep learning models based on the color value of the target color.

[0090] In operation 506, the determination device 149 may determine the optimal mixing conditions for color values ​​constituting the target color using the deep learning model selected by the model selection device 147. In one embodiment, the determination device 149 may determine a sample corresponding to the optimal mixing condition from a plurality of samples using the deep learning model.

[0091] In order to determine the optimal mixing condition between the samples, the determination device 149 may use a color difference value between the color value of the sample and the color value of the target color.

[0092] In one embodiment, the determining device 149 may determine a sample having a color difference value less than a preset value from among the multiple samples as the optimal mixing condition.

[0093] In one embodiment, the determining device 149 may determine the sample having the smallest color difference value among the plurality of samples as the optimal mixing condition.

[0094] The terms "include", "including" and "having" described above, unless otherwise expressly stated to the contrary, all mean that the corresponding components may be included and should be understood to include another component rather than exclude another component. Unless otherwise defined herein, all terms used herein (including technical terms or scientific terms) have the same meaning as those generally understood by those skilled in the art to which the embodiments disclosed in this specification belong. Common terms (such as terms defined in dictionaries) should be understood to have a meaning consistent with the meaning in the context of the prior art and should not be understood to have an idealized or overly formalized meaning unless otherwise clearly defined in this specification.

[0095] The above description is only used to illustrate the technical concepts disclosed in this specification. Those skilled in the art may make various modifications and changes to the fields to which the embodiments disclosed in this specification belong without departing from the essential features of the embodiments disclosed in this specification. Therefore, the embodiments disclosed in this specification are not intended to limit the technical concepts of the embodiments disclosed in this specification, but are intended to explain the technical concepts of the embodiments disclosed in this specification, and the scope of the technical concepts disclosed in this specification is not limited by the embodiments. The scope of protection disclosed in this specification should be interpreted by the attached claims, and all equivalents thereof should be understood to be included within the scope of this specification.

Claims

1. A method for operating a color mixing device, the method comprising: determining, based on the color value of the target color, a region including the color value of the target color from one or more regions included in the color space; generating one or more samples based on the color value of the target color; Selecting a deep learning model for the identified area from among multiple pre-learned artificial intelligence (AI) models; as well as Using the selected AI model, determine the sample with the best mixing conditions for the color values ​​that constitute the target color from the samples, Among them, multiple AI models are learned based on a learning-specific data set, wherein the mixing ratio of each reference color in one or more reference colors is set arbitrarily.

2. The method according to claim 1, wherein Generated samples include: A sample is generated based on the primary mixing condition in which the reference colors are arbitrarily mixed within a ratio from a minimum mixing ratio to a maximum mixing ratio of each reference color constituting the determined area.

3. The method according to claim 2, wherein: Generated samples include: Samples were generated based on secondary mixing conditions where the primary mixing conditions were arbitrarily varied within a preset range.

4. The method according to claim 1, wherein Make sure to include: The sample whose color difference between the color value of the target color and the color value of the sample is less than a preset value is determined as the optimal mixing condition.

5. The method according to claim 1, wherein Make sure to include: The sample with the minimum color difference between the color value of the target color and the color value of the sample is determined to be the optimal mixing condition.

6. A color mixing device comprising: an area determining device configured to determine, based on the color value of the target color, an area including the color value of the target color from one or more areas included in the color space; a sample generating device configured to generate one or more samples based on a color value of a target color; a model selection device configured to select an AI model for the determined area from a plurality of pre-learned AI models; and determining means for determining, from the samples, samples of optimal mixing conditions for color values ​​constituting the target color by using the selected AI model, Among them, multiple AI models are learned based on a learning-specific data set, wherein the mixing ratio of each reference color in one or more reference colors is set arbitrarily.

7. The color mixing device according to claim 6, wherein: The sample generating device is configured to: A sample is generated based on the primary mixing condition in which the reference colors are arbitrarily mixed within a ratio from a minimum mixing ratio to a maximum mixing ratio of each reference color constituting the determined area.

8. The color mixing device according to claim 7, wherein: The sample generating device is configured to: Samples were generated based on secondary mixing conditions where the primary mixing conditions were arbitrarily varied within a preset range.

9. The color mixing device according to claim 6, wherein: Make sure the device is configured to: The sample whose color difference between the color value of the target color and the color value of the sample is less than a preset value is determined as the optimal mixing condition.

10. The color mixing device according to claim 6, wherein: Make sure the device is configured to: The sample that minimizes the color difference between the color value of the target color and the color value of the sample is determined as the optimal mixing condition.

Images