JP2021107781A - Production method of coating material and prediction method of color data - Google Patents

Abstract

To provide a production method of a coating material for obtaining coating of various colors including bright colors in which optical characteristics are difficult to predict, which is based on computer toning and can finish toning with a small number of prototypes regardless of the skill of an operator.SOLUTION: A method for producing a coating material based on computer toning using a device provided with a database in which at least compound composition data Y of a composition containing one or more kinds of color materials and corresponding color data X are registered, and a computer in which a color matching calculation logic using the data registered in the database is operated, includes predetermined steps S101-S111.SELECTED DRAWING: Figure 6

Description

The present invention relates to a method for producing a coating material and a method for predicting color data. More specifically, the present invention relates to a method for producing a paint based on computer toning and a system for predicting color data of a coating film. The present invention also relates to a computer toning system, a system for predicting color data of a coating film, and application software for controlling and operating these systems.

In recent years, due to the diversification of personal tastes and the improvement of beauty, the number of bright colors based on bright pigments such as metal powder and bright mica has increased as the color of various industrial products, especially automobiles. ing. When repairing such a brilliant color, it is common practice to apply a repair paint containing a brilliant pigment to the repaired portion.

When repairing a brilliant color, it is necessary to adjust not only the color tone but also the brilliant feeling so that the repaired part and the like are not conspicuous. However, when preparing a repair paint for a brilliant color that satisfies both the color tone and the brilliant feeling, even a skilled worker needs to repeat trial and error to prototype a large number of repair paints. In some cases. Further, for an inexperienced worker, it is necessary to repeat trial and error more than a skilled worker to make a large number of prototypes of repair paint, which is a very difficult task.
As the operator repeats trial and error, the work time becomes long, and in addition to problems such as a long repair period and a decrease in efficiency, there are also problems such as cost and disposal based on the repair paint that cannot be used even though it is prototyped.

For this reason, many studies have been made in order to quickly and efficiently produce the repair paint regardless of factors such as the skill level of the operator. As one of them, the color data of the target color obtained by the color measurement with the colorimeter and the color data of the color sample whose compounding composition is known are used to obtain the target color using a computer. Studies have been made on the use of computer color matching (CCM) systems to obtain compounding compositions. However, a brilliant color, for example, a metallic coating color containing a brilliant pigment has an angle dependence of spectral reflectance, and it has been difficult to deal with it in a conventional CCM.
In addition, in the toning of repair paints so far, it is often necessary to perform finer toning with this as a starting point after obtaining an approximate paint toning composition (sometimes abbreviated as an approximate composition). In particular, in the preparation of repair paints that produce brilliant colors, conventional CCMs have limitations in the accuracy and reproducibility of the closest blending, and the number of steps required by the operator to repeat trial and error of toning is still large. Therefore, there was a problem in producing the repair paint quickly and efficiently.

According to Patent Document 1, in a color matching system using a computer, the total spectral reflectance of an unknown color panel is determined by a spectrophotometer, and this reflectance data is sent to the computer, and the computer sends the K value (light absorption coefficient) of the pigment. ) And the pre-recorded data representing the S value (light scattering coefficient) are mathematically processed to perform logical color matching.
Patent Document 2 describes a colorimeter, a micro-brilliance measuring device, a plurality of paint formulations, color data and micro-brilliance data corresponding to each paint formulation, color characteristic data and micro-brilliance of a plurality of primary color paints. Characteristic data is registered, and a computer in which a color matching calculation logic operates, a computer toning device composed of the same, and a toning method using the computer are disclosed.
Pat. Sensation characteristic data is registered, and a color matching method using a computer color matching device composed of a computer on which a color matching calculation logic operates and a computer color matching device composed of the same is disclosed.

In Patent Document 4, as a toning method of a metallic coating color useful as a final fine toning of a repair paint, a toning combination of a metallic paint is based on a difference when a front color and a scab color are visually compared. It is stated that when changing the color, the characteristics based on the scale of the front color and the scab color of the paint color and the compounding conversion index of each metallic primary color paint in which the brightness of the front color does not change are used.
Patent Document 5 describes determining or predicting the visual texture parameters of a paint by an artificial neural network based on the color components used in the paint distribution composition. However, optical properties such as spectroscopic reflection properties are determined or predicted by a physical model for a known paint distribution composition. It is expected that a lot of trial and error will be required when preparing a repair paint for bright colors whose optical characteristics are difficult to predict.
In Patent Document 6, in order to determine a coating color having a desired texture or a coating color belonging to a desired color category in a brilliant coating film or the like, data such as the spectral reflectance of the coating color and the micro-brilliance feeling are used. A method for creating a paint color database and a method for searching for a paint color using the database are described, including a step of training a neural network. However, no specific operation or the like when preparing a repair paint for a brilliant color whose optical characteristics are difficult to predict is not disclosed.

Tokukousho 50-28190 Gazette Japanese Unexamined Patent Publication No. 2001-221690 International Publication No. 2002/004567 Japanese Unexamined Patent Publication No. 2004-224966 Special Table 2019-500488 Gazette International Publication No. 2008/156147

An object of the present invention is a method for manufacturing a paint for obtaining a wide variety of paint colors including bright colors whose optical characteristics are difficult to predict, and the number of trial manufactures is small regardless of the skill level of the operator. It is to provide a method of manufacturing a paint based on computer toning which can finish toning with.
An object of the present invention is to provide a method for predicting the color data of a coating film capable of predicting the color data of the coating film of a coating film having a wide variety of compositions including a bright pigment and the like with high accuracy.
An object of the present invention is a computer toning system for preparing a paint for obtaining a wide variety of paint colors, including bright colors whose optical characteristics are difficult to predict, and for the skill level of an operator. Regardless, it is to provide a computer toning system that can finish toning with a small number of trial productions.
An object of the present invention is to provide a system for predicting the color data of a coating film capable of predicting the color data of the coating film of a coating film having a wide variety of compositions including a bright pigment and the like with high accuracy.

The present inventors have conducted diligent studies to solve the above-mentioned problems, and have found that the above-mentioned problems can be solved by adopting the following configuration, and have completed the present invention.
Specifically, it is as follows.
[Item 1]
A database in which at least the compounding composition data Y and the corresponding color data X of the composition containing one or more kinds of color materials are registered, a computer in which the color matching calculation logic using the data registered in the database is operated, and a computer. A method for producing a paint based on computer toning, which comprises the following steps S101 to S111.
(S101) A step of inputting training data to the computer using the data registered in the database (S102) An artificial intelligence model in which the training data is machine-learned and color data X is estimated from the compounding composition data Y. the at least one step of generating (S103) to obtain the target color data _{X t} colors that the target (S104) the target color data _{X t,} the step of inputting to the computer (S105) using a computer search, with obtaining the target color data X _t approximation to find the color data X _n1 and search color approximation formulation composition data Y _n1 corresponding to the data X _n1, comparing said target color data X _t and the search color data X _n1 Step for determining pass / fail (S106) After obtaining candidate compounding composition data Y _{ni which is predicted to give target color data X t} when the step is not passed in the step S105, at least one of the above trainings was performed. using an artificial intelligence model and / or the non-artificial intelligence model prediction formula, together with obtaining the predicted color data X _ni predicted from the candidate blend composition data Y _ni, and the color data X _t and the predicted color data X _ni (S107) A step of determining pass / fail (S107) After obtaining candidate compounding composition data Y _{ni which is predicted to give target color data X t} using a computer when the result is not passed in the S106 step, at least 1 is described above. using species learned artificial intelligence model and / or prediction expression other than the artificial intelligence model, along with obtaining a predicted color data X _ni predicted from the candidate blend composition data Y _ni, the predicted color and the color data X _t A step of repeating the step of comparing with the data X _ni and determining pass / fail until it passes (S108) A step of _{obtaining pass compounding composition data Y C1} when any of the steps S105 to S107 is passed (S109). Step of preparing the actual candidate paint CM _Ci based on the accepted compounding composition data Y _{C1 , obtaining the coated plate of the actual candidate paint CM Ci} , and acquiring the measured color data X _Ci (S110) The color data X _t and the above. Step of determining pass / fail by comparison with actual measurement color data X _Ci and / or comparison of the target color with the color _{of the coating plate of the actual candidate paint CM Ci} (S111) When the process does not pass in the S110 step. In addition, a step of repeating the steps S106 to S110.

[Item 2]
Item 2. The method according to Item 1, wherein in the step S103, the target color data _Xt is the color data of the coating film containing the bright pigment.
[Item 3]
If the step S111 does not pass, _{the difference Δ between the predicted color data X ni} and the measured color data X _Ci is input to the computer as a correction coefficient α, and then the steps S106 to S111 are repeated in Item 1 or 2. The method described.

[Item 4]
A database in which at least the compounding composition data Y and the corresponding color data X of the composition containing one or more kinds of color materials are registered, a computer in which the color matching calculation logic using the data registered in the database is operated, and a computer. A method for predicting color data of a coating film using an apparatus comprising the above method, which comprises the following steps S201 to S209.
(S201) A step of inputting training data to the computer using the data registered in the database (S202) An artificial intelligence model in which the training data is machine-learned and color data X is estimated from the compounding composition data Y. Step of generating at least one type (S203) Step of obtaining compounding composition data Y _{CM of paint CM t} for predicting color data of coating film (S204) Step of inputting the compounding composition data Y _CM into the computer (S205) If necessary, the search using a computer, the step (S206) of obtaining search color data X _n1 corresponding to the blending composition data Y _CM said S205 if the corresponding search color data X _n1 is not found in step or, in the case of not performed the S205 step, from the blend composition data Y _CM, wherein at least one of the learned artificial intelligence model, or, with the at least one learned artificial intelligence model than the artificial intelligence model _{Step of obtaining predicted color data X m1} using the prediction formula of (S207) If necessary, _{the measured color data X CM} of the coated plate coated with the _{paint CM t} is acquired, and the predicted color data X _m1 is obtained. Process to compare

[Item 5]
The step of generating at least one of the artificial intelligence models is
(I) A step of learning an artificial intelligence model by using each compounding composition data Y and each color data X as learning data relating to one or more kinds of compositions containing no glitter pigment.
(Ii) A step of learning an artificial intelligence model by using each compounding composition data Y and each color data X as learning data relating to one or more kinds of compositions containing a brilliant pigment.
Item 4. The method according to any one of Items 1 to 4.
[Item 6]
The step of generating at least one of the artificial intelligence models is
Content of light-reflecting pigment in composition, content of photo-interfering pigment, content of orientation control agent, content of light-reflecting pigment in composition for each hue, each hue of photo-interfering pigment One or more data selected from different contents, the content of each hue of the colorant, and the sum of two or more of those contents, and / or
Shape data of the coloring material contained in the composition,
The method according to any one of Items 1 to 5, which comprises a step of training an artificial intelligence model by using the above as training data.
[Item 7]
Item 6. The method according to any one of Items 1 to 6, wherein the step of switching to use a prediction formula other than the artificial intelligence model is included in the repeating step when the determination step does not pass.
[Item 8]
The compounding composition data Y and the corresponding color data X of the composition containing one or more kinds of colorants registered in the database include actual measurement data or data calculated based on the actual measurement data and the actual measurement data. Item 8. The method according to any one of Items 1 to 7.
[Item 9]
Item 2. The method according to any one of Items 1 to 8, which is used when repair painting a vehicle.

[Item 10]
A database in which at least the compounding composition data Y and the corresponding color data X of the composition containing one or more kinds of color materials are registered, a computer in which the color matching calculation logic using the data registered in the database is operated, and a computer. A computer toning system comprising the following means S301 to S311.
(S301) Means for inputting training data to the computer using the data registered in the database (S302) An artificial intelligence model in which the training data is machine-learned and color data X is estimated from the compounding composition data Y. the at least one means for generating (S303) means for obtaining a target color data _{X t} colors that the target (S304) the target color data _{X t,} using the means (S305) a computer to be input to the computer search, with obtaining the target color data X _t approximation to find the color data X _n1 and search color approximation formulation composition data Y _n1 corresponding to the data X _n1, comparing said target color data X _t and the search color data X _n1 Means for determining pass / fail (S306) After obtaining candidate compounding composition data Y _{ni which is predicted to give target color data X t} when the means S305 does not pass, at least one of the above trainings was performed. using an artificial intelligence model and / or the non-artificial intelligence model prediction formula, together with obtaining the predicted color data X _ni predicted from the candidate blend composition data Y _ni, and the color data X _t and the predicted color data X _ni comparing, if not pass the means for determining the acceptability (S307) said means S306, after obtaining the candidate formulation composition data Y _ni predicted to give a target color data X _t using a computer, said at least using species learned artificial intelligence model and / or prediction expression other than the artificial intelligence model, along with obtaining a predicted color data X _ni predicted from the candidate blend composition data Y _ni, the predicted color and the color data X _t Means for comparing with data X _ni and repeating the means for determining pass / fail until it passes (S308) Means _{for obtaining pass compounding composition data Y C1} when any of the means S305 to S307 passes (S309). Means for preparing the actual candidate paint CM _Ci based on the accepted compounding composition data Y _C1 , obtaining the coated plate of the actual candidate paint CM _Ci , and acquiring the measured color data X _Ci (S310) The color data X _t and the above. Means for determining pass / fail by comparison with actual measurement color data X _Ci and / or comparison between the target color and the color _{of the coating plate of the actual candidate paint CM Ci} (S311) When the means S310 does not pass. Means that repeat the means S306 to S310.

[Item 11]
A database in which at least the compounding composition data Y and the corresponding color data X of the composition containing one or more kinds of color materials are registered, a computer in which the color matching calculation logic using the data registered in the database is operated, and a computer. A system for predicting color data of a coating film using an apparatus including the following means S401 to S409.
(S401) Means for inputting training data to the computer using the data registered in the database (S402) An artificial intelligence model in which the training data is machine-learned and color data X is estimated from the compounding composition data Y. Means for generating at least one type (S403) Means for obtaining compounding composition data Y _{CM of paint CM t} for predicting color data of a coating film (S404) Means for inputting the compounding composition data Y _CM into the computer (S405) If necessary, means for acquiring search color data X _{n1 corresponding to the compounding composition data Y CM} by a search using a computer (S406) When the search color data X _n1 corresponding to the means S405 is not searched. or, when not subjected to said means S405, from the blend composition data Y _CM, wherein at least one of the learned artificial intelligence model, or, with the at least one learned artificial intelligence model than the artificial intelligence model _{Means for obtaining predicted color data X m1} using the prediction formula of (S407) If necessary, _{the measured color data X CM} of the coated plate coated with the _{paint CM t} is acquired, and the predicted color data X _m1 is obtained. Means for comparison [Item 12]
Item 10. The system according to Item 10 or 11, wherein the system includes an automatic blending means for automatically blending and toning blending based on the obtained blending composition data.
[Item 13]
The application software for controlling and operating the system according to any one of Items 10 to 12.

According to the present invention, it is a method for manufacturing a paint for obtaining a wide variety of paint colors including bright colors whose optical characteristics are difficult to predict, and the number of trial manufactures is small regardless of the skill level of the operator. Provided is a method for producing a paint based on computer toning, which can be completed with.
According to the present invention, there is provided a method for predicting the color data of a coating film capable of predicting the color data of the coating film of a coating film having a wide variety of compositions including a bright pigment and the like with high accuracy.
According to the present invention, it is a computer toning system for preparing a paint for obtaining a wide variety of paint colors, including bright colors whose optical characteristics are difficult to predict, and it depends on the skill level of the operator. Regardless, a computer toning system that can finish toning with a small number of trial productions is provided.
According to the present invention, there is provided a system for predicting the color data of a coating film capable of predicting the color data of the coating film of a coating film having a wide variety of compositions including a bright pigment and the like with high accuracy.
As a result, the burden on the worker is reduced by reducing the work time and the number of man-hours such as the number of tonings, the reduction of waste and energy saving by reducing the number of paints to be produced, and stable toning regardless of the skill of the worker. It is extremely useful in industry because it is possible to obtain effects such as improvement of workability by being able to perform color tone prediction.

Schematic configuration diagram showing an embodiment of an apparatus used in the method according to the present invention. Schematic block diagram showing another embodiment of the apparatus used in the method according to the present invention. Schematic configuration diagram of a neural network in the artificial intelligence model of the present invention Schematic configuration diagram showing an embodiment of variable angle color measurement by the multi-angle spectrophotometer of the present invention. Schematic configuration diagram showing another embodiment of angled color measurement by the multi-angle spectrophotometer of the present invention. A flowchart showing an embodiment of a coating material manufacturing method based on computer toning according to the present invention. A flowchart showing an embodiment of a method for predicting color data of a coating film according to the present invention.

The configuration and operation of the method, system and application software of the present invention will be described in detail with reference to FIGS. 1 to 7 and by embodiments exemplified below. The means and processes herein are not limited as long as they can perform the operations, functions or processes described for those means and processes. Further, the present invention is not limited by the embodiments exemplified below as long as it does not deviate from the gist of the present invention.

[Device]
The method for producing a paint based on computer toning and the method for predicting the color data of a coating film of the present invention are a database in which at least the color data X and the compounding composition data Y of a composition containing one or more kinds of color materials are registered. A device including a computer on which a color matching calculation logic using the data registered in the database is operated is used.
Further, the computer toning system of the present invention and the system for predicting the color data of the coating film include a device including the database and a computer on which the color matching calculation logic operates.

1 and 2 are schematic configuration diagrams showing embodiments of an apparatus used in a method for producing a paint based on the computer toning of the present invention and a method for predicting color data of a coating film. This device is also used as a device included in the computer toning system of the present invention and the system for predicting the color data of the coating film.
As shown in FIGS. 1 and 2, the apparatus D includes a database 1 and a computer 2.
In the present invention, the apparatus D may include two or more databases 1 and may include two or more computers 2. Further, the database 1 and the computer 2 may be integrated.
Further, the device D used in the present invention includes an input device 31 (32), an output device 41 (42), a display device 51 (52), a colorimeter 61 (62), and an imaging device 71 (72), if necessary. ), (Automatic) compounding machine 81 (82) and the like, it may be provided with one or more devices having one or more functions. Further, one or more of these devices may be integrated with a database and / or a computer.

Here, for the database 1, for example, a known recording device, server, or the like can be used. Further, as the computer 2, a commercially available personal computer, a mobile terminal, a smartphone or the like can be used.
The input device 31 (32) is a known input device such as a keyboard, a touch panel, and a reading device, and the output device 41 (42) is a known output device such as a printing device or a data writing device. As 52), a known display device such as a display can be used.
The colorimeter 61 (62) is a known colorimeter such as a (multi-angle) spectrophotometer, a colorimeter, and a color difference meter, and the imaging device 71 (72) is a CCD camera, a solid-state image sensor, etc. ) A known imaging device such as an infrared spectroscopic imaging device can be used, and as the (automatic) compounding machine, a known (automatic) compounding machine including an electronic balance device or the like can be used.
In the apparatus used in the present invention, the database, the computer and the device are connected so as to be able to send and receive data to and from each other by a communication means of wire, wireless or a combination thereof, or a means via a recording medium. .. Examples of the communication means include one or more combinations of various communication networks such as LAN (local area network), WAN (wide area network), the Internet, and a telephone network.

FIG. 1 is an example of a device in which one database 1 and one computer 2 are connected so as to be able to send and receive data to and from each other. One or more of the input device 31, the output device 41, the display device 51, the colorimeter 61, the imaging device 71, and the (automatic) compounding machine 81 may be connected to the database 1, and the computer 2 may be connected to the computer 2. Also, one or more of the input device 32, the output device 42, the display device 52, the colorimeter 62, the imaging device 72, and the (automatic) compounding machine 82 may be connected. One or more of the colorimeters 61, 62, the imaging devices 71, 72, and the (automatic) compounding machines 81, 82 connected to the database 1 or the computer 2 are measured and compounded by a command from the database 1 or the computer 2. And so on. Further, the measured data or the like can be transmitted to the database 1 or the computer 2, and finally the data can be registered in the database.
In the device of FIG. 1, the database 1 may be formed in a recording device in the computer 2. In such a case, the device alone does not communicate with the database 1 and is an independent computer. It is possible to perform toning and prediction of color data of the coating film. In addition, the database can be maintained by updating (updating) the data registered in the database at an appropriate timing as needed, so that the worker can perform the work based on the latest data. Can be done.

FIG. 2 is an example of a device in which two or more computers 21 to 2X are connected to one database 1. FIG. 2 shows an example in which four computers 21 to 24 are connected. The database 1 may be one in which two or more databases 1 are communicably connected. By increasing the number of databases 1, it is possible to increase the number of computers that can be connected in a communicable manner.
In FIG. 2, an input device 31, an output device 41, and a display device 51 may be connected to the database 1, and any one or more of a colorimeter, an imaging device, and an (automatic) compounding machine (not shown). Equipment may be connected.
In FIG. 2, computers 21 to 24 are connected to one or more of an input device 32, an output device 42, a display device 52, a colorimeter 62, an imaging device 72, and an (automatic) compounding machine 82. You may.
The device of FIG. 2 corresponds to a device in which a plurality of computers 2 are connected by using the database 1 as a server. For example, a computer 2 of a worker (user) can be connected to a database 1 managed by a paint company or the like through a communication line (for example, an internet line, a telephone line, etc.) so that data communication can be performed.

FIG. 3 is a schematic configuration diagram showing a neural network 9 (a programmatic reproduction of the function of nerve cells in the brain) in an artificial intelligence model that estimates color data X from compounding composition data Y. The artificial intelligence model is generated by inputting the learning data to the computer and machine learning the learning data using the data registered in the database. As shown in FIG. 3, the neural network 9 is configured to include three processing layers (three neuron layers) of an input layer 91, a hidden layer 92, and an output layer 93.
The input layer 91 contains at least 1 to i processing elements called input nodes 911 to 91i and is coupled to hidden nodes 921 to 92j in the hidden layer of the network. Each unit of the input layer 91 corresponds to one or more types of feature amounts related to the compounding composition data Y.
The hidden layer 92 has at least 1 to j processing elements called hidden nodes 921, and is coupled to the output node 931 of the output layer 93 of the network. Each hidden layer 92 (hidden nodes 921 to 92j) exists between the input layer 91 (input nodes 911 to 91i) and the output layer 93 (output nodes 913 to 93k). The number of hidden nodes 921-92j can be varied by increasing the number of hidden nodes added to the network function to model the complexity of the I / O relationship.
The output layer 93 is organized to have at least 1 to 1 processing elements called output nodes 931-93k. The processing elements or nodes are interconnected so that the relationship between the formulation data and the color data can be calculated during network execution. Each unit of the output layer 93 corresponds to each component amount of one or more types related to the color data X.
In the neural network 9, data flows in only one direction, and each node only transmits a signal to one or more nodes and does not receive feedback.

In the artificial intelligence model, the input nodes 911 to 91i in the input layer 91 correspond to one input variable (input element; parameter) in each compounding composition data by one input node. Further, the output nodes 913 to 93k in the output layer 93 correspond to one output node (output element; parameter) in each color data.
In the artificial intelligence model, the number of hidden nodes 921 to 92j in the hidden layer 92 can be increased or decreased depending on the complexity of the input / output relationship. Each connection between each input node of the input layer 91, between an input node and a hidden node, between a hidden node and an output node, and between each output node of the output layer 93 has a connection weight associated therewith, and further. Each of the hidden node 92 and the output node 93 may have one or more additional threshold weights.
Artificial intelligence using neural networks 9 is particularly advantageous in complex systems or phenomena where analysis is complex and deriving models from human knowledge for use in computer specialized systems can be a daunting task.
The neural network 9 can be generated on the server computer side that constitutes the database in the device as shown in FIG. 2, for example. As a result, the connected worker (user) can receive high-quality data at any time regardless of geographical requirements and the like. Further, by improving the security of the server computer, it is possible to prevent data modification, neural network destruction, etc. due to unauthorized access.

<Database>
The database in the present invention is configured by registering (recording) color data and compounding composition data in association with each other. In addition to color data and compounding composition data, various data related to color or composition can be registered in the database of the present invention in association with each other. In the present invention, the data registered in the database is preferably a very large number of data (so-called big data). Specifically, it is 5,000 or more, 10,000 or more, and more preferably 20,000 or more. These data can be added, changed, and deleted at will.
The compounding composition data Y and the corresponding color data X of the composition containing one or more kinds of colorants registered in the database include actual measurement data or data calculated based on the actual measurement data and the actual measurement data. Examples of the data calculated based on the measured data include various parameter values calculated using the measured data and using a predetermined mathematical formula, and predicted values that can be calculated based on the measured data.
The database may be a single database, or may be a plurality of databases associated with each other by at least one common information element or a plurality of databases not associated with each other.
The database can be installed on a server that can communicate with a computer and can be operated remotely. In addition, at least a part of the database is provided in a recording unit (memory, etc.) of a computer, a colorimeter, a micro-brilliance measuring device, an automatic compounding machine, and other devices that acquire or use data registered in the database. be able to.
When there are a plurality of databases, each database may be connected by wire or wirelessly. In addition, the database may be connected by wire or wirelessly to one or more of a computer, a colorimeter, a micro-brightness measuring device, an automatic compounding machine, and other devices that acquire or use data registered in the database. ..

(Composition containing one or more kinds of coloring materials)
The coloring material contained in the composition containing one or more kinds of coloring materials of the present invention causes the composition to develop a color, for example, a coloring pigment, a dye, a bright pigment (light reflecting pigment, light interfering pigment, etc.). It is a material with a function.
The composition of the present invention may be a paint obtained by mixing raw materials including one or more primary color paints and color-matching them to a desired color. In addition, the composition of the present invention may contain a coloring pigment paste, a bright pigment paste, an orientation control agent, a gloss control agent, and various additives used in the field of paints and the like.

(Color data)
In the present invention, the color data registered in the database includes data related to color and data related to appearance characteristics such as texture, brilliance, and luster. These data can be obtained by measuring a coating film obtained from a composition containing one or more kinds of coloring materials using an instrument such as a colorimeter. Further, at least a part of various color data obtained by the measurement may be calculated by arithmetic processing. The data obtained by measuring with the device may be corrected for errors due to measurement fluctuations or the like between the measuring devices, if necessary.
Further, the K value (light absorption coefficient) and the S value (light scattering coefficient) of the composition itself may be used as color data. The K value and the S value can be obtained, for example, by numerically processing the composition and the color measurement data of the diluted color of the composition.
The data related to color and / or the data related to appearance characteristics are, for example, measuring devices such as a colorimeter, a multi-angle spectrophotometer, a laser metallic sensation measuring device, a variable angle spectrophotometer, a gloss meter, and a micro brilliance measuring device. It is either directly acquired by the measurement using or calculated from the data acquired by the measurement.
Further, the data related to color and / or the data related to appearance characteristics can include data related to one or more illumination angles, one or more observation angles, or a combination thereof.

Among the color data registered in the database, data related to color includes data representing lightness, saturation, and hue, or data in which color can be specified by calculation. For example, XYZ color system (X, Y, Z value), RGB color system, L * a * b * color system (L *, a *, b * value), Hunter Lab color system (L, a) , B value), L * C * h color system (L *, C *, h value) specified in CIE (1994), Mansell color system (H, V, C value), etc. Can be based. In the present invention, among the color data registered in the database, the data related to color can be data based on any one or more color systems. Preferably, the data is based on the L * a * b * color system or the L * C * h color system, which is widely used in various fields including the repair coating field.
Among the color data registered in the database, the data related to the appearance characteristics is, for example, the texture felt when observing the surface to be measured such as a coating film containing a bright pigment, which is macroscopically observed. Macroscopic brilliance, which is the perceived texture, micro-brilliance, which is the texture perceived by microscopic observation, depth sensation (depth), sharpness, etc. can be mentioned.

-Macro brilliance-
Examples of the macroscopic brilliance include multi-angle spectral reflectance obtained by irradiating the surface to be measured with uniform light, receiving the reflected light at each angle, and measuring the color. In addition, the FF value (flipflop value), which represents the flipflop phenomenon in which the color (brightness, saturation, hue) changes depending on the illumination and observation angle when the color measurement surface is observed from a long distance, is the opposite of the incident light. The IV value (intensity value value) indicating the visual brightness of the highlight side between 10 and 25 degrees of the opening angle from the positively reflected light appearing on the side, and the SV value (scatter) indicating the brightness of the front of the highlight side. value value), cFF value, metal feeling index, depth feeling index, sharpness, gloss value indicating gloss, and the like.
The macroscopic brilliance can be directly obtained by using, for example, a multi-angle spectrophotometer, a laser metallic sensation measuring device, a variable-angle spectrophotometer, a gloss meter, or the like, and can be calculated from these. For example, as the multi-angle spectrophotometer, BYK-Mac i (trade name, manufactured by BYK), MA-68II (trade name, manufactured by X-Rite) and the like can be used. The FF value, IV value and SV value can also be obtained by using the laser type metallic feeling measuring device Alcorp (registered trademark) LMR-200 (trade name, manufactured by Kansai Paint Co., Ltd.).

The multi-angle spectral reflectance is a spectral reflectance measured by a spectrophotometer capable of multi-angle colorimetric measurement, and is represented by R (x, λ). Here, R is the spectral reflectance (Reflectance), and is represented by the spectral reflectance% calibrated by the calibration plate attached to the measuring instrument. x is the light receiving angle and is represented by the declination from the specularly reflected light. λ is a wavelength, and the visible light range of 400 to 700 nm is measured at 10 nm intervals (number of wavelengths: 31). The angle of incidence is -45 degrees, which is generally standard.
In the present invention, the light receiving angle x is shaded from highlight (25 degrees, 15 degrees, -15 degrees) and face (45 degrees) when the incident angle is 45 degrees. ) (75 degrees, 110 degrees), any one angle or more, preferably three angles or more. Preferably, as shown in FIG. 4, the light receiving angle can be 6 angles of -15 degrees, 15 degrees, 25 degrees, 45 degrees, 75 degrees and 110 degrees. Further, as shown in FIG. 5, the light receiving angles can be set to 5 angles of 15 degrees, 25 degrees, 45 degrees, 75 degrees and 110 degrees.

The FF value (flip-flop value) indicates the degree of change in the L value (brightness) depending on the observation angle (light receiving angle). The flip-flop corresponds to the difference in brightness between the highlight direction (the direction closer to the specular reflection direction of light) and the shade direction (the direction away from the specular reflection direction of light). The larger the FF value, the larger the change in the L value (brightness) depending on the observation angle (light receiving angle), indicating that the flip-flop property is excellent.
The FF value is determined by irradiating the surface to be colored with light from an angle of 45 degrees using a multi-angle spectrophotometer, and measuring the L value (brightness) of a light receiving angle of 15 degrees and a light receiving angle of 110 degrees. It can be calculated by the formula of.
FF value = L value with a light receiving angle of 15 degrees / L value with a light receiving angle of 110 degrees.

The IV value is a Y value in the XYZ color system obtained from the spectral reflectance measured from the direction in which the declination is 15 degrees.
The SV value is a Y value in the XYZ color system obtained from the spectral reflectance measured from the direction in which the declination is 45 degrees.
The FF value is the XYZ color obtained from the spectral reflectance measured from the direction of the declination of 45 degrees, where the Y value in the XYZ color system obtained from the spectral reflectance measured from the direction of the declination of 15 degrees is Y15. It is obtained by 2 × (Y15-Y45) / (Y15 + Y45), where the Y value in the system is Y45.
The cFF value is the spectral reflection measured from the direction of the declination of 45 degrees, where the c * value in the L * c * h color system obtained from the spectral reflectance measured from the direction of the declination of 15 degrees is c * 15. It is obtained by 2 × (c * 15-c * 45) / (c * 15 + c * 45), where c * 45 is the c * value in the L * c * h color system obtained from the rate.
The metal feeling index is obtained from Y15 and FF by Y15 × FF ^2.
The depth sensation index represents the depth sensation given by the brilliant pigment, and the L * and c * values in the L * c * h color system obtained from the spectral reflectance at a representative angle are L * R and c, respectively. Let * R be used, and use these to obtain c * R / L * R. For example, using a multi-angle spectrophotometer, light is irradiated from an angle of 45 degrees, and c * 75, which is a c * value at a light receiving angle of 75 degrees, and L * 75, which is an L * value at a light receiving angle of 75 degrees, are obtained. It can be measured and the depth of the shade obtained by the following formula can be used.
Shade depth = C * 75 / L * 75
The sharpness is determined by sqrt (L * R2 + c * R2) using the above L * R and c * R.

-Micro brilliance-
The micro-brilliance sensation is a sensibility regarding two-dimensional brightness non-uniformity such as glitter and particle sensation, which is expressed by a brilliant pigment in the surface to be measured when the color-measured surface is observed from a short distance.

The micro-brightness can be obtained by using a micro-brightness measuring instrument. Examples of the micro-brilliance measuring device include a light irradiation device that irradiates a bright coating surface with light, a CCD camera that forms an image by photographing the light-irradiated coating surface at an angle at which the irradiation light does not enter. Examples thereof include a micro-brightness measuring device equipped with an image analysis device connected to a CCD camera and analyzing the image.

To measure the micro-brightness of the surface to be measured using a micro-brightness measuring device, first, the surface to be measured is irradiated with pseudo (artificial) sunlight. The angle of light irradiation on the surface to be measured is usually in the range of 5 to 60 degrees, preferably 10 to 20 degrees based on the vertical line of the surface to be measured, and in particular, about 15 degrees with respect to the vertical line. Is preferable. The shape of the light irradiation region is not particularly limited, but it is usually circular, and the irradiation area on the surface to be measured is usually within the range ^{of 1 to 10,000 mm 2 of the surface to be measured.} Suitable, but not limited to this range. The illuminance of the irradiation light is usually preferably in the range of 100 to 2,000 lux.

The surface to be measured is irradiated with light, and of the reflected light based on the light, the surface to be measured to be irradiated with light is photographed with a CCD (Charge Couple Device) camera at an angle at which the specular reflected light is not incident. The imaging angle may be any angle at which the specularly reflected light does not enter, but the vertical direction with respect to the surface to be measured is particularly preferable. Further, the angle between the shooting direction of the CCD camera and the specularly reflected light is preferably in the range of 10 to 60 degrees. The measurement range of the light-irradiated surface to be measured by the CCD camera is not particularly limited as long as it is uniformly irradiated with light, but usually includes the central portion of the irradiated portion and the measurement area. Is in the range ^{of 1 to 10,000 mm 2} , preferably 10 to 600 mm ^2.

The image taken by the CCD camera is a two-dimensional image, which is divided into a large number (usually 10,000 to 1,000,000) sections (pixels, pixels), and the brightness in each section is measured. In the present invention, "brightness" means "a digital amount of digital gradation indicating the shading value of each section of a two-dimensional image obtained by taking a picture with a CCD camera, and corresponding to the brightness of the subject". .. The digital gradation, which means the brightness of each section output from the 8-bit resolution CCD camera, shows a value of 0 to 255.

In the two-dimensional image taken by the CCD camera, the section corresponding to the portion where the reflected light of the bright pigment is strong has a strong glittering feeling, so that the brightness is high, and the section corresponding to the portion where the reflected light is not so is low. Further, even in the section corresponding to the portion where the reflected light of the bright pigment is strong, the brightness changes depending on the size, shape, angle, material and the like of the bright pigment. The brightness can be displayed for each section, and in the present invention, the brightness distribution of the two-dimensional image taken by the CCD camera can be displayed three-dimensionally based on the brightness in each section. This three-dimensional distribution map of brightness is divided into peaks, valleys, and flat areas. The height and size of the peaks indicate the degree of brilliance caused by the brilliant pigment, and the higher the peaks, the more brilliant the sensation. Shown, valleys and flat areas show no or little brilliance and show reflection of light primarily by colored pigments or substrates.

The analysis of the image taken by the CCD camera can be performed by an image analysis device connected to the CCD camera. As the image analysis software used in this image analysis apparatus, for example, WinLoof (trade name: manufactured by Mitani Corporation) is suitable.
In image analysis, "glitter" (perception of irregular and fine brilliance caused by the light directly reflected from the bright pigment in the surface to be measured) and "graininess" (as much as possible glitter is expressed). When observing the surface to be measured under difficult lighting conditions, the perception generated from the irregular / non-directional pattern (random pattern) that occurs due to the orientation and overlap of the bright pigments in the surface to be measured containing the bright material). It is preferable to quantitatively evaluate each of these separately because the correlation with the sensory evaluation in visual observation is high and the variation due to individual differences at the time of measurement is small.

As a suitable method for quantitatively measuring the feeling of glitter, for example, the following measuring method can be mentioned. A two-dimensional image obtained by photographing the light-irradiated surface to be measured with a CCD camera is divided into a large number of sections, and the brightness of each section is summed over all of the sections to obtain a total value. The value is divided by the total number of sections to obtain the average brightness x, and the threshold value α is set to a value equal to or higher than the average brightness x. It is usually appropriate that the threshold value α is the sum of the average luminance x and y (y is a number of 24 to 40, preferably 28 to 36, more preferably 32).

Then, the value of the threshold value α is subtracted from the brightness of each of the above sections, and the subtracted values whose subtracted values are positive values are summed up to obtain the total volume V which is the total sum. Further, the total area S, which is the total number of compartments having brightness equal to or higher than the threshold value α (the total number of compartments having the threshold value α or higher obtained by binarizing at the threshold value α), is obtained. Since it is considered that the luminance peak can be approximated to a cone or a pyramid, the average height PHavα of the luminance peak is obtained by multiplying the value obtained by dividing the total volume V by the total area S, that is, the following equation PHavα = 3V / S.
The value obtained by.
Further, a threshold value β which is equal to or more than the average brightness x and equal to or less than the threshold value α is set. The threshold value β is equal to or less than the threshold value α, and it is usually appropriate that the average brightness x and z (z is a number of 16 to 32, preferably 20 to 28, more preferably 24).

Then, the value of the threshold value β is subtracted from each luminance of the above-mentioned section, and the subtracted values whose subtracted values are positive values are summed up to obtain the total volume W which is the total sum. In addition, the total area A, which is the total number of compartments having brightness equal to or higher than the threshold value β (the total number of compartments having the threshold value β or higher obtained by binarizing with the threshold value β), is obtained. Since the average height PHavβ of the luminance peak at the threshold value β is considered to be similar to a cone or a pyramid, the value obtained by dividing the total volume W by the total area A is tripled, that is, the following equation PHavβ = 3W / A
Can be the value obtained by.

Further, the average particle area of the optical particles can be obtained from the total area A at the threshold value β and the number C of optical particles exhibiting brightness equal to or higher than the threshold value β. In the present invention, the "optical particle" means "an independent continuum whose brightness is equal to or higher than a threshold value on a two-dimensional image". Assuming that the shape of the optical particles is a circle, the diameter D of the circle having the same area as the average particle area is calculated by the following formula.

The average tail spread ratio PSav of the luminance peak is calculated from the above PHavβ and L by the following formula PSav = D / PHavβ.
Get by.

The brightness value BV is calculated from the above-mentioned average brightness peak height PHavα and the average brightness peak spread ratio PSav obtained as described above by the following formula BV = PHavα + a · PSav.
(In the formula, a is 300 when PHavα is less than 25, 1050 when PHavα is more than 45, and when PHavα is a number of 25 to 45, the following formula a = 300 + 37. 5 × (PHavα-25)
Is the value indicated by)
Can be calculated approximately by.

In the preferred method of the present invention, the "glittering feeling" of the bright coating film can be quantitatively measured by the brilliance value BV obtained as described above, and the brilliance value BV and the "glittering feeling" by visual observation can be measured. The correlation with the sensory evaluation result is high even when the difference in the density and the difference in brightness of the bright material in the coating film is large.

The graininess is represented by the MGR value. The MGR value is one of the measures of microscopic brilliance, which is the texture when observed microscopically, and is a parameter representing the graininess in highlights (observing a multi-layer coating film from the vicinity of specular reflection with respect to incident light). Is. The MGR value is obtained by imaging a multi-layer coating film with a CCD camera at an incident angle of 15 degrees and a light receiving angle of 0 degrees, and performing two-dimensional Fourier transform processing on the obtained digital image data, that is, two-dimensional brightness distribution data. From the obtained power spectrum image, only the spatial frequency region corresponding to the grain feeling was extracted, and the calculated measurement parameters were further taken from 0 to 100, and a linear relationship with the grain feeling was maintained. It is a measured value obtained by converting so as to be obtained. The one without a grainy feeling is set to 0, and the one with the most grainy feeling is almost 100.

As a suitable method for quantitatively measuring the "grain feeling", as described above, a light-irradiated bright coating surface is photographed with a CCD camera to obtain a two-dimensional image, and the two-dimensional image is obtained in 2D. From the spatial frequency spectrum obtained by the dimensional Fourier transform, the power of the low spatial frequency component is integrated and the two-dimensional power spectrum integrated value obtained by normalizing with the DC component is obtained, and the particle feeling of the coating film is obtained from this two-dimensional power spectrum integrated value. There is a method of quantitatively evaluating.
An image of the spatial frequency spectrum when measuring the two-dimensional power spectrum integrated value obtained by extracting the low spatial frequency component from the image of the spatial frequency spectrum after the two-dimensional Fourier transform and performing integration and normalization with the DC component. The extraction region of the low spatial frequency component extracted from is preferably a region in which the linear density representing the resolution is a numerical value in the range of the lower limit value of 0 lines / mm to the upper limit value of 2 to 13.4 lines / mm. A region of 0 lines / mm to 4.4 lines / mm is suitable from the viewpoint of increasing the correlation with the sensory evaluation result of "grain feeling" by visual observation. The larger the two-dimensional power spectrum integral value, the greater the graininess.

The two-dimensional power spectrum integral value (hereinafter, may be abbreviated as "IPSL") can be obtained by the following equation.

(In the equation, ν is the spatial frequency, θ is the angle, P is the power spectrum, 0 to L is the extracted low spatial frequency region, and L means the upper limit of the extracted frequency.)
Further, based on the brilliance value BV, the following linear expression MBV = (BV-50) / 2
It is also possible to evaluate the "glittering feeling" by the value of MBV calculated by.
The value of MBV is 0 for those without a feeling of glitter, and a value of about 100 for the one with the most glittering feeling, and the larger the value is, the larger the value is. The MBV value is sometimes referred to as the HB value (Hi-light Brilliant value).

Further, the "particle feeling" can be evaluated by the value of MGR calculated by the following linear formula based on the two-dimensional power spectrum integral value (IPSL).
If the IPSL value is 0.32 or more,
MGR = [(IPSL × 1000) -285] / 2
If the IPSL value is in the range 0.15 <IPSL <0.32,
MGR = [IPSL × (35 / 0.17)-(525/17)] / 2
When the IPSL value is 0.15 or less, MGR = 0.
The value of MGR is a value in which the one having no grain feeling of the bright material is 0 and the one having the most grain feeling of the bright material is almost 100, and the one having "grain feeling" shows a larger value. The value of MGR is sometimes referred to as the HG value (Hi-light Graininess value).

Further, the micro-brightness can be evaluated by a numerical value (micro-brightness index) obtained by indexing the micro-brightness calculated by the following formula, which comprehensively expresses the micro-brightness based on the above-mentioned MBV and MGR values. ..
Micro brilliance index = (MGR + α ・ MBV) / (1 + α)
From the examination of the surface to be colored having many brilliant sensations, the value of α is preferably 1.80 to 1.40, more preferably 1.63, which matches well with the visual brilliant sensation. You can get the result. The micro-brilliance index is 0 when there is no brilliance (no brilliance or graininess), and the value with the most brilliance (the most brilliance and particle sensation) is approximately 100.

(Mixed composition data)
In the database, compounding composition data of a composition containing one or more kinds of coloring materials is registered.
The compounding composition data includes data relating to each compounding component such as one or more kinds of coloring materials, binders, additives and the like contained in the composition, and the respective compounding amounts. Further, when a commercially available product is used as the composition, the product name (product number) can be used as the compounding composition data. For example, it can be used when a commercially available product whose compounding composition data is unknown is used, or when the composition is managed by the product number.
In the present invention, the shape, chemical properties, and the like of each component such as one or more kinds of coloring materials, binders, and additives contained in the composition can be registered as compounding composition data. Examples of the shape include the shape of a coloring material (spherical, scaly, fibrous, etc.), average primary particle size, average secondary particle size, average dispersed particle size, particle size distribution, aspect ratio, thickness, and the like. Chemical properties include molecular weight, molecular weight distribution, discoloration temperature, reactivity and the like.

In the present invention, when the composition contains a bright pigment such as a light-reflecting pigment or a light-interfering pigment, the content of the light-reflecting pigment, the content of the light-interfering pigment, and the orientation of the bright pigment are controlled. The content of the orientation control agent to be used is also registered in the database as compounding composition data.
Further, the content of the colorant for each hue, the content of the light-reflecting pigment for each hue, and the content of the light-interfering pigment for each hue, which are contained in the composition, are also stored in the database as compounding composition data. To be registered in.
Here, the definition of each hue is the L * C * h table devised based on the L * a * b * color system, which was defined by the International Commission on Illumination in 1976 and adopted in JIS Z 8729. It can be done in a color system.
For example, in the L * C * h color system chromaticity diagram calculated based on the spectral reflectance when light emitted from 45 degrees to the coating film is received at 45 degrees with respect to the positively reflected light, red. The system color is defined as a color in which the hue angle h is within the range of −45 degrees or more and less than 45 degrees when the a * red direction is 0 degrees. Similarly, an orange color is a color in which the hue angle h is within the range of 45 degrees or more and less than 67.5 degrees when a * red direction is 0 degrees, and a yellow color is a hue angle h. Is within the range of 67.5 degrees or more and less than 135 degrees when a * red direction is 0 degrees, and greenish colors are when the hue angle h is a * red direction is 0 degrees. As a color within the range of 135 degrees or more and less than -135 degrees, a bluish color is within the range of -135 degrees or more and less than -45 degrees when the hue angle h is a * red direction is 0 degrees. Each is defined as a color that is.

(Painting condition data)
Painting condition data K may be further registered in the database.
The painting condition data K is all data related to painting. For example, information on painting equipment used for painting (type of painting equipment, manufacturer of painting equipment, model number, etc.), painting condition information (painting temperature, humidity at the time of painting, dry film, etc.) Thickness, paint solid content, painting distance, painting speed, etc.), painter information (name, painting skill, painting tendency, habit, etc.), drying condition information (drying temperature, drying humidity, drying equipment manufacturer, drying equipment model number), etc. can give.

<Computer>
The computer included in the device used in the present invention means an electronic device having a calculation function and an information processing function such as a supercomputer (supercomputer), a desktop computer, a laptop computer, a personal computer such as a mobile computer, a tablet terminal, and a smartphone. is doing. The computer may be installed at any place including the work site, or may be carried by an operator or the like.

The computer includes a recording unit, a calculation unit, a control unit, and may further include an input / output unit, a communication unit, and the like. Further, in the present invention, by incorporating it into a colorimeter or the like provided with recording, calculation, control, input / output functions, etc., it can be realized integrally with the colorimeter or the like. It is also possible to record color data and compounding composition data in a computer recording unit and provide a database in the recording unit.
If the database is located outside the computer, the computer is connected to the database by wire or wirelessly.
The computer further inputs various devices for measuring data related to color and / or data related to appearance characteristics, an automatic compounding machine, other arithmetic devices, a keyboard, a mouse, a bar code reader, a touch panel, an image recognition device, and the like. It may be connected to an device, a monitor screen, an output device such as a printing device, or the like by wire or wirelessly.
If necessary, application software (program) for executing the method of the present invention or for controlling and operating the system of the present invention is installed in the computer so that necessary control and operation can be performed. You may be.

<Automatic compounding machine>
The device may include an automatic compounding machine that automatically mixes and adjusts each compounding component based on the compounding composition data. The automatic compounding machine can be connected to the computer or database by wire or wirelessly.
The automatic compounding machine has at least an electronic balance that automatically weighs the weight or capacity of each compounding component such as a coloring material, and an injector that injects each weighed compounding component into the compounding machine.
By using the automatic compounding machine, high-precision weighing can be performed automatically, human error in compounding can be reduced, compounding can be performed quickly, and any amount of toning can be adjusted. Finished paint can be easily prepared. In addition, production control can be facilitated by recording the compounding work. The automatic compounding machine may be one that automates all the operations related to compounding, or may be such that the operator can perform a part of fine color adjustment and the like.

[Manufacturing method of paint based on computer toning]
FIG. 6 is a flowchart for executing the method for producing a paint based on the computer toning of the present invention. The flow shown in FIG. 6 is only one embodiment of the present invention.
The method for producing a paint based on computer toning of the present invention includes a database in which at least color data X and compounding composition data Y of a composition containing one or more kinds of coloring materials are registered, and data registered in the database. It is a method including the following steps S101 to S111 using a computer toning device including a computer on which the used color matching calculation logic operates.
Hereinafter, the steps S101 to S111 will be described in detail.

<S101 process>
The step S101 is a step of inputting learning data into the computer using the data registered in the database.

In the present invention, a composition containing one or more kinds of coloring materials and not containing a glittering pigment, learning data using compounding composition data and color data, and one or more kinds of coloring materials and 1 It is preferable to separately prepare the compounding composition data of the composition containing the brilliant pigments of more than one kind and the learning data using the color data, and input them separately.
The present inventor has found that by creating separate learning data based on the presence or absence of a bright pigment and inputting them separately, the matching rate in computer toning is significantly improved.

In the present invention, when the compounding composition data of a composition containing one or more kinds of coloring materials and one or more kinds of bright pigments is used as learning data, the content of the light-reflecting pigment and the light-interfering pigment are used. It is preferable to use one or more types of data selected from the content, the content of the orientation control agent, and the sum of one or more of them as learning data.

In the present invention, when the compounding composition data of the composition containing one or more kinds of coloring materials and one or more kinds of bright pigments is used as the learning data, the data of the content of the bright pigments for each hue is learned. It is preferable to use the data for use. Specifically, one or more types of data selected from the content of each hue of the light-reflecting pigment in the composition, the content of each hue of the light-interfering pigment, and the content of each hue of the colorant. Is preferably used as training data.
The definition of the color of the glitter pigment is the L * C * h table devised based on the L * a * b * color system, which was defined by the International Commission on Illumination in 1976 and adopted in JIS Z 8729. It can be done in a color system.

For example, in the L * C * h color system chromaticity diagram calculated based on the spectral reflectance when light emitted from 45 degrees to the coating film is received at 45 degrees with respect to the positively reflected light, red. The system color is defined as a color in which the hue angle h is within the range of −45 degrees or more and less than 45 degrees when a * red direction is 0 degrees. Similarly, the orange color is a color in which the hue angle h is within the range of 45 degrees or more and less than 67.5 degrees when the red direction is 0 degrees, and the yellow color is the hue angle h. Is within the range of 67.5 degrees or more and less than 135 degrees when a * red direction is 0 degrees, and greenish colors are when the hue angle h is a * red direction is 0 degrees. As a color within the range of 135 degrees or more and less than -135 degrees, a bluish color is within the range of -135 degrees or more and less than -45 degrees when the hue angle h is a * red direction is 0 degrees. Each is defined as a color that is.

In the present invention, when the compounding composition data of the composition containing one or more kinds of coloring materials is used as the learning data, it is preferable to use the shape data of the coloring materials contained in the composition as the learning data. Specifically, the shape of the coloring material such as a coloring pigment or a bright pigment (spherical, scaly, fibrous, etc.), the average primary particle size of the coloring material, the average secondary particle size, the average dispersed particle size, and the particle size distribution. , Aspect ratio, thickness and other shape data are preferably used as learning data.

Input to the computer can be performed by transmitting data by wire, wireless, or a combination of these communication means, or by means via a recording medium. Examples of the input using the communication means include one or more combinations of various communication networks such as LAN (local area network), WAN (wide area network), the Internet, and a telephone network. Input by means via a recording medium can be performed by reading data from a recording medium such as a magnetic recording medium, an optical recording medium, or a paper recording medium by using an appropriate reading means.

<S102 process>
The step S102 is a step of machine learning the learning data and generating at least one type of artificial intelligence model that estimates the color data X from the composition data Y. Examples of the artificial intelligence model in the present invention include a decision tree using gradient boosting, linear regression, logistic regression, simple perceptron, MLP, neural network, support vector machine, random forest, Gaussian process, Bayesian network, and k-nearest neighbor method. , And other models used in machine learning. In the present invention, it is preferable that the artificial intelligence model uses a neural network.
In the present invention, by constructing a neural network and training the neural network using the learning data input in the step S101, at least one kind of artificial intelligence model that estimates the color data X from the composition data Y is generated. can do.

The learning of the artificial intelligence model (neural network) is performed using the learning data input to the computer in the step S101. As the learning data, at least the color data X and the composition compounding data Y relating to the composition containing one or more kinds of coloring materials are used. As an algorithm of the neural network, a known backpropagation method, which is one of the supervised learning methods, can be used. A neural network by setting the learning rate (real value between 0 and 1), which is a parameter representing the learning speed, and the margin of error (real value between 0 and 1), which is the permissible value of the error of the output value in learning. To learn. Thereby, one or more kinds of feature amounts related to the blended composition data Y can be associated with one or more kinds of feature amounts related to the color data X of the coating film of the paint film based on the blended composition data Y. Using the learned network, it is possible to predict the composition composition data satisfying the color data and the color data based on the composition composition data by the feed forward calculation. The learned network can make these predictions without performing labor-intensive experimental confirmations such as cost and time.

In the present invention, in the step of generating at least one kind of the artificial intelligence model, (i) each compounding composition data Y and each color data X relating to one or more kinds of compositions not containing a brilliant pigment are used as learning data. And (ii) each compounding composition data Y and each color data X related to one or more kinds of compositions containing a brilliant pigment are used as learning data, and the artificial intelligence model is used. It is preferable to include a step of learning. At that time, in the step (ii), one or more kinds selected from the content of the light-reflecting pigment in the composition, the content of the photo-interfering pigment, the content of the orientation control agent, and the sum of one or more of them are selected. It is preferable to include a step of using the data as training data and training an artificial intelligence model.
In the step of training the artificial intelligence model, the order of the steps (i) and the step (ii) is not particularly limited, and the step (ii) may be performed after the step (i), or after the step (ii). Step (i) may be performed. In the present invention, learning data is created as separate data from the viewpoint of whether or not the composition contains a bright pigment, and by training this, an artificial intelligence model capable of making more accurate predictions can be performed. Can be generated.

In the present invention, the step of generating at least one of the artificial intelligence models is the content of the light-reflecting pigment, the content of the light-interfering pigment, and the content of the orientation control agent such as amorphous silica in the composition. And one or more kinds of data selected from the sum of one or more of them are used as training data, and it is preferable to include a step of training an artificial intelligence model. As a result, it is possible to obtain compounding composition data that can be accurately adapted to the color data of the surface to be measured, which particularly contains a bright pigment.

In the present invention, the step of generating at least one of the artificial intelligence models is the content of the light-reflecting pigment in the composition for each hue, the content of the photo-interfering pigment for each hue, and the colorant. It is preferable to include a step of learning an artificial intelligence model by using one or more kinds of data selected from the contents of each hue as learning data. As a result, it is possible to obtain compounding composition data that can be more accurately adapted to the color data of the surface to be measured, which particularly contains a bright pigment.

In the present invention, it is preferable that the step of generating at least one of the artificial intelligence models includes a step of learning the artificial intelligence model by using the shape data of the coloring material contained in the composition as learning data. This makes it possible to more accurately match the sensibilities caused by the particles in the color data of the surface to be measured.
In the present invention, the coloring material includes not only ordinary colorants such as inorganic color pigments and organic color pigments, but also bright pigments such as particulate or flake-like (scaly) glass, metal, silica, and alumina, and flakes. It contains light-coherent pigments such as glass having a shape and an interfering property (for example, silica-coated glass flakes), silica, and alumina. In addition, the shape data includes data such as shape appearance such as spherical, flake-shaped, and fibrous, particle size, particle size distribution, thickness, aspect ratio, fiber length, and fiber diameter.

<S103 process>
S103 is a step of obtaining a target color data X _t colors a target.
Examples of the target color data _Xt include color data for all colors possessed by painted products, molded products, natural structures, and the like. In particular, it is preferable to use the color data of the painted object.
The present invention has thus far computer toning was difficult, as the target color data X _t color data of the coating film containing the bright pigment can be performed accurately toned. Thus, S103 is the target color data X _t in step, it is preferable that the color data of the coating film containing the bright pigment. Of course, the target color data X _t at S103 step may be color data of the coating film not containing the bright pigment.

The elements constituting the target color data can be the same as the elements constituting the color data registered in the database. For example, it can be color data measured by an instrument or color data calculated from the color data.
As an instrument for obtaining color data, if it is an instrument capable of measuring the color of a bright coating film (metallic coating film, pearl color coating film, etc.), solid color coating film, etc. and acquiring color data, it is measured. There are no particular restrictions on the principle, the method of calculating color data using measured values, and the like, and conventionally known ones can be used. For example, a colorimeter such as a single-angle spectrophotometer, a multi-angle spectrophotometer, a colorimeter, a color difference meter, and a variable-angle spectrophotometer equipped with a light source that irradiates the surface to be measured, an image pickup device, and a micro-brilliance feeling. One or more measuring instruments such as measuring instruments and instruments such as color swatches can be used. In addition, a data processing device that processes various color data obtained from those instruments can be arbitrarily used.

Target color data X _t is the worker by using the various instruments, can be obtained by measuring the detected material directly. In addition, various instruments may be automatically acquired based on a program or the like. Further, it may be calculated based on these color measurement data.
In the present invention, performs a measurement of the color measurement surface using a multi-angle spectrophotometer, it is preferable to obtain the target color data X _t.

If the target color data X _t cannot be obtained by directly measuring the color object to be measured, the color data obtained from the product name of the color object to be measured can be used as the _{target color data X t.} can. For example, when the target color data _Xt is color data related to an automobile, the target color data can be set based on the paint data obtained from the product name, model number, year, serial number, etc. of the automobile. ..

<S104 process>
The step S104 is a step of inputting the target color data _Xt into the computer.
The input to the computer, wired, and communication means of the wireless or a combination thereof, by means through the recording medium, by receiving the transmitted data computer from various devices for measuring and / or calculating the color data X _t It can be carried out. Examples of the input using the communication means include one or more combinations of various communication networks such as LAN (local area network), WAN (wide area network), the Internet, and a telephone network. Input by means via a recording medium can be performed by reading data from a recording medium such as a magnetic recording medium, an optical recording medium, or a paper recording medium by using an appropriate reading means.
In addition, input can be performed using a keyboard, a mouse, a bar code reader, a touch panel, a voice input device, an image recognition device, or other input means connected to the computer or provided by the computer.

<S105 process>
S105 step, the search using a computer, together with obtaining the target color data _{X t} approximation to find the color data _{X n1} and search color approximation formulation composition data _{Y n1} corresponding to the data _{X n1,} the target color data _{X t} This is a step of comparing the search color data X _n1 with the search color data X n1 to determine pass / fail.

S105 step may be a step corresponding to the computer color search (CCS), among many color data registered in the database, retrieves the color data which approximates the target color data X _t, search Color It is acquired as data X _n1.
Here, the color data registered in the database is, for example, the color data of a known color sample book, the color data of a coated plate produced in the past, and the like, all of which are associated with the color data and the corresponding compounding composition data. .. Therefore, by obtaining the search color data X _n1 , it is possible to easily obtain the _{approximate compounding composition data Y n1} which is the corresponding compounding composition data.

The search color data X _n1 includes the corresponding elements constituting the _{target color data X t} for each of one or more of the elements constituting the color data (for example, each value in the L * a * b * color system). By comparison, it is possible to search and obtain data whose value difference, degree of agreement, error rate, etc. are within a certain range. The fixed range may be set by the operator based on experience or the like, or may be set by a computer.

In the step S105, when a pass / fail judgment is made by comparing _{the target color data X t} with the search color data X _n1 , one or more of the elements constituting the _{search color data X n1 and the target color data X t1 are used.} This can be done by focusing on one or more of the constituent elements and comparing each corresponding constituent element. In the pass / fail judgment, for example, a threshold value may be set for the difference, the degree of coincidence, the error rate, etc. in each component, and the pass / fail judgment may be made by the device or the operator with reference to the threshold value. At that time, weighting may be performed among the components, reflecting the viewpoint of a skilled worker and the like.

<S106 process>
S106 process, if it does not pass in the S105 step, the target color after obtaining provide data X _t and the predicted candidate blending composition data Y _ni, said at least one learned artificial intelligence model and using a computer / or by using a prediction expression other than the artificial intelligence model, along with obtaining a predicted color data X _ni predicted from the candidate blend composition data Y _ni, comparing the predicted color data X _ni and the color data X _t, This is a step of determining pass / fail.

As a method of obtaining the candidate compounding composition data Y _ni which is predicted to give the target color data X _t using a computer, for example, a method known as computer color matching (CCM), which is a color using a computer. Calculations based on combined calculation logic and calculations by mathematical optimization can be mentioned.
Calculated based on the color matching calculation logic using computer, for example, on the basis of various color data and composition formulation data corresponding to that registered in the database, and comparing the target color data X _t various color data, the difference , One or more compounding compositions considered to be the most rational are determined _{as candidate compounding composition data Y ni} by calculating so that the degree of agreement and the like are within a certain range. It can be done by using various functions that compose the calculation logic and modifying an arbitrary composition or an approximate composition in a small iterative process. At that time, a logical command in the form of a rule can be created to assist the calculation speed and the accuracy of the adjustment algorithm.

The calculation by mathematical optimization is such that the candidate compounding composition data Y _ni refers to the color data recorded in the database, for example, in the direction of reducing the error for each coordinate axis constituting the color data obtained by the approximate compounding composition. It is possible to search for components having characteristic information that acts. For example, in the L * a * b * color system, when the error is ΔL * = L * _2- L * ₁ , Δa * = a * _2- a * ₁ , and Δb * = b * _2- b * _1. In addition, when the error ΔL * on the L * axis is a positive value, a component having characteristic information acting in the direction of decreasing the value of L * 2 is searched, and the error ΔL * on the L * axis is negative. If the value is, the component having the characteristic information that increases the value of L * 2 is searched. Similarly, when the error Δa * on the a * axis is a positive value, the _{component having the characteristic information (green) that reduces the value of a * 2} is the component, and the error Δa * on the a * axis is a negative value. If the component has characteristic information (red) that increases the value of a * ₂ , the characteristic information that decreases the value of _{b * 2} when the error Δb * on the b * axis is a positive value. A component having (blue) is searched, and a component having characteristic information (yellow) that increases the value _{of b * 2} when the error Δb * on the b * axis is a negative value is searched. As a result, in the approximate blending composition, each coordinate axis of the color system constituting the color space is made to act in the direction of reducing the error, that is, by adding a component for giving predetermined characteristic information, the target is achieved. Candidate compounding composition data Y _ni that is close to the color can be obtained.
If, when the components having the characteristic information that acts in a direction to reduce the error is not found, it based on the target color data X _t, after obtaining a more suitable new approximation blending composition, obtaining a candidate formulation composition data Can be done.

In addition, the candidate compounding composition data Y _ni refers to the compounding composition data obtained by CCM with reference to corrections by the operator (for example, color data of a known color sample book, color data of a coated plate produced in the past, own experience, etc.). It can also be obtained by making corrections by the operator), making corrections using the computer, making corrections using an artificial intelligence model, and the like.
Furthermore, when the color materials or compositions that can be used at a work site such as an automobile repair shop are limited, the candidate compounding composition data Y is based only on the color materials or compositions that can be used at the work site. You can also get _ni.

The candidate compounding composition data Y _ni can be output by a display means, a printing means, or the like. In addition, it can be transmitted from the computer to a device or the like that carries out the next process without outputting.

_{When obtaining the predicted color data X ni} in the step S106, a prediction formula other than the artificial intelligence model and / or the artificial intelligence model is used. In the present invention, it is possible to switch between using an artificial intelligence model and using a prediction formula other than the artificial intelligence model.
In the present invention, the step S106 can be a step of obtaining the _{predicted color data X ni by using only one kind of artificial intelligence model.} Further, in the second and subsequent steps S106, the step of obtaining the _{predicted color data X ni can be performed without using the artificial intelligence model.
Examples of the predicted color data X ni} that can be obtained in the step S106 include a wide variety of color data recorded in the database. In the present invention, the predicted color data X _ni preferably includes multi-angle spectral reflectance and / or brilliance parameters. By including the multi-angle spectral reflectance and / or the brilliance parameter in the predicted color data X _ni, it is possible to perform more accurate toning even for a brilliant color whose optical characteristics are difficult to predict.

In the step S106, as a _{method of obtaining the predicted color data X ni} using the learned artificial intelligence model, each unit of the input layer in the neural network of the trained artificial intelligence model has a feature amount _{of the candidate compounding composition data Y ni.} Just enter. The candidate compounding composition data Y _ni input to the input layer is transmitted while being weighted between each node and each layer, and is output as color data from each unit of the output layer.

_{In the step S106, as a method of obtaining the predicted color data X ni} using a prediction formula other than the learned artificial intelligence model, various prediction formulas known in the field of toning by CCM can be used. Examples of such a prediction formula include a method using the optical density formula of Kubelka Munk, a prediction formula of the two-constant method based on Duncan's color mixing theory formula, a method using fuzzy inference, and other color data or compounding composition data. Examples include a method of comparing with a computer and indexing the degree of matching of each.

The method using the optical density formula of Kubelka Munch and the color mixing theory formula of Duncan is as follows. The light scattering coefficient and light absorption coefficient of each color material contained in the composition containing one or more kinds of color materials and the compounding ratio of each color material were obtained, and the "light absorption coefficient after color mixing" was obtained from the optical density formula of Kubelka Munk. / "Light scattering coefficient after color mixing" is calculated, and this value can be used to obtain the spectral reflectance using Duncan's color mixing theory formula. In addition, the "light absorption coefficient / light scattering coefficient" can be calculated from the spectral reflectance of the target color, and the blending ratio of each primary color paint such as a color material or composition required to match the color can be obtained. .. By performing this calculation for each wavelength of the visible spectrum, it is possible to determine the pigment compounding ratio for achieving the target color.

Here, the optical density formula of Kubelka Munch is as follows.

(K / S) _λ : Kubelka Munch's optical density function at wavelength λ K : Light absorption coefficient S : Light scattering coefficient R _λ : Reflectance at wavelength λ λ: Wavelength

Ｋ_ｍ：混色後の光吸収係数
Ｓ_ｍ：混色後の光散乱係数
Ｋ_ｉ：着色剤ｉの光吸収係数
Ｓ_ｉ：着色剤ｉの光散乱係数
Ｐ_ｉ：着色剤ｉの配合比率 Duncan's color mixing theoretical formula is as follows.

K _m : Light absorption coefficient _{after color mixing S m} : Light scattering coefficient after color mixing K _i : Light absorption coefficient of _{colorant i S i} : Light scattering coefficient of _{colorant i Pi} : Blending ratio of colorant i

The optical density formula of Kubelka Munk is obtained by calculating the ratio of the light absorption coefficient and the light scattering coefficient from the spectral reflectance. In order to perform the color mixing calculation using Duncan's color mixing theory formula, the light absorption coefficient and the light scattering coefficient It is necessary to obtain each of the light scattering coefficients. As a method for obtaining the light absorption coefficient and the light scattering coefficient, a known method can be used, and for example, a relative method or an absolute method can be used.
At this time, in order to further improve the prediction accuracy, in order to correct the influence on the measurement of the spectral reflectance due to the internal specular reflection and the difference in the refractive index generated at the interface between the resin layer and the air layer forming the paint, Sanderson's After converting to the ideal state reflectance using the formula, the color mixture calculation can be performed. In addition, iterative calculation by the Newton-Rapson method can be used as a method of adjusting the blending ratio of the colorant in order to match the blending of the colorant with the target color, and the target reflectance and the predicted reflectance are color-matched. For the evaluation of sex, the color values XYZ, L * a * b *, etc. calculated from the reflectance are used, and the convergence calculation is performed by the Newton-Rapson method while evaluating the difference between the target value and the predicted value. Alternatively, an isometric method can be used in which the convergence calculation is performed while evaluating the sum of squares of the difference between the target reflectance and the predicted reflectance.

Fuzzy inference is a method of defining ambiguity by using the membership function in fuzzy set theory. Many proposals have been made so far as a specific method of fuzzy inference, and any method may be used in the present invention. For example, the fuzzy reasoning method devised by Mandani can be used.

The pass / fail determination can be performed, for example, by individually comparing _{each element constituting the color data X t} and each element constituting the predicted color data X _ni. For example, when the color data X _t and the predicted color data X _ni include an element using the L * a * b * color system and an element obtained from the element, in addition to each of L *, a * and b *, the color difference ΔE Can be compared to make a pass / fail judgment. At that time, a threshold value for the difference, the degree of agreement, the error rate, etc. in each component may be set, and the pass / fail judgment may be performed with reference to this.
In the present invention, when the predicted color data X _ni is color data related to brilliance, it is preferable to perform pass / fail determination using multi-angle spectral reflectance and / or brilliance parameters.
When determining pass / fail, if necessary, the operator may be notified of improvements and the like for making the _{predicted color data X ni} closer to the target color data X _{t regardless of pass / fail.}

<S107 process>
In the step S107, after obtaining the candidate compounding composition data Y _{ni which is predicted to give the target color data X t} by using a computer when the step S106 does not pass, the at least one learned artificial intelligence model and the above-mentioned artificial intelligence model and the step S107. / or by using a prediction expression other than the artificial intelligence model, along with obtaining a predicted color data X _ni predicted from the candidate blend composition data Y _ni, comparing the predicted color data X _ni and the color data X _t, This is a process of repeating the process of determining pass / fail until it passes.
In the method of obtaining the candidate compounding composition data Y _{ni predicted to give the target color data X t} using a computer in the step S107, the _{candidate compounding predicted to give the target color data X t} using the computer in the step S106 It is the same as the method for obtaining the composition data Y _ni.
Further, in the step S107, the method of determining pass / fail is the same as the method of determining pass / fail in the step S106.

In step S107, when obtaining the predicted color data X _ni predicted from the candidate compounding composition data Y _ni , the same as that used in step S106, other than at least one learned artificial intelligence model and / or artificial intelligence model. Prediction formula can be used. Further, a step (means) for switching may be provided so as to use only one of at least one learned artificial intelligence model or a prediction formula other than the artificial intelligence model. In the present invention, it is preferable to use a prediction formula different from the prediction formula used when the failure is rejected. _{For example, if the predicted color data X ni} obtained by using at least one learned artificial intelligence model in the S106 step does not pass, the prediction formula other than the artificial intelligence model is used in the next S107 step. Is preferable.
The prediction formula can be switched manually, or can be set to be automatically performed when a predetermined condition is satisfied. In the present invention, it is preferable to use a prediction formula other than the artificial intelligence model at least once out of a plurality of S107 steps.

<S108 process>
_{The step S108 is a step of obtaining pass compounding composition data Y C1} that has passed any of the steps S105 to S107. In the present invention, the accepted compounding composition data Y _C1 may be output, or the data may be transmitted without being output.
The accepted compounding composition data Y _C1 can include compounding composition data of the coating composition obtained by the toning method. For example, data such as a blending ratio of a plurality of commercially available toning paints, a blending ratio of a color material component such as a pigment and a toning paint, and a blending ratio of one or more kinds of coloring materials can be given.
In addition, the accepted compounding composition data Y _C1 can include data on the components and / or the compounding amount thereof necessary for eliminating the difference between the accepted compounding composition and the approximate compounding composition and / or the candidate compounding composition. For example, the difference of one or more compounding components when the acceptable compounding composition is compared with the approximate compounding composition or the candidate compounding composition, the difference of one or more compounding components when comparing the approximate compounding composition and the candidate compounding composition, etc. One or more types of data can be mentioned. These difference data correspond to the fine color adjustment composition data used when performing fine color adjustment from a specific compound composition, and are useful for simplifying the color adjustment work.

When outputting the pass composition composition data Y _C1 , it is possible to use a monitor, a display, a mobile terminal device, a mobile phone such as a smartphone, or an arbitrary output device capable of displaying or outputting information or an image based on a signal. can. Further, an output device such as a printing device capable of displaying information or an image on an appropriate medium such as paper or plastic based on a signal can also be used.
The output of the pass blending composition data Y _C1 may be the output of the internal computer, in this case, pass blending composition data Y _C1 outputted by the internal computer, automatic compounding device, the terminal device, data recording device, data It is sent to a recording medium or the like via a communication means or the like.
Further, the accepted compounding composition data Y _C1 may be transmitted to an automatic compounding device, a terminal device, a data recording device, a data recording medium, or the like through a communication means or the like without being output.

<S109 process>
The step S109 is a step of preparing the actual candidate paint CM _{Ci based on the accepted compounding composition data Y C1} , obtaining a coated plate of the actual candidate paint CM _Ci , and acquiring the measured color data X _Ci .

The method for preparing the actual candidate paint CM _Ci is not particularly limited, and it can be carried out by a known method when preparing the paint. For example, each component constituting the actual candidate paint CM _Ci can be placed in a blending container and mixed with a stirrer, a dispersion device or the like as necessary to prepare.
_{In the present invention, the actual candidate paint CM Ci} may be prepared by transmitting the data of the accepted compounding composition calculated by the computer to an automatic compounding machine equipped with an electronic balance or the like via a wired or wireless network. .. As a result, the actual candidate paint CM _Ci can be easily prepared even if the worker is not an expert.

The method for obtaining the coated plate of the actual candidate paint CM _Ci is also not particularly limited, and can be carried out by a known method when preparing the coated plate. For example, one or more layers of toning paint are formed on the base material so as to have a hiding film thickness or more, and a clear paint film is formed on the uppermost layer so as to have a dry film thickness of, for example, 10 to 100 μm. A method of forming a coated plate in this way can be mentioned. When forming each coating film, it may be dried and cured by heating if necessary. When drying / curing by heating, it may be carried out collectively after all the coating films have been formed, or it may be carried out each time the coating film is formed. The coating plate of the actual candidate coating material CM _Ci may be produced fully automatically by using an automatic coating apparatus using a robot or the like, or may be produced by performing a part of the steps by an operator.

The base material used in the method of the present invention is not particularly limited, and the base material used for producing a test coating plate for toning can be used. For example, a metal plate, paper, a plastic film and the like can be mentioned. The size of the base material is not particularly limited as long as the color can be measured and the color tone can be visually confirmed. For example, the length of one side is generally about 5 to 20 cm. Is the target.

The process of measuring the color of the coated plate of the actual candidate paint CM _{Ci and acquiring the measured color data X Ci} is a color meter, a multi-angle spectrophotometer, a laser metallic sensation measuring device, a variable-angle spectrophotometer, a gloss meter, and a micro. It can be acquired directly by measurement using a measuring device such as a brilliance measuring device, or by calculation using the data acquired by the measurement.

<S110 process>
In the step S110, pass / fail is determined by comparing the _{color data X t} with the measured color data X _Ci and / or comparing the target color with the color _{of the coating plate of the actual candidate paint CM Ci.} It is a process.

Determination of Compliance is a color painting plate of the the color X _t that the target actual candidate paint CM _Ci, it can be performed by comparing visually the operator. Further, for example, the target color data _Xt or each element constituting the _{target color data Xt or the color data obtained by measuring the coating plate of the actual candidate paint CM Ci} with a colorimeter or the like or each element constituting the same can be obtained. , Each can be individually compared. At that time, as in the case of the pass / fail judgment in the steps S106 and S107, threshold values for the difference, the degree of agreement, the error rate, etc. in each component are set, and the pass / fail judgment is made by the device or the operator with reference to these thresholds. It may be.
In the present invention, the pass / fail judgment may be performed by combining the pass / fail judgment visually by the operator and the pass / fail judgment performed by the device or the worker based on the color data or each element constituting the same. ..
At the time of pass / fail determination, if necessary, the operator may be notified of improvements in the composition of the _{actual candidate paint CMCi regardless of pass / fail.}

If it passes in the step S110, the paint can be prepared based on the compounding composition of _{the actual candidate paint CMCi.} Further, if necessary, it is possible to add a step of finely adjusting the color by the operator without using a computer to bring the color closer to the target color.

<S111 process>
The S111 step is a step of repeating the steps S106 to S110 when the pass / fail determination in the step S110 does not pass.
In the present invention, even if the pass / fail judgment in the S110 step fails twice or more, the steps S106 to S110 can be repeated until the pass / fail judgment in the S110 step passes.

When the steps S106 to S110 are repeated, in the steps S106 and / or S107, as in the previous time, the predicted color data X _{ni is used by using at least one learned artificial intelligence model and / or a prediction formula other than the artificial intelligence model.} Can be obtained. Further, a step (means) for switching may be provided so as to use only one of at least one learned artificial intelligence model or a prediction formula other than the artificial intelligence model. In the present invention, when the predicted color data X _ni obtained by using at least one learned artificial intelligence model does not pass, it is preferable to switch to using a prediction formula other than the artificial intelligence model next.

When repeating steps S106 to S110 without passing in step S111, and _{when obtaining predicted color data X ni} in steps S106 and / or step S107, at least one type of learned artificial intelligence model and / or other than the artificial intelligence model is used. Of the prediction formulas, it is possible to switch to using the prediction formula that was not used last time. The prediction formula can be switched manually, or can be set to be automatically performed when a predetermined condition is satisfied. In the present invention, it is preferable to use a prediction formula other than the artificial intelligence model at least once out of a plurality of S106 steps and / or S107 steps.
Further, when the process does not pass in the S111 step, it is preferable to repeat the steps S105 to S111 after inputting the difference Δ _{between the predicted color data X ni} and the measured color data X _{Ci as the correction coefficient α into the computer.}
As a result, the actual candidate coating material _CMCi can be prepared within 5 times, preferably within 3 times, more preferably within 2 times in the S110 step, and a coating material capable of obtaining the desired color can be prepared.

<Application of paint manufacturing method based on computer toning>
The method for producing a paint based on the computer toning of the present invention can be used when preparing a paint in order to obtain a desired color. Further, it can be used for identification in the compounding composition of the paint and for modifying the compounding composition.
In particular, the method for producing a paint based on the computer toning of the present invention includes colored articles such as vehicles or parts thereof such as automobiles and motorcycles, large vehicles such as trucks, buses, trains and monorails or parts thereof, and other parts. It can be used to prepare repair paints to be applied to industrial products and the like. The colored article may in particular have a single-layer or multi-layer coating. In particular, the effect of the present invention can be maximized in the case of a multi-layer coating film in which a clear coating film is provided on a coating film containing a bright pigment such as a metallic coating color or a pearl glossy color.

[Method of predicting color data of coating film]
FIG. 7 is a flowchart when executing the method of predicting the color data of the coating film of the present invention. The flow shown in FIG. 7 is only one embodiment of the present invention.
The method for predicting the color data of the coating film of the present invention is a database in which at least the color data X and the compounding composition data Y of the composition containing one or more kinds of coloring materials are registered, and the data registered in the database. This is a method including the following steps S201 to S209 using a computer toning device including a computer on which the used color matching calculation logic operates.
Here, the S201 step and the S202 step are the same as the S101 step and the S102 step, respectively. Hereinafter, the steps S203 to S209 will be described in detail.

<S203 process>
The S203 step is a step of obtaining the compounding composition data Y _{CM of the paint CM t} that predicts the color data of the coating film.
_{The paint CM t} that predicts the color data of the coating film can be a paint for which the color or the color data is to be acquired without actually preparing the paint. As a result, for example, when a large number of colors are prototyped at one time, the color or color data can be easily obtained without performing the steps of preparing the paint, preparing the coating film by painting, and measuring the color data of the coating film. Can be obtained.
Blending composition data Y _CM is data relating to each type (trade name, product number, etc.) and blending amount of binders, coloring pigments, additive components, etc. contained in the paint. Specifically, it can be the same as the data related to the composition and the blending amount, which constitutes the blending composition data registered in the database.

<S204 process>
S204 step, the compounded composition data _{Y CM,} a step of inputting to the computer.
The data can be input by using the same means as the means for inputting the _{target color data Xt into the computer in the step S104.}

<S205 process>
The step S205 is a step of acquiring the search color data X _{n1 corresponding to the compounding composition data Y CM} by a search using a computer, if necessary.
S205 step may be a step similar to computer color search (CCS), among many blending composition data registered in the database, retrieves the blending composition data approximating said blend composition data Y _CM , The color data corresponding to the obtained compounding composition data is acquired as the _{search color data X n1.}
Here, the compounding composition data registered in the database is, for example, the compounding composition data of a commercially available paint, the compounding composition data of a paint produced in the past, etc., and the compounding composition data and the corresponding color data are all associated with each other. There is. Therefore, to obtain a formulation composition data that approximates the blend composition data Y _CM, it can be easily obtained the corresponding color data as search color data X _n1.

The formulation composition data for approximating the blend composition data Y _CM, when searching from a number of formulation composition data, one or more elements constituting the blending composition data (e.g., content of the pigment of a specific color, etc.) For each, it can be obtained by comparing with the corresponding element and searching for the value difference, the degree of agreement, the error rate, etc. within a certain range. The fixed range may be set by the operator based on experience or the like, or may be set by a computer.
In the present invention, the S205 step is performed as needed, and there is no problem even if it is not performed.

<S206 process>
S206 process, if the search color data _{X n1} addressed in the S205 step is not found, or, in the case of not performed the S205 step, from the blend composition data _{Y CM,} and said at least one learning artificial This is a step of obtaining _{predicted color data X m1} by using an intelligence model or at least one learned artificial intelligence model and a prediction formula other than the artificial intelligence model.

In the present invention, the predicted color data X _m1 can be obtained _{from the compounding composition data Y CM} by using at least one artificial intelligence model. _{Further, the predicted color data X m1} can be obtained by using at least one type of learned artificial intelligence model and a prediction formula other than the artificial intelligence model in combination. Here, the prediction formula other than the artificial intelligence model and the method using the artificial intelligence model, such as the method using the artificial intelligence model, can be the same as those described in the step S106.

<S207 process>
The step S207 is a step of _{acquiring the actually measured color data X CM} of the coated plate coated with the _{paint CM t} and comparing it with the predicted color data X _{m1 as needed.}
By carrying out the step S207 and feeding back the result, it becomes possible to predict the color data of the coating film with higher accuracy.
For example, if there is a large discrepancy between the measured color data X _CM and the predicted color data X _m1 , only the prediction formula other than the artificial intelligence model is used to obtain the predicted color data again and compare it _{with the measured color data X CM.} , Can give feedback.
Further, after inputting the difference Δ between the predicted color data X _m1 and the actually measured color data X _CM into the computer as the correction coefficient β, the steps S205 to S207 can be repeated.

<Application of the method of predicting the color data of the coating film>
The method of predicting the color data of the coating film of the present invention can be used to predict the color tone of the coating film when preparing a coating material used for painting such as vehicle painting.
By using the method of predicting the color data of the coating film of the present invention, it is possible to accurately predict the color for a large number of specified paint distribution compositions without having to generate a specific paint for each paint distribution composition. Become. In addition, the paint distribution composition having the smallest deviation from the designated color can be easily selected from a large number of paint distribution composition candidates. This makes it possible to prepare paints for each of a large number of paint distribution compositions, and then obtain the corresponding color data without actually applying the paint to the material to be coated to prepare a coated plate and then performing measurement.

[Computer toning system]
The computer toning system of the present invention
A database in which color data (X) and compounding composition data (Y) of a composition containing one or more kinds of coloring materials are registered, and a computer in which a color matching calculation logic using the data registered in the database is operated. A computer toning system comprising the following means S301 to S311.
(S301) Means for inputting training data to the computer using the data registered in the database (S302) An artificial intelligence model in which the training data is machine-learned and color data X is estimated from the compounding composition data Y. the at least one means for generating (S303) means for obtaining a target color data _{X t} colors that the target (S304) the target color data _{X t,} using the means (S305) a computer to be input to the computer search, with obtaining the target color data X _t approximation to find the color data X _n1 and search color approximation formulation composition data Y _n1 corresponding to the data X _n1, comparing said target color data X _t and the search color data X _n1 Means for determining pass / fail (S306) After obtaining candidate compounding composition data Y _{ni which is predicted to give target color data X t} when the means S305 does not pass, at least one of the above trainings was performed. using an artificial intelligence model and / or the non-artificial intelligence model prediction formula, together with obtaining the predicted color data X _ni predicted from the candidate blend composition data Y _ni, and the color data X _t and the predicted color data X _ni comparing, if not pass the means for determining the acceptability (S307) said means S306, after obtaining the candidate formulation composition data Y _ni predicted to give a target color data X _t using a computer, said at least using species learned artificial intelligence model and / or prediction expression other than the artificial intelligence model, along with obtaining a predicted color data X _ni predicted from the candidate blend composition data Y _ni, the predicted color and the color data X _t Means for comparing with data X _ni and repeating the means for determining pass / fail until it passes (S308) Means _{for obtaining pass compounding composition data Y C1} when any of the means S305 to S307 passes (S309). Means for preparing the actual candidate paint CM _Ci based on the accepted compounding composition data Y _C1 , obtaining the coated plate of the actual candidate paint CM _Ci , and acquiring the measured color data X _Ci (S310) The color data X _t and the above. Means for determining pass / fail by comparison with actual measurement color data X _Ci and / or comparison between the target color and the color _{of the coating plate of the actual candidate paint CM Ci} (S311) When the means S310 does not pass. Means that repeat the means S306 to S310.

Here, "a composition containing one or more kinds of coloring materials", "a database in which color data (X) and compounding composition data (Y) of a composition containing one or more kinds of coloring materials are registered" and " The "computer on which the color matching calculation logic using the data registered in the database operates" is substantially the same as the "database" and "computer" in the apparatus used in the method for manufacturing paint based on the computer toning. can do.

Further, since the means S301 to S311 correspond to the means for carrying out the steps S101 to S111 in the method for producing a paint based on the computer toning of the present invention, the steps S101 to S112 have been substantially described. It can be the same means as the means. Further, the automatic blending means can be the same as the automatic blending means by the automatic blending machine used in the method for producing a paint based on the computer toning of the present invention.

[System for predicting color data of coating film]
The system for predicting the color data of the coating film of the present invention is a database in which color data (X) and compounding composition data (Y) of a composition containing one or more kinds of coloring materials are registered, and is registered in the database. It is a system for predicting color data of a coating film including a computer on which a color matching calculation logic using the above data is operated, and includes the following means S401 to S407.
(S401) Means for inputting training data to the computer using the data registered in the database (S402) An artificial intelligence model in which the training data is machine-learned and color data X is estimated from the compounding composition data Y. Means for generating at least one type (S403) Means for obtaining compounding composition data Y _{CM of paint CM t} for predicting color data of a coating film (S404) Means for inputting the compounding composition data Y _CM into the computer (S405) If necessary, means for acquiring search color data X _{n1 corresponding to the compounding composition data Y CM} by a search using a computer (S406) When the search color data X _n1 corresponding to the means S405 is not searched. or, when not subjected to said means S405, from the blend composition data Y _CM, wherein at least one of the learned artificial intelligence model, or, with the at least one learned artificial intelligence model than the artificial intelligence model _{Means for obtaining predicted color data X m1} using the prediction formula of (S407) If necessary, _{the measured color data X CM} of the coated plate coated with the _{paint CM t} is acquired, and the predicted color data X _m1 is obtained. Means of comparison

Here, "a composition containing one or more kinds of coloring materials", "a database in which color data (X) and compounding composition data (Y) of a composition containing one or more kinds of coloring materials are registered" and " The "computer on which the color matching calculation logic using the data registered in the database operates" is substantially the same as the "database" and "computer" in the apparatus used in the method for predicting the color data of the coating film of the present invention. Can be.

Further, since the means S401 to S407 correspond to the means for carrying out the steps S201 to S207 in the method for predicting the color data of the coating film of the present invention, the steps S201 to S209 have been substantially described. It can be the same means as the means. Further, the automatic blending means can be the same as the automatic blending means by the automatic blending machine used in the method for predicting the color data of the coating film of the present invention.

[Application software]
The present invention also relates to application software for controlling and operating the computer toning system and / or a system for predicting color data of a coating film.
The application software of the present invention functions to control and operate the system of the present invention, thereby functioning to carry out the method of the present invention.
The application software of the present invention may be stored in advance in a recording device such as an HDD (Hard Disk Drive) or a flash memory possessed by the system of the present invention or a device or the like that executes each means constituting the system. .. Further, it may be installed in a device or the like by using a wireless or wired communication means, a detachable recording medium such as a DVD, a CD-ROM, or a USB memory.

Hereinafter, the present invention will be described in more detail with reference to Examples. However, the present invention is not limited to the following examples.

[Example]
86 types of primary colors and pearls were selected for each of the Retan PG80, Retan PG Hybrid Eco, Retan WB Eco EV, and Retan Eco Fleet (all trade names manufactured by Kansai Paint Co., Ltd.) series. The compounding composition data and color data were acquired and registered in the database. There are 155 types of color data at 400 to 700 nm at 5 angles measured with a multi-angle spectrophotometer (incident angle 45 degrees, light receiving angle highlights 15 degrees, 25 degrees, face 45 degrees, shade 75 degrees, 110 degrees). Reflection spectrum data was used. The training data created using the data registered in the database was input to a computer and machine-learned by a neural network to generate an artificial intelligence model that estimates color data X from compounding composition data Y.
Prepare 100 coated plates (including about 14% of 100 colors with different colors and coating compositions, metallic and / or pearl coatings) coated with a paint whose compounding composition data is unknown, and obtain each color data. At the same time, the data was input to a computer color matching device (manufactured by Kansai Paint Co., Ltd.) equipped with a trained artificial intelligence model and a prediction formula other than the artificial intelligence model at the same time, and color matching work was performed using the artificial intelligence model. The toning load during toning work was evaluated based on the toning accuracy and reduction effect. For the toning accuracy, the rate of passing the toning within 2 times was determined, and the evaluation was made on the following 1 to 5 grades. Further, the reduction effect is shown as a ratio when the number of man-hours (time) required to obtain the final pass toning composition is set to 100 in the comparative example. The results are shown in Table 1. As shown in this embodiment, it means that there was a reduction effect of 60% from the conventional toning man-hours.
5: Pass rate is 85% or more and 4: Pass rate is 75% or more and less than 85% 3: Pass rate is 65% or more and less than 75% 2: Pass rate is 55% or more and less than 65% 1: Pass rate is less than 54%

[Comparison example]
Acquired the same color primary color and pearl primary color 86 kinds of compounding composition data and color data used in the examples, and registered them in a conventional computer color matching device (manufactured by Kansai Paint Co., Ltd.) not equipped with an artificial intelligence model. did.
For the same 100 coated plates used in Example 1, a candidate compounding composition was obtained using the above computer color matching device, and color matching work was performed until a passing compounding was obtained by a skilled person for 5 years or more. This was repeated and the toning load was evaluated in the same manner as in Example 1. The results are shown in Table 1.

1 Database 2, 21-24 Computer 31, 32 Input device 41, 42 Output device 51, 52 Display device 61, 62 Colorimeter 71, 72 Imaging device 81, 82 (Automatic) Formulator 9 Neural network 91 Input layer 911 Input Node 92 Hidden layer 921 Hidden node 93 Output layer 931 Output node

Claims (13)

A database in which at least the compounding composition data Y and the corresponding color data X of the composition containing one or more kinds of color materials are registered, a computer in which the color matching calculation logic using the data registered in the database is operated, and a computer. A method for producing a paint based on computer toning, which comprises the following steps S101 to S111.
(S101) A step of inputting training data to the computer using the data registered in the database (S102) An artificial intelligence model in which the training data is machine-learned and color data X is estimated from the compounding composition data Y. the at least one step of generating (S103) to obtain the target color data _{X t} colors that the target (S104) the target color data _{X t,} the step of inputting to the computer (S105) using a computer search, with obtaining the target color data X _t approximation to find the color data X _n1 and search color approximation formulation composition data Y _n1 corresponding to the data X _n1, comparing said target color data X _t and the search color data X _n1 Step for determining pass / fail (S106) After obtaining candidate compounding composition data Y _{ni which is predicted to give target color data X t} when the step is not passed in the step S105, at least one of the above trainings was performed. using an artificial intelligence model and / or the non-artificial intelligence model prediction formula, together with obtaining the predicted color data X _ni predicted from the candidate blend composition data Y _ni, and the color data X _t and the predicted color data X _ni (S107) A step of determining pass / fail (S107) After obtaining candidate compounding composition data Y _{ni which is predicted to give target color data X t} using a computer when the result is not passed in the S106 step, at least 1 is described above. using species learned artificial intelligence model and / or prediction expression other than the artificial intelligence model, along with obtaining a predicted color data X _ni predicted from the candidate blend composition data Y _ni, the predicted color and the color data X _t A step of repeating the step of comparing with the data X _ni and determining pass / fail until it passes (S108) A step of _{obtaining pass compounding composition data Y C1} when any of the steps S105 to S107 is passed (S109). Step of preparing the actual candidate paint CM _Ci based on the accepted compounding composition data Y _{C1 , obtaining the coated plate of the actual candidate paint CM Ci} , and acquiring the measured color data X _Ci (S110) The color data X _t and the above. Step of determining pass / fail by comparison with actual measurement color data X _Ci and / or comparison of the target color with the color _{of the coating plate of the actual candidate paint CM Ci} (S111) When the process does not pass in the S110 step. In addition, a step of repeating the steps S106 to S110. The method according to claim 1, wherein in the step S103, the target color data _Xt is the color data of the coating film containing the bright pigment. If the step S111 does not pass, the steps _{S106 to S111 are repeated after the difference Δ between the predicted color data X ni} and the measured color data X _Ci is input to the computer as a correction coefficient α, claim 1 or 2. The method described in. A database in which at least the compounding composition data Y and the corresponding color data X of the composition containing one or more kinds of color materials are registered, a computer in which the color matching calculation logic using the data registered in the database is operated, and a computer. A method for predicting color data of a coating film using an apparatus comprising the above method, which comprises the following steps S201 to S209.
(S201) A step of inputting training data to the computer using the data registered in the database (S202) An artificial intelligence model in which the training data is machine-learned and color data X is estimated from the compounding composition data Y. Step of generating at least one type (S203) Step of obtaining compounding composition data Y _{CM of paint CM t} for predicting color data of coating film (S204) Step of inputting the compounding composition data Y _CM into the computer (S205) If necessary, the search using a computer, the step (S206) of obtaining search color data X _n1 corresponding to the blending composition data Y _CM said S205 if the corresponding search color data X _n1 is not found in step or, in the case of not performed the S205 step, from the blend composition data Y _CM, wherein at least one of the learned artificial intelligence model, or, with the at least one learned artificial intelligence model than the artificial intelligence model _{Step of obtaining predicted color data X m1} using the prediction formula of (S207) If necessary, _{the measured color data X CM} of the coated plate coated with the _{paint CM t} is acquired, and the predicted color data X _m1 is obtained. Process to compare The step of generating at least one of the artificial intelligence models is
(I) A step of learning an artificial intelligence model by using each compounding composition data Y and each color data X as learning data relating to one or more kinds of compositions containing no glitter pigment.
(Ii) A step of learning an artificial intelligence model by using each compounding composition data Y and each color data X as learning data relating to one or more kinds of compositions containing a brilliant pigment.
The method according to any one of claims 1 to 4, wherein the method comprises. The step of generating at least one of the artificial intelligence models is
Content of light-reflecting pigment in composition, content of photo-interfering pigment, content of orientation control agent, content of light-reflecting pigment in composition for each hue, each hue of photo-interfering pigment One or more data selected from different contents, the content of each hue of the colorant, and the sum of two or more of those contents, and / or
Shape data of the coloring material contained in the composition,
The method according to any one of claims 1 to 5, which comprises a step of training an artificial intelligence model by using the above as training data. The method according to any one of claims 1 to 6, wherein the step of switching to use a prediction formula other than the artificial intelligence model is included in the repeating step when the determination step does not pass. The compounding composition data Y and the corresponding color data X of the composition containing one or more kinds of colorants registered in the database include actual measurement data or data calculated based on the actual measurement data and the actual measurement data. The method according to any one of claims 1 to 7. The method according to any one of claims 1 to 8, which is used when repair painting a vehicle. A database in which at least the compounding composition data Y and the corresponding color data X of the composition containing one or more kinds of color materials are registered, a computer in which the color matching calculation logic using the data registered in the database is operated, and a computer. A computer toning system comprising the following means S301 to S311.
(S301) Means for inputting training data to the computer using the data registered in the database (S302) An artificial intelligence model in which the training data is machine-learned and color data X is estimated from the compounding composition data Y. the at least one means for generating (S303) means for obtaining a target color data _{X t} colors that the target (S304) the target color data _{X t,} using the means (S305) a computer to be input to the computer search, with obtaining the target color data X _t approximation to find the color data X _n1 and search color approximation formulation composition data Y _n1 corresponding to the data X _n1, comparing said target color data X _t and the search color data X _n1 Means for determining pass / fail (S306) After obtaining candidate compounding composition data Y _{ni which is predicted to give target color data X t} when the means S305 does not pass, at least one of the above trainings was performed. using an artificial intelligence model and / or the non-artificial intelligence model prediction formula, together with obtaining the predicted color data X _ni predicted from the candidate blend composition data Y _ni, and the color data X _t and the predicted color data X _ni comparing, if not pass the means for determining the acceptability (S307) said means S306, after obtaining the candidate formulation composition data Y _ni predicted to give a target color data X _t using a computer, said at least using species learned artificial intelligence model and / or prediction expression other than the artificial intelligence model, along with obtaining a predicted color data X _ni predicted from the candidate blend composition data Y _ni, the predicted color and the color data X _t Means for comparing with data X _ni and repeating the means for determining pass / fail until it passes (S308) Means _{for obtaining pass compounding composition data Y C1} when any of the means S305 to S307 passes (S309). Means for preparing the actual candidate paint CM _Ci based on the accepted compounding composition data Y _C1 , obtaining the coated plate of the actual candidate paint CM _Ci , and acquiring the measured color data X _Ci (S310) The color data X _t and the above. Means for determining pass / fail by comparison with actual measurement color data X _Ci and / or comparison between the target color and the color _{of the coating plate of the actual candidate paint CM Ci} (S311) When the means S310 does not pass. Means that repeat the means S306 to S310. A database in which at least the compounding composition data Y and the corresponding color data X of the composition containing one or more kinds of color materials are registered, a computer in which the color matching calculation logic using the data registered in the database is operated, and a computer. A system for predicting color data of a coating film using an apparatus including the following means S401 to S409.
(S401) Means for inputting training data to the computer using the data registered in the database (S402) An artificial intelligence model in which the training data is machine-learned and color data X is estimated from the compounding composition data Y. Means for generating at least one type (S403) Means for obtaining compounding composition data Y _{CM of paint CM t} for predicting color data of a coating film (S404) Means for inputting the compounding composition data Y _CM into the computer (S405) If necessary, means for acquiring search color data X _{n1 corresponding to the compounding composition data Y CM} by a search using a computer (S406) When the search color data X _n1 corresponding to the means S405 is not searched. or, when not subjected to said means S405, from the blend composition data Y _CM, wherein at least one of the learned artificial intelligence model, or, with the at least one learned artificial intelligence model than the artificial intelligence model _{Means for obtaining predicted color data X m1} using the prediction formula of (S407) If necessary, _{the measured color data X CM} of the coated plate coated with the _{paint CM t} is acquired, and the predicted color data X _m1 is obtained. Means of comparison The system according to claim 10 or 11, wherein the system includes an automatic blending means for automatically blending and toning blending based on the obtained blending composition data. Application software for controlling and operating the system according to any one of claims 10 to 12.

Images