Design assistance device, design assistance method, and program

Abstract

To provide a design assistance device which can efficiently perform optical design of anti-reflection films.SOLUTION: A design assistance device provided herein comprises: a model generation unit for generating a model that uses thickness information of an anti-reflection film as an explanatory variable and color information of light diffusely reflected by the anti-reflection film as an objective variable; and a film thickness search unit for searching for a film thickness condition under which the color information satisfies a target condition using the model.SELECTED DRAWING: Figure 4

Description

Technical Field

[0001] The present disclosure relates to a design support device, a design support method, and a program.

Background Art

[0002] In recent years, from the viewpoint of aesthetics, a method of installing a transparent substrate such as cover glass on the front surface of an image display device such as a liquid crystal display (LCD) has been used. And, in order to prevent external light from reflecting into such a transparent substrate, a transparent substrate provided with an antireflection film is known.

[0003] For example, Patent Document 1 discloses an antireflection film - attached transparent substrate having a transparent substrate having two main surfaces and, on one main surface of the transparent substrate, a diffusion layer and an antireflection film in this order from the transparent substrate side.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in the prior art, there is room for improving the efficiency of the optical design of the antireflection film. For example, since it is difficult to perform a physical simulation of the diffuse - reflected light of the antireflection film, the film - thickness conditions for which the optical characteristics of the diffuse - reflected light satisfy the target are experimentally explored, which takes a great deal of time.

[0006] One aspect of the present disclosure aims to provide a design support device capable of efficiently performing the optical design of an antireflection film in view of the above - described technical problems.

Means for Solving the Problems

[0007] A design support apparatus according to an aspect of the present disclosure includes a model generation unit that generates a model using the film thickness information of an antireflection film as an explanatory variable and the color information of diffused reflection light by the antireflection film as an objective variable, and a film thickness search unit that searches for a film thickness condition under which the color information satisfies a target condition based on the model.

Advantages of the Invention

[0008] According to an aspect of the present disclosure, the optical design of an antireflection film can be efficiently performed.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Modes for Carrying Out the Invention

[0010] Hereinafter, each embodiment of the present disclosure will be described with reference to the accompanying drawings. In the present specification and drawings, components having substantially the same functional configuration are denoted by the same reference numerals, and redundant description is omitted.

[0011] [Embodiment] One embodiment of the present disclosure is a design support system for supporting the optical design of an antireflection film. The design support system in the present embodiment has a function of optimizing the film thickness of the antireflection film so as to realize the target diffused reflection spectroscopic characteristics.

[0012] Conventionally, in order to improve the visibility of a display device, a combination of anti-glare property by diffuse reflection and reflection suppression by an anti-reflection film has been carried out. Further, in a micro LED display manufactured by connecting (tiling) a large number of small LED panels, it is required to perform an optical design of diffuse reflection so that the color deviation of each panel becomes less noticeable.

[0013] In order to perform optical design efficiently, physical simulation is utilized. However, although physical simulation is possible for specularly reflected light, diffuse reflected light is difficult to physically simulate because it changes depending on the surface shape of the base material. This means that it is technically difficult to predict the reflection characteristics from the surface shape of the base material, so tuning of the calculation formula becomes a trial-and-error method and requires a huge amount of time. Therefore, in optical design by physical simulation, it may not be possible to complete the design within a reasonable development period. For this reason, in the prior art, the film thickness of the anti-reflection film has been experimentally determined so that the optical characteristics of the diffuse reflected light fall within the target range.

[0014] When experimentally determining the film thickness of the anti-reflection film, there is a problem that it takes a lot of time for the design. For example, in an anti-reflection film having a large number of film layers, since the combination of the film thicknesses of each film layer becomes enormous, it takes time to search for the film thickness conditions. Further, for example, even if the film thickness is determined based on the optical characteristics of the diffuse reflected light, there may be a case where a film thickness condition with a noticeable color deviation is selected. Furthermore, even if a promising film thickness condition is discovered by experiment, it is impossible to evaluate whether it is the optimal solution. And when the surface unevenness state changes by changing the base material, the past verification results cannot be applied and the optical design has to be done again, and there are also such problems.

[0015] One embodiment of the present disclosure aims to efficiently perform the optical design of an antireflection film. In this embodiment, the film thicknesses of each film layer are optimized using a combinatorial optimization method so that the diffuse reflection spectral characteristics of the antireflection film meet the target. In the combinatorial optimization method, a model is used in which the film thickness information of the antireflection film is an explanatory variable and the color information of the diffuse reflected light by the antireflection film is an objective variable. On one side, according to this embodiment, the film thickness conditions that satisfy the target conditions for the color information of the diffuse reflected light can be searched with a small number of experiments. As a result, the optical design of the antireflection film can be efficiently performed.

[0016] <Outline of a transparent substrate with an antireflection film> In this embodiment, the antireflection film to be designed will be described with reference to FIG. 1. The antireflection film in this embodiment may be, for example, the one used in the transparent substrate with an antireflection film disclosed in Patent Document 1.

[0017] FIG. 1 is a cross-sectional view schematically showing an example of a transparent substrate with an antireflection film. As shown in FIG. 1, the transparent substrate 1 with an antireflection film includes a substrate 10, a barrier layer 20, an antireflection film 30, and an antifouling film 40.

[0018] The substrate 10 includes a transparent substrate 11 and a diffusion layer 12. In the example shown in FIG. 1, the diffusion layer 12 is formed on one main surface of the transparent substrate 11, and the antireflection film 30 is formed on the diffusion layer 12.

[0019] The transparent substrate 11 includes at least one of glass and resin. The transparent substrate 11 may include both glass and resin. Hereinafter, when the transparent substrate 11 includes glass, the transparent substrate is also referred to as a glass substrate. Also, when the transparent substrate 11 includes resin, the transparent substrate is also referred to as a resin substrate.

[0020] When the transparent substrate includes glass, the type of glass is not particularly limited, and glasses having various compositions can be used. The thickness of the glass substrate is not particularly limited, but may be, for example, 0.2 mm or more and 5 mm or less.

[0021] When the transparent substrate contains a resin, the type of the resin is not particularly limited, and resins having various compositions can be used. The shape of the resin substrate is preferably film-like. When the resin substrate is in the form of a film, i.e., a resin film, its thickness is not particularly limited, but it may be 20 μm or more and 300 μm or less.

[0022] The case where the transparent substrate 11 contains both glass and a resin includes, for example, a composite substrate in which a glass substrate and a resin substrate are laminated. More specifically, the transparent substrate 11 may be, for example, in a form having a resin substrate on a glass substrate.

[0023] The diffusion layer 12 is a layer having a function of diffusing specularly reflected light and reducing glare and reflection. Examples of the diffusion layer 12 include an antiglare layer in which a function of diffusing specularly reflected light (antiglare property) is imparted to a hard coat layer. One side of the antiglare layer has an uneven shape, thereby increasing the haze value and imparting antiglare property by external scattering or internal scattering.

[0024] The diffusion layer 12 may be formed directly on the transparent substrate 11, or a laminate composed of a resin substrate - antiglare layer may be prepared in advance and bonded to a glass substrate or the like to obtain a configuration in which the diffusion layer 12 is provided on a composite substrate of a glass substrate and a resin substrate. Such a laminate is one in which the diffusion layer 12 is formed on a film-like resin substrate. Specific examples of the laminate composed of a resin substrate - antiglare layer include an antiglare PET film and an antiglare TAC film.

[0025] By subjecting the transparent substrate 11 to a surface treatment, the diffusion layer 12 may be formed on the surface layer of the transparent substrate 11 itself. For example, when using a glass substrate, a method of subjecting the glass main surface to a surface treatment to form desired unevenness can be used. Specifically, a method of performing a chemical treatment on the main surface of the glass substrate, for example, a frosting treatment, can be mentioned.

[0026] In the example shown in FIG. 1, a barrier layer 20 is provided between the diffusion layer 12 and the antireflection film 30. When the transparent substrate 11 includes a resin substrate, by providing the barrier layer 20 between the transparent substrate 11 and the antireflection film 30, it is possible to suppress the influence of moisture and oxygen that attempt to penetrate from the resin substrate into the antireflection film, and there are advantages such as the optical properties being less likely to change.

[0027] The antireflection film 30 is, for example, a multilayer film having a laminated structure in which at least two dielectric layers having different refractive indexes are laminated. In the example shown in FIG. 1, the antireflection film 30 includes low refractive index layers 31 and 33 having a relatively low refractive index and high refractive index layers 32 and 34 having a relatively high refractive index, and the low refractive index layers and the high refractive index layers are alternately laminated. However, the number of layers of the antireflection film 30 is not limited to four layers, and may be three layers or less, or five layers or more.

[0028] Hereinafter, each film layer of the antireflection film 30 will be referred to as a first low refractive index layer 31, a first high refractive index layer 32, a second low refractive index layer 33, and a second high refractive index layer 34 in order from the surface side of the transparent substrate 1 with the antireflection film. Also, let the film thickness of the first low refractive index layer 31 be t1 (nm), the film thickness of the first high refractive index layer 32 be t2 (nm), the film thickness of the second low refractive index layer 33 be t3 (nm), and the film thickness of the second high refractive index layer 34 be t4 (nm).

[0029] The low refractive index layers 31 and 33 are mainly composed of an oxide of Si (SiOx). The high refractive index layers 32 and 34 are mainly composed of a mixed oxide of at least one oxide selected from Group A consisting of Mo and W and at least one oxide selected from Group B consisting of Si, Nb, Ti, Zr, Ta, Al, Sn, and In. Here, "mainly" means the component having the highest content (by mass) in each layer, for example, it means being composed of containing 70% by mass or more of the corresponding component.

[0030] The antifouling film 40 is provided on the antireflection film 30 from the viewpoint of protecting the outermost surface of the antireflection film 30. The antifouling film 40 may be provided at a location where there is a possibility of contact with a person's hand or the like, and whether or not to provide the antifouling film 40 can be selected according to the use of the transparent substrate 1 with the antireflection film.

[0031] The transparent substrate 1 with an antireflection film is suitably used for the cover glass of an image display device, particularly for the cover glass of a display obtained by tiling a plurality of displays (for example, an LED panel, etc.), such as a cover glass of a large-sized micro LED display. Alternatively, the transparent substrate 1 with an antireflection film is also suitably used when the haze required for the transparent substrate with an antireflection film is relatively high and the color change depending on the angle is more likely to be remarkable. In addition, the transparent substrate 1 with an antireflection film is suitably used for the cover glasses of various image display devices such as liquid crystal displays, organic EL displays, and electronic paper displays.

[0032] The image display device using the transparent substrate 1 with an antireflection film is suitably used, for example, for smartphones, tablet terminals, laptop personal computers, in-vehicle devices such as car navigation systems, digital advertisements such as digital signage, etc.

[0033] <Overall Configuration> The overall configuration of the design support system in this embodiment will be described with reference to FIG. 2. FIG. 2 is a block diagram showing an example of the overall configuration of the design support system.

[0034] As shown in FIG. 2, the design support system 1000 includes a design support device 100, a terminal device 200, and an experimental device 300. The design support device 100, the terminal device 200, and the experimental device 300 are connected so as to be able to perform data communication via a communication network N such as a LAN (Local Area Network) or the Internet. Note that the experimental device 300 does not have to be connected to the communication network N as long as data input / output can be performed offline using a portable storage medium or the like.

[0035] The design support device 100 is an information processing device such as a personal computer, a workstation, or a server that supports the optical design of an antireflection film. The design support device 100 receives design conditions from the terminal device 200 and outputs film thickness conditions of the antireflection film that satisfy the design conditions. The design support device 100 searches for film thickness conditions that satisfy the target diffusion reflection spectral characteristics from the film thickness conditions of the antireflection film based on a combinatorial optimization method.

[0036] The terminal device 200 is an information processing terminal such as a personal computer, a smartphone, or a tablet terminal operated by a user of the design support system 1000. The terminal device 200 transmits the design conditions specified by the user to the design support device 100. The terminal device 200 presents the film thickness conditions searched by the design support device 100 to the user.

[0037] The experimental device 300 is a measuring instrument capable of measuring the optical characteristics of diffused reflected light by the antireflection film. In the present embodiment, the experimental device 300 measures L*, a*, and b* under a D65 light source. L*, a*, and b* under a D65 light source can be measured, for example, using a spectrophotometer CM-M6 manufactured by Konica Minolta. The measuring method of L*, a*, and b* under a D65 light source is described in detail, for example, in Patent Document 1.

[0038] Note that the overall configuration of the design support system 1000 shown in FIG. 2 is an example, and there can be various system configuration examples depending on the application and purpose. For example, one or more of the design support device 100, the terminal device 200, and the experimental device 300 may be included in multiple units in the design support system 1000. For example, the design support device 100 may be realized by a plurality of computers or may be realized as a cloud computing service. For example, the design support system 1000 may be realized by a stand-alone computer. The classification of devices such as the design support device 100, the terminal device 200, and the experimental device 300 shown in FIG. 2 is an example.

[0039] <Hardware Configuration> The design support device 100 and the terminal device 200 in this embodiment may be realized by, for example, a computer. FIG. 3 is a block diagram showing an example of the hardware configuration of a computer.

[0040] As shown in FIG. 3, the computer 500 has a CPU (Central Processing Unit) 501, a ROM (Read Only Memory) 502, a RAM (Random Access Memory) 503, an HDD (Hard Disk Drive) 504, an input device 505, a display device 506, a communication I / F (Interface) 507, and an external I / F 508. The CPU 501, the ROM 502, and the RAM 503 form a so-called computer. Each hardware of the computer 500 is interconnected via a bus line 509. Note that the input device 505 and the display device 506 may be connected to the external I / F 508 for use.

[0041] The CPU 501 is an arithmetic unit that realizes the control and functions of the entire computer 500 by reading programs and data from a storage device such as the ROM 502 or the HDD 504 onto the RAM 503 and executing processing. The computer 500 may have a GPU (Graphics Processing Unit) in addition to or instead of the CPU 501.

[0042] The ROM 502 is an example of a non-volatile semiconductor memory (storage device) that can hold programs and data even when the power is turned off. The ROM 502 functions as a main storage device that stores various programs, data, etc. necessary for the CPU 501 to execute various programs installed in the HDD 504. Specifically, the ROM 502 stores boot programs such as BIOS (Basic Input / Output System) and EFI (Extensible Firmware Interface) that are executed when the computer 500 is started up, as well as data such as OS (Operating System) settings and network settings.

[0043] RAM 503 is an example of a volatile semiconductor memory (storage device) in which programs and data are erased when the power is turned off. RAM 503 is, for example, DRAM (Dynamic Random Access Memory), SRAM (Static Random Access Memory), or the like. RAM 503 provides a work area in which various programs installed in HDD 504 are expanded when executed by CPU 501.

[0044] HDD 504 is an example of a non-volatile storage device that stores programs and data. Programs and data stored in HDD 504 include an OS, which is basic software that controls the entire computer 500, and applications that provide various functions on the OS. Note that instead of HDD 504, the computer 500 may use a storage device (for example, SSD: Solid State Drive or the like) that uses a flash memory as a storage medium.

[0045] The input device 505 is a touch panel, operation keys or buttons, a keyboard or a mouse, a microphone that inputs sound data such as voice, etc., which are used by the user to input various signals.

[0046] The display device 506 is composed of a display such as a liquid crystal or an organic EL (Electro-Luminescence) that displays a screen, a speaker that outputs sound data such as voice, etc.

[0047] The communication I / F 507 is an interface for connecting to a communication network and enabling the computer 500 to perform data communication.

[0048] The external I / F 508 is an interface with an external device. Examples of the external device include a drive device 510 and the like.

[0049] The drive device 510 is a device for setting the recording medium 511. The recording medium 511 here includes media that record information optically, electrically, or magnetically, such as CD-ROMs, flexible disks, magneto-optical disks, etc. The recording medium 511 may also include semiconductor memories that record information electrically, such as ROMs and flash memories. Thus, the computer 500 can read from and write to the recording medium 511 via the external I / F 508.

[0050] Note that various programs installed in the HDD 504 are installed, for example, when the distributed recording medium 511 is set in the drive device 510 connected to the external I / F 508 and the various programs recorded on the recording medium 511 are read by the drive device 510. Alternatively, various programs installed in the HDD 504 may be installed by being downloaded from another network different from the communication network via the communication I / F 507.

[0051] <Functional Configuration> The functional configuration of the design support system 1000 in this embodiment will be described with reference to FIG. 4. FIG. 4 is a block diagram showing an example of the functional configuration of the design support system.

[0052] ≪Design Support Device≫ As shown in FIG. 4, the design support device 100 includes a design condition acquisition unit 101, a film thickness condition generation unit 102, a model generation unit 103, a model storage unit 104, a film thickness search unit 105, a measurement result acquisition unit 106, a model update unit 107, a convergence determination unit 108, and a result output unit 109.

[0053] The design condition acquisition unit 101, the film thickness condition generation unit 102, the model generation unit 103, the film thickness search unit 105, the measurement result acquisition unit 106, the model update unit 107, the convergence determination unit 108, and the result output unit 109 are realized, for example, by the processing executed by the CPU 501 on the RAM 503 by the program expanded from the HDD 504 shown in FIG. 3.

[0054] The model memory unit 104 is realized by, for example, the RAM 503 or the HDD 504 shown in FIG. 3.

[0055] The design condition acquisition unit 101 acquires the design conditions of the antireflection film. The design condition acquisition unit 101 may receive information indicating the design conditions from the terminal device 200. The design condition acquisition unit 101 may receive the design conditions input to the input device 505 of the design support device 100.

[0056] The design information at least includes film thickness range information. The film thickness range information is information indicating the range of film thickness that each film layer of the antireflection film can take. That is, the film thickness range information includes the upper limit value and the lower limit value of each of the film thicknesses t1 to t4 of each film layer of the antireflection film. Note that the film thicknesses t1 to t4 may be continuous values or discrete values as long as they are within the range of film thickness that each film layer of the antireflection film can take.

[0057] The design information may further include the convergence conditions of the combinatorial optimization method, the target conditions of the diffuse reflection spectroscopic characteristics, the number of film layers of the antireflection film, the type and physical properties of the substrate 10, the composition of each film layer, the film thickness of the film layers other than the low refractive index layer and the high refractive index layer, the model of the experimental device 300, and the like.

[0058] The film thickness condition generation unit 102 generates the film thickness conditions of the antireflection film based on the design conditions acquired by the design condition acquisition unit 101. For each film layer of the antireflection film, the film thickness condition generation unit 102 samples values within the range indicated by the film thickness range information and combines the values of each film thickness to generate a plurality of film thickness conditions. The sampling interval of the film thickness values may be determined in advance according to the computing resources. The sampling interval may be different for each film layer or the same for all film layers.

[0059] The model generation unit 103 generates a prediction model based on a plurality of film thickness conditions generated by the film thickness condition generation unit 102. The prediction model is a machine learning model that uses the film thickness information of the antireflection film as an explanatory variable and the color information of the diffused reflected light by the antireflection film as an objective variable. The film thickness information is information indicating a combination of the film thicknesses of each film layer of the antireflection film. As an example, the prediction model may be a Gaussian process regression model.

[0060] In the present embodiment, the color information of the diffused reflected light is the degree of similarity between the color of the diffused reflected light and the target color. Specifically, the degree of similarity may be a distance in a color space. Examples of the color space include the Lab color space or the XYZ color space. In the present embodiment, the Lab color space is used as the color space.

[0061] In the present embodiment, the target color is achromatic. Therefore, in the present embodiment, the color information of the diffused reflected light may be the distance between the coordinates specified by the optical characteristics (L*, a*, and b*) of the diffused reflected light and the origin (L* = 0, a* = 0, b* = 0). The distance ΔE from the origin in the Lab color space can be calculated, for example, by Equation (1).

[0062]

Equation

[0063] The model generation unit 103 samples a predetermined number of film thickness conditions from the film thickness conditions generated by the film thickness condition generation unit 102. The model generation unit 103 may sample the film thickness conditions randomly or by using an experimental design method or the like.

[0064] For the plurality of sampled film thickness conditions, the model generation unit 103 acquires measurement results obtained by measuring the optical characteristics (L*, a*, and b*) of the diffused reflected light using an antireflection film that satisfies each film thickness condition. The L*, a*, and b* of the diffused reflected light are measured by an experiment using the experimental apparatus 300.

[0065] The model generation unit 103 generates a dataset including a plurality of sampled film thickness conditions and measurement results corresponding to each film thickness condition. The model generation unit 103 generates a prediction model by learning the generated dataset.

[0066] The prediction model generated by the model generation unit 103 is stored in the model storage unit 104. The prediction model stored in the model storage unit 104 is updated by the model update unit 107.

[0067] Based on the prediction model read from the model storage unit 104, the film thickness search unit 105 searches for a film thickness condition under which the color information of the diffuse reflected light satisfies the target condition from the film thickness conditions generated by the film thickness condition generation unit 102. The film thickness search unit 105 outputs the searched film thickness condition.

[0068] In the present embodiment, the target condition includes maximizing the similarity between the color of the diffuse reflected light and the target color (that is, minimizing the distance between L*, a*, and b* of the diffuse reflected light and the origin), or making the similarity between the color of the diffuse reflected light and the target color greater than a predetermined value (that is, making the distance between L*, a*, and b* of the diffuse reflected light and the origin less than a reference value). Here, the reference value may be the distance between L*, a*, and b* of the diffuse reflected light measured with an existing antireflection film and the origin.

[0069] The target condition may include a constraint condition that the color information of the diffuse reflected light should satisfy. As an example, the constraint condition may be at least one of an upper limit value and a lower limit value of the similarity between the color of the diffuse reflected light and the target color. By using the target condition including the constraint condition, it is possible to search for a film thickness condition that satisfies the target condition from among the film thickness conditions that satisfy the constraint condition.

[0070] In this embodiment, in order to set the target color to achromatic, the film thickness search unit 105 searches for film thickness conditions under which the color of the diffusely reflected light approaches achromatic. When the color of the diffusely reflected light approaches achromatic, the influence of the diffusely reflected light by the antireflection film on visibility can be reduced. As a result, the color deviation of the transparent substrate with the antireflection film is less noticeable, and the visibility of the display using the transparent substrate is improved.

[0071] The film thickness search unit 105 searches for film thickness conditions by a combinatorial optimization method. Examples of the combinatorial optimization method include Bayesian optimization, genetic algorithm, particle swarm optimization, annealing method, local search method, etc. Among them, Bayesian optimization is more preferable as the combinatorial optimization method.

[0072] (Acquisition function of Bayesian optimization) In Bayesian optimization, the acquisition function and the search range may be switched according to the progress of the search. For example, at the beginning of the search, the search may be performed using the acquisition function EI (Expected Improvement), in the middle of the search, the search range may be narrowed and the search may be performed using the acquisition function EI, and at the end of the search, the search may be performed using the acquisition function PI (Probability of Improvement).

[0073] The acquisition function EI is a function for obtaining an objective variable that maximizes the expected value for improvement from the current objective variable value in the existing samples. By searching for the explanatory variable that maximizes the acquisition function EI from a wide search range, it can be expected that an explanatory variable that greatly improves the objective variable will be searched.

[0074] The acquisition function PI is a function for obtaining an objective variable that maximizes the probability for improvement from the current objective variable value. By searching for the explanatory variable that maximizes the acquisition function PI from the search range specified by the acquisition function EI, it is expected that an explanatory variable that improves the objective variable will be searched.

[0075] Note that the acquisition functions available for Bayesian optimization are not limited to EI and PI. Examples of acquisition functions other than EI and PI include PTR (Probability in Target Range) or UCB / LCB (Upper / Lower Confidence Bound). PTR is an acquisition function that obtains an objective function that maximizes the probability of updating the maximum value within a certain range. UCB / LCB is an acquisition function that takes into account the average value and the upper and lower limit values of the confidence interval.

[0076] (Batch Bayesian Optimization) In Bayesian optimization, batch Bayesian optimization that searches for multiple explanatory variables at once may be used. By adopting batch Bayesian optimization, experiments can be conducted under multiple film thickness conditions at once, so the time required for the experiments can be made more efficient. The number of explanatory variables to be searched in one batch Bayesian optimization may be arbitrarily determined according to the cost of the experiment.

[0077] The measurement result acquisition unit 106 acquires measurement results obtained by measuring the optical characteristics (L*, a*, and b*) of the diffuse reflected light using an antireflection film that satisfies the film thickness conditions searched by the film thickness search unit 105. The L*, a*, and b* of the diffuse reflected light are measured by an experiment using the experimental apparatus 300.

[0078] The model update unit 107 updates the prediction model stored in the model storage unit 104 based on the measurement results acquired by the measurement result acquisition unit 106. The model update unit 107 updates the prediction model by re-learning the dataset to which the film thickness conditions searched by the film thickness search unit 105 and the measurement results acquired by the measurement result acquisition unit 106 are added.

[0079] The convergence determination unit 108 determines whether or not a predetermined convergence condition is satisfied. The convergence condition may be included in the design conditions acquired by the design condition acquisition unit 101, or may be predetermined. As an example, the convergence condition may be that the film thickness condition in which the diffuse reflection spectral characteristic exceeds the reference value has been searched by the film thickness search unit 105, or that the diffuse reflection spectral characteristic has not improved under the film thickness condition searched by the film thickness search unit 105, etc.

[0080] The result output unit 109 outputs the film thickness condition searched by the film thickness search unit 105 as an optimization result. When it is determined by the convergence determination unit 108 that the convergence condition is satisfied, the result output unit 109 outputs a predetermined number of film thickness conditions predicted to have high diffuse reflection spectral characteristics among the film thickness conditions searched by the film thickness search unit 105. Specifically, the result output unit 109 outputs a predetermined number of film thickness conditions as the optimization result from the ones with smaller distances between the L*, a*, and b* of the diffuse reflected light and the origin. The result output unit 109 may output the film thickness condition predicted to have the highest diffuse reflection spectral characteristic. The result output unit 109 may output the film thickness condition searched by the film thickness search unit 105 together with the predicted value of the diffuse reflection spectral characteristic.

[0081] ≪Terminal device≫ As shown in FIG. 4, the terminal device 200 includes a design condition input unit 201, a film thickness condition presentation unit 202, a measurement result input unit 203, and a result presentation unit 204.

[0082] The design condition input unit 201, the film thickness condition presentation unit 202, the measurement result input unit 203, and the result presentation unit 204 are realized, for example, by the processing executed by the CPU 501 for the program expanded from the HDD 504 shown in FIG. 3 onto the RAM 503.

[0083] The design condition input unit 201 receives the input of design conditions according to the user's operation. The design condition input unit 201 transmits information indicating the received design conditions to the design support device 100.

[0084] The film thickness condition presentation unit 202 presents the film thickness conditions output from the design support device 100 to the user. For example, the film thickness condition presentation unit 202 displays the film thickness conditions on the display device 506 of the terminal device 200.

[0085] The user refers to the presented film thickness conditions and obtains an anti-reflection film-coated transparent substrate on which an anti-reflection film satisfying the film thickness conditions is formed. The user may create an anti-reflection film-coated transparent substrate so as to satisfy the presented film thickness conditions. Using the experimental device 300, the user measures the L*, a*, and b* of the diffuse reflected light using the obtained anti-reflection film-coated transparent substrate.

[0086] The measurement result input unit 203 receives the input of the measurement results of the L*, a*, and b* of the diffuse reflected light measured using an anti-reflection film that satisfies the film thickness conditions presented by the film thickness condition presentation unit 202 in response to the user's operation. The measurement result input unit 203 transmits the received measurement results to the design support device 100.

[0087] The result presentation unit 204 presents the optimization result output from the design support device 100 to the user. For example, the result presentation unit 204 displays the optimization result on the display device 506 of the terminal device 200.

[0088] <Processing procedure> The design support method executed by the design support system 1000 in the present embodiment will be described with reference to FIG. 5. FIG. 5 is a flowchart showing an example of the design support method.

[0089] In step S1, the design condition input unit 201 of the terminal device 200 receives the input of design conditions in response to the user's operation. Next, the design condition input unit 201 transmits information indicating the received design conditions to the design support device 100.

[0090] In step S2, the design condition acquisition unit 101 of the design support apparatus 100 receives information indicating the design conditions from the terminal device 200. Next, the design condition acquisition unit 101 acquires the design conditions from the received information indicating the design conditions. Then, the design condition acquisition unit 101 sends the acquired design conditions to the film thickness condition generation unit 102.

[0091] In step S3, the film thickness condition generation unit 102 of the design support apparatus 100 receives the design conditions from the design condition acquisition unit 101. Next, the film thickness condition generation unit 102 acquires film thickness range information from the received design conditions. Subsequently, the film thickness condition generation unit 102 generates the film thickness conditions for the antireflection film based on the film thickness range information. Then, the film thickness condition generation unit 102 sends the generated film thickness conditions to the model generation unit 103 and the film thickness search unit 105.

[0092] In step S4, the model generation unit 103 of the design support apparatus 100 receives the film thickness conditions from the film thickness condition generation unit 102. Next, the model generation unit 103 samples a predetermined number of film thickness conditions from the received film thickness conditions. Subsequently, the model generation unit 103 acquires measurement results obtained by measuring the L*, a*, and b* of the diffuse reflected light using an antireflection film that satisfies the sampled film thickness conditions.

[0093] Next, the model generation unit 103 generates a dataset including the sampled multiple film thickness conditions and the measurement results corresponding to each film thickness condition. Subsequently, the model generation unit 103 learns a prediction model based on the generated dataset. Thereby, a learned prediction model is generated. Then, the model generation unit 103 stores the learned prediction model and the dataset in the model storage unit 104.

[0094] In step S5, the film thickness search unit 105 of the design support apparatus 100 receives the film thickness conditions from the film thickness condition generation unit 102. Next, the film thickness search unit 105 reads out the prediction model from the model storage unit 104. Subsequently, the film thickness search unit 105 searches for the film thickness conditions under which the color information of the diffuse reflected light satisfies the target conditions based on the read prediction model.

[0095] In step S6, the film thickness search unit 105 of the design support apparatus 100 transmits information indicating the film thickness condition searched in step S5 to the terminal device 200.

[0096] In step S7, the film thickness condition presentation unit 202 of the terminal device 200 receives information indicating the film thickness condition from the design support apparatus 100. Next, the film thickness condition presentation unit 202 displays the received information indicating the film thickness condition on the display device 506 of the terminal device 200.

[0097] The user obtains a transparent substrate with an antireflection film formed thereon that satisfies the film thickness condition displayed on the display device 506 of the terminal device 200. Next, the user uses the experimental apparatus 300 to measure L*, a*, and b* of the diffuse reflected light using the obtained transparent substrate with an antireflection film. Thereby, the user obtains a measurement result of measuring the optical characteristics of the diffuse reflected light by the antireflection film that satisfies the film thickness condition received from the design support apparatus 100.

[0098] In step S8, the measurement result input unit 203 of the terminal device 200 receives an input of the measurement result obtained in step S7 according to the user's operation. Next, the measurement result input unit 203 transmits information indicating the received measurement result to the design support apparatus 100.

[0099] In step S9, the measurement result acquisition unit 106 of the design support apparatus 100 receives information indicating the measurement result from the terminal device 200. Next, the measurement result acquisition unit 106 acquires the measurement result from the received information indicating the measurement result. Then, the measurement result acquisition unit 106 sends the acquired measurement result to the model update unit 107.

[0100] In step S10, the model update unit 107 of the design support apparatus 100 receives the measurement result obtained in step S9 from the measurement result acquisition unit 106. Next, the model update unit 107 reads out the data set stored in the model storage unit 104. Subsequently, the model update unit 107 adds the film thickness condition searched in step S5 and the measurement result obtained in step S9 to the read data set.

[0101] Subsequently, based on the dataset with the measurement results obtained in step S9 added, the model update unit 107 learns the prediction model. As a result, the learned prediction model is updated. Then, the model generation unit 103 stores the updated prediction model and the dataset in the model storage unit 104.

[0102] In step S11, the convergence determination unit 108 of the design support apparatus 100 determines whether or not a predetermined convergence condition is satisfied. If the convergence condition is satisfied (YES), the convergence determination unit 108 proceeds to step S12. On the other hand, if the convergence condition is not satisfied (NO), the convergence determination unit 108 returns the process to step S5.

[0103] After returning to step S5, the design support apparatus 100 executes the processes from step S5 to step S11 again. Note that the design support apparatus 100 may switch the search range of the film thickness condition and the acquisition function according to the progress of the search. In this way, the design support apparatus 100 repeats the processes from step S5 to step S11 until the convergence condition is satisfied.

[0104] In step S12, the result output unit 109 of the design support apparatus 100 transmits information indicating the optimization result to the terminal device 200. The information indicating the optimization result includes the film thickness condition searched in step S5. When step S5 is executed multiple times, the information indicating the optimization result includes a predetermined number of film thickness conditions predicted to have high diffuse reflection spectroscopic characteristics. The information indicating the optimization result may also be one film thickness condition predicted to have the highest diffuse reflection spectroscopic characteristics.

[0105] In step S13, the result presentation unit 204 of the terminal device 200 receives the information indicating the optimization result from the design support apparatus 100. Next, the result presentation unit 204 acquires the optimization result from the received information indicating the optimization result. Then, the result presentation unit 204 displays the acquired optimization result on the display device 506 of the terminal device 200.

[0106] The user can utilize the optimization results displayed on the display device 506 of the terminal device 200 in the optical design of the antireflection film. For example, the user can manufacture a transparent substrate with an antireflection film by forming each film layer on the transparent substrate so as to satisfy the film thickness conditions shown in the optimization results. Further, for example, the user can manufacture an image display device using the manufactured transparent substrate with an antireflection film.

[0107] <Example> Specific examples will be described for the design support method described in the above embodiment. However, the present disclosure is not limited to the following examples.

[0108] (Antireflection film) The antireflection film to be designed is the antireflection film shown in FIG. 1. That is, it is a four-layer antireflection film in which low refractive index layers and high refractive index layers are alternately laminated. Hereinafter, the film thickness of the first low refractive index layer will be denoted as t1, the film thickness of the first high refractive index layer will be denoted as t2, the film thickness of the second low refractive index layer will be denoted as t3, and the film thickness of the second high refractive index layer will be denoted as t4.

[0109] (Initial sampling) Step 1. The search range for each film layer of the antireflection film was determined as follows. These were determined based on past knowledge or experimental data. In this case, considering combinations of film thicknesses in 1 nm units, there are 1,519,392 (about 1.5 million) cases. ·41 ≤ t1 (nm) ≤ 124 ·59 ≤ t2 (nm) ≤ 177 ·18 ≤ t3 (nm) ≤ 55 ·2 ≤ t4 (nm) ≤ 6

[0110] Step 2. Lists were generated by sampling values at a predetermined sampling interval from the above search ranges for each of the film thicknesses t1 to t4. For the film thicknesses t1 to t3, the sampling interval was 3 nm, and for the film thickness t4, the sampling interval was 1 nm.

[0111] Step 3. By combining the values in the lists of film thicknesses t1 to t4 respectively, film thickness conditions were generated. In this embodiment, the number of generated film thickness conditions is 58,240.

[0112] Step 4. In order to sample uniformly from the generated film thickness conditions, random numbers were generated by the Sobol method, and 16 film thickness conditions were sampled based on the generated random numbers. Note that the sampling method is not limited to the Sobol method, and for example, sampling may be performed by a Latin hypercube or a design of experiments method.

[0113] Step 5. Experiments were conducted using an antireflection film that satisfies the 16 sampled film thickness conditions, and L*, a*, and b* of the diffuse reflected light corresponding to each film thickness condition were measured. Then, based on the measured L*, a*, and b* of the diffuse reflected light, the diffuse reflection spectral characteristics (the distance between the coordinates specified by L*, a*, and b* and the origin) were calculated.

[0114] Step 6. Using a dataset (a combination of film thickness conditions and diffuse reflection spectral characteristics) with 16 film thickness conditions, the kernel function with the highest accuracy was determined by cross-validation. The following k1 to k4 were used as candidates for the kernel function. In cross-validation, the root mean squared error (RMSE) was calculated as the accuracy. In this embodiment, the kernel function k1 resulted in the highest accuracy. · k1: Radial Basis Function Kernel · k2: Matern Kernel · k3: Dot-product Kernel · k4: Exp-Sine-Squared Kernel

[0115] (First experiment in the initial exploration stage) Step 7. Using a Gaussian process regression model with the kernel function k1, the mean and standard deviation of the predicted values of the diffuse reflection spectral characteristics for each of the film thickness conditions generated in Step 3 were calculated.

[0116] Step 8. Using the average and standard deviation calculated in Step 7, the acquisition function EI was calculated, and the film thickness condition that maximizes the acquisition function EI was explored.

[0117] Step 9. The film thickness condition explored in Step 8 and the predicted value of the diffuse reflection spectroscopic characteristics corresponding to the film thickness condition were added to the dataset, and the Gaussian process regression model was updated. Then, Steps 7 to 8 were executed again using the updated Gaussian process regression model.

[0118] Step 10. Step 9 was executed again, and a total of three film thickness conditions and the predicted values of the diffuse reflection spectroscopic characteristics corresponding to each film thickness condition were obtained.

[0119] Step 11. Experiments were conducted using each film thickness condition obtained in Step 10, and the measured values of the diffuse reflection spectroscopic characteristics for each film thickness condition were obtained.

[0120] (Second experiment in the initial exploration stage) Step 12. The film thickness conditions obtained in Step 11 and the measured values of the diffuse reflection spectroscopic characteristics were added to the dataset, and the Gaussian process regression model was updated. Then, Steps 7 to 11 were executed again using the updated Gaussian process regression model.

[0121] (Third experiment in the middle exploration stage) Step 13. Since the range in which the diffuse reflection spectroscopic characteristics are improved was specified, the exploration ranges of the film thicknesses t1 to t4 of the antireflection film were changed as follows. ·76.5 ≤ t1 (nm) ≤ 86.5 ·116.5 ≤ t2 (nm) ≤ 124.5 ·37.5 ≤ t3 (nm) ≤ 47.5 ·2.0 ≤ t4 (nm) ≤ 4.0

[0122] Step 14. For each of the film thicknesses t1 to t4, a list was generated by sampling values with a sampling interval of 0.5 nm.

[0123] Step 15. By combining the values in the lists of each of the film thicknesses t1 to t4, a film thickness condition was generated. In this embodiment, the number of generated film thickness conditions is 37,485.

[0124] Step 16. The film thickness conditions (37,485 cases) generated in Step 15 were added to the film thickness conditions (58,240 cases) generated in Step 3, and the Gaussian process regression model was updated. Then, using the updated Gaussian process regression model, the mean and standard deviation of the predicted values of the diffuse reflection spectroscopic characteristics for each film thickness condition were calculated.

[0125] Step 17. Steps 8 to 11 were executed again using the mean and standard deviation calculated in Step 16.

[0126] (Final stage of exploration, 4th experiment) Step 18. Since the diffuse reflection spectroscopic characteristics did not improve under the film thickness conditions explored in Step 8, the acquisition function EI was changed to the acquisition function PI.

[0127] Step 19. Using the mean and standard deviation calculated in Step 16, the acquisition function PI was calculated, and the film thickness condition that maximizes the acquisition function PI was explored.

[0128] Step 20. Steps 9 to 11 were executed again using the film thickness conditions explored in Step 19.

[0129] (Exploration completed) Step 21. Since the diffuse reflection spectroscopic characteristics did not improve under the film thickness conditions explored in Step 19, the exploration was terminated. Film thickness conditions with improved diffuse reflection spectroscopic characteristics compared to the existing antireflection film could be discovered through initial sampling and four experiments.

[0130] FIG. 6 is a diagram showing the diffuse reflection spectroscopic characteristics in the examples. In FIG. 6, the vertical axis represents the diffuse reflection spectroscopic characteristics (ΔE), and the minimum value, maximum value, and average value of the diffuse reflection spectroscopic characteristics (the distance between the L*, a*, and b* of the diffuse reflected light and the origin) measured in the initial sampling and the first to fifth experiments (Experiment 1 to Experiment 5) in the examples are plotted. In this example, since the diffuse reflection spectroscopic characteristics are defined as the distance ΔE from the origin, it can be evaluated that the smaller the value (the closer to the horizontal axis), the better the diffuse reflection spectroscopic characteristics.

[0131] In this example, the experiment is performed with 16 samples in the initial sampling. Also, since three film thickness conditions are experimented per time by batch Bayesian optimization, the experiment is performed with 3 samples in each experiment.

[0132] As shown in FIG. 6, the diffuse reflection spectroscopic characteristics improved from the initial sampling to Experiment 4, and the diffuse reflection spectroscopic characteristics decreased in Experiment 5 compared to Experiment 4. It is shown that the possibility of further improvement in the diffuse reflection spectroscopic characteristics is low even if the experiment is repeated, and the optimal film thickness conditions were explored up to Experiment 4.

[0133] <Effects of the Embodiment> The design support device 100 in this embodiment searches for the film thickness conditions under which the color information of the diffuse reflected light satisfies the target conditions based on a model that uses the film thickness information of the antireflection film as an explanatory variable and the color information of the diffuse reflected light by the antireflection film as an objective variable. Conventionally, since it has been difficult to design the diffuse reflection spectroscopic characteristics of the antireflection film by physical simulation, a design that satisfies the target of the diffuse reflection spectroscopic characteristics has been experimentally explored. In one aspect, according to this embodiment, the optical design of the antireflection film can be efficiently performed.

[0134] The design support device 100 may update the model based on the color information measured using the antireflection film that satisfies the searched film thickness conditions. The design support device 100 may repeatedly execute updating the model and searching for the film thickness conditions. According to this embodiment, the film thickness conditions for improving the diffuse reflection spectroscopic characteristics can be accurately searched.

[0135] The color information of the diffused reflected light may be the degree of similarity between the color of the diffused reflected light and the target color. According to the present embodiment, in order to search for the film thickness condition based on the degree of similarity between the color of the diffused reflected light and the target color, an optical design in which the color of the diffused reflected light approaches the target color can be efficiently performed.

[0136] The target color may be achromatic. According to the present embodiment, in order to search for the film thickness condition based on the degree of similarity between the color of the diffused reflected light and the achromatic color, an optical design of an antireflection film in which the color of the diffused reflected light approaches the achromatic color and the influence of the diffused reflected light is suppressed can be efficiently performed.

[0137] The design support device 100 may search for the film thickness condition by a combinatorial optimization method. The combinatorial optimization method may be Bayesian optimization. According to the present embodiment, in order to search for the film thickness condition by a combinatorial optimization method such as Bayesian optimization, the film thickness condition can be searched efficiently and accurately.

[0138] The antireflection film may have a plurality of film layers with different refractive indices. When there are a plurality of film layers with different refractive indices, the number of combinations of film thicknesses becomes enormous, so it is extremely difficult to experimentally search for the optimal film thickness condition. According to the present embodiment, even for an antireflection film having a plurality of film layers with different refractive indices, an optical design can be efficiently performed.

[0139] [Other Embodiments] ≪Spectral Residual≫ In the above embodiment, as the diffused reflection spectral characteristic, a configuration using the color information of the diffused reflected light has been described. The color information of the diffused reflected light was taken as the degree of similarity between the diffused reflected light and the target color as an example. Specifically, the distance between the coordinates specified by L*, a*, and b* of the diffused reflected light and the origin was used.

[0140] The color information of the diffused reflected light may be, as another example, the spectral residual between the diffused reflected light and the target color. The spectral residual may be the sum of the absolute values of the differences in spectra in a predetermined wavelength range.

[0141] FIG. 7 is a diagram showing an example of a spectral residual. FIG. 7 is a reflection spectrum with the horizontal axis representing wavelength (nm) and the vertical axis representing reflectance. f(x)1 is the reflection spectrum of the diffused reflected light by the antireflection film, and f(x)2 is the reflection spectrum of the target color. Wavelength λ1 and wavelength λ2 are the upper and lower limits of a predetermined wavelength range. In the wavelength range from wavelength λ1 to wavelength λ2, the sum of the absolute values of the differences between the spectrum f(x)1 and the spectrum f(x)2 (i.e., the areas of regions d1 and d2 sandwiched between the spectrum f(x)1 and the spectrum f(x)2) is the spectral residual.

[0142] When the film thickness conditions are optimized so that the spectral residual becomes small, the color of the diffused reflected light approaches the target color. If the target color is achromatic, the color of the diffused reflected light by the antireflection film approaches achromatic, and the influence of the diffused reflected light by the antireflection film on visibility can be reduced.

[0143] <<Transfer Learning>> In the above-described embodiment, the design conditions other than the film thickness of each film layer of the antireflection film are fixed, and the film thickness conditions of the antireflection film are optimized. Examples of the design conditions other than the film thickness include the number of film layers of the antireflection film, the type and physical properties of the substrate 10, the composition of each film layer, the film thickness of film layers other than the low refractive index layer and the high refractive index layer, the model type of the experimental apparatus 300, and the like.

[0144] A prediction model learned by being fixed under certain design conditions can be used to search for the optimal film thickness conditions under other design conditions by transfer learning. Specifically, for a dataset collected under various film thickness conditions with the design conditions other than the film thickness fixed, a dataset collected by changing any one or more of the design conditions other than the film thickness is added, and a prediction model is learned based on that dataset. For example, when it is desired to change the substrate 10 to one with different type or physical properties, if the learned prediction model is subjected to transfer learning, the diffused reflection spectroscopic characteristics of the transparent substrate with the antireflection film using the new substrate 10 can be accurately predicted. As a result, the film thickness conditions of the antireflection film when using the new substrate 10 can be optimized with a small number of experimental times.

[0145] [Supplementary Explanation] In this specification, having another layer, film, etc. on the main surface of a substrate such as a transparent substrate, on a layer such as a diffusion layer, or on a film such as an antireflection film is not limited to the mode in which the other layer, film, etc. is provided in contact with the main surface, layer, or film, and any mode in which a layer, film, etc. is provided in the upper direction thereof may be sufficient. For example, having a diffusion layer on the main surface of a transparent substrate means that the diffusion layer may be provided in contact with the main surface of the transparent substrate, or any other arbitrary layer, film, etc. may be provided between the transparent substrate and the diffusion layer.

[0146] Each function of the embodiments described above can be realized by one or a plurality of processing circuits. Here, the "processing circuit" in this specification means a processor programmed to execute each function by software like a processor implemented by an electronic circuit, an ASIC (Application Specific Integrated Circuit) designed to execute each function described above, a DSP (Digital Signal Processor), an FPGA (Field Programmable Gate Array), and devices such as conventional circuit modules.

[0147] As described above in detail for the embodiments of the present disclosure, the embodiments disclosed this time are illustrative in all respects and not restrictive. The embodiments can be modified and improved in various forms without departing from the scope of the appended claims and their gist. Matters described in the plurality of embodiments above can also adopt other configurations and can be combined within a non - conflicting range.

Description of Reference Numerals

[0148] 1: Transparent substrate with antireflection film 10: Substrate 11: Transparent substrate 12: Diffusion layer 20: Barrier layer 30: Antireflection film 31: First low - refractive - index layer 32: First high - refractive - index layer 33: Second low - refractive - index layer 34: Second highest refractive index layer 40: Antifouling film 100: Design support device 101: Design condition acquisition unit 102: Film thickness condition generation unit 103: Model generation unit 104: Model storage unit 105: Film thickness search unit 106: Measurement result acquisition unit 107: Model update unit 108: Convergence determination unit 109: Result output unit 200: Terminal device 201: Design condition input unit 202: Film thickness condition presentation unit 203: Measurement result input unit 204: Result presentation unit 300: Experimental device 500: Computer 501: CPU 502: ROM 503: RAM 504: HDD 505: Input device 506: Display device 507: Communication I / F 508: External I / F 509: Bus line 510: Drive device 511: Recording medium 1000: Design support system S1: Input design conditions S2: Acquire design conditions S3: Generate film thickness conditions S4: Generate prediction model S5: Search for film thickness conditions S6: Output film thickness conditions S7: Experiment based on film thickness conditions S8: Input measurement results S9: Acquire measurement results S10: Update prediction model S11: Do the convergence conditions hold? S12: Output optimization results S13: Display optimization results d1, d2: Regions λ1, λ2: Wavelengths f(x)1, f(x)2: Spectra N: Communication network ΔE: Distance from the origin in the Lab color space

Claims

1. A model generation unit configured to generate a model having the film thickness information of the antireflection film as an explanatory variable and the color information of the diffuse reflected light by the antireflection film as an objective variable; A film thickness search unit configured to search for a film thickness condition under which the color information satisfies a target condition based on the model; A design support device comprising:

2. The design support device according to claim 1, Further comprising a model update unit configured to update the model based on the color information measured using the antireflection film that satisfies the film thickness condition searched by the film thickness search unit. Design support device.

3. The design support device according to claim 2, Updating the model by the model update unit; Searching for the film thickness condition by the film thickness search unit; Repeatedly executing, Design support device.

4. The design support device according to any one of claims 1 to 3, The color information is the similarity between the color of the diffuse reflected light and a target color. Design support device.

5. The design support device according to claim 4, The target color is achromatic. Design support device.

6. The design support device according to claim 4, The similarity is a distance in a color space. Design support device.

7. The design support device according to claim 4, The similarity is the absolute value of the difference between spectra. Design support device.

8. The design support device according to any one of claims 1 to 3, wherein the film thickness search unit is configured to search for the film thickness condition by a predetermined combination optimization method. Design support device.

9. The design support device according to claim 8, wherein the combination optimization method is Bayesian optimization. Design support device.

10. The design support device according to any one of claims 1 to 3, wherein the antireflection film has a plurality of film layers with different refractive indices. Design support device.

11. A computer, generating a model having the film thickness information of the antireflection film as an explanatory variable and the color information of the diffused reflected light by the antireflection film as an objective variable; searching for a film thickness condition in which the color information satisfies a target condition based on the model; and a design support method for executing the above.

12. A program for causing a computer to, generate a model having the film thickness information of the antireflection film as an explanatory variable and the color information of the diffused reflected light by the antireflection film as an objective variable; search for a film thickness condition in which the color information satisfies a target condition based on the model; and execute the above.

Images