Footwear component design device and footwear component design method

Abstract

In a footwear component design device 250, a measurement data acquisition unit 212 acquires measurement data of a three-dimensional coordinate group indicating anatomical features corresponding to a three-dimensional homologous model of a foot shape including at least a sole and a foot side surface of a subject. A model determination unit 222 evaluates the similarity between the coordinates between the three-dimensional coordinate group in the measurement data and the three-dimensional coordinate groups in a plurality of three-dimensional homologous models, thereby determining a three-dimensional homologous model 104 similar to the foot shape of the subject from the plurality of three-dimensional homologous models 104. A basic design determination unit 232 determines basic design data corresponding to the determined three-dimensional homologous model 104 from a plurality of basic design data. A design processing unit 234 generates footwear component design data corresponding to the foot shape of the subject on the basis of the determined basic design data.

Description

Footwear component design device and footwear component design method

[0001] The present disclosure relates to a footwear component design device and a footwear component design method, and more particularly to a technology for designing an insole.

[0002] It is desirable to offer a wide range of shoe and insole sizes to accommodate a variety of foot shapes and sizes. However, expanding the range of sizes entails increased manufacturing costs and the risk of having a large inventory, so in reality, most shoe manufacturers focus on offering sizes that are in high demand.

[0003] A technology for selecting a type of shoe that fits a foot shape based on three-dimensional data obtained by measuring the foot shape in three dimensions is known (see, for example, Patent Document 1). In Patent Document 1, two-dimensional shape feature data is generated separately for the toe, midsection, and heel from the three-dimensional foot shape data. Similarly, feature data for each part of the shoe last is generated, and the collected feature data for a large number of shoe lasts is clustered and stored with a code assigned to each cluster. The relationship between the foot shape feature data and the feature data of the shoe last that fits it is learned using a neural network, and rules are established for determining the feature data of the shoe last from the foot shape feature data.

[0004] Meanwhile, in recent years, in response to diversifying tastes, there has been an increase in products that can be customized and personalized by consumers (see, for example, Patent Documents 2 to 4). Also, a method for manufacturing products with a high degree of design freedom using additive manufacturing, also known as 3D printing, is known (see, for example, Patent Documents 5 to 7).

[0005] Japanese Patent Application Laid-Open No. 2000-90272 Japanese Patent Application Laid-Open No. 2022-60213 Japanese Patent Application Laid-Open No. 2023-71642 Japanese Patent Application Laid-Open No. 2023-523141 Japanese Patent Application Laid-Open No. 2018-501865 Japanese Patent Application Laid-Open No. 2017-522917 Japanese Patent Application Laid-Open No. 2020-163154

[0006] Even if additive manufacturing increases product design freedom, there is a possibility that the design may end up with a shape that is difficult to manufacture or that is not normal for the product. To avoid such designs, it is essential to confirm the validity of the design in advance, but there is also a risk that the design freedom may be narrowed in an effort to ensure design validity. There is also a risk that the suitability for customization and personalization may be impaired.

[0007] The present disclosure has been made in light of these circumstances, and its purpose is to provide a technology that achieves both high design freedom and design validity for footwear components.

[0008] In order to solve the above-mentioned problems, a footwear component design device according to one aspect of the present disclosure includes: a model memory unit that pre-stores three-dimensional coordinate groups that indicate anatomical features of the foot as multiple three-dimensional homologous models for multiple foot shapes; a design memory unit that stores multiple basic design data as three-dimensional design data to be used for additive manufacturing of footwear components corresponding to each of the foot shapes of the multiple three-dimensional homologous models; a measurement data acquisition unit that acquires measurement data of the three-dimensional coordinate groups that indicate anatomical features corresponding to the three-dimensional homologous models of a foot shape that includes at least the sole and lateral surfaces of a subject's foot; a model determination unit that determines a three-dimensional homologous model that resembles the subject's foot shape from the multiple three-dimensional homologous models by evaluating the similarity between the three-dimensional coordinate groups in the measurement data and the three-dimensional coordinate groups in the multiple three-dimensional homologous models; a basic design determination unit that determines basic design data corresponding to the determined three-dimensional homologous model from the multiple basic design data; and a design processing unit that generates footwear component design data that corresponds to the subject's foot shape based on the determined basic design data.

[0009] Another aspect of the present disclosure is a footwear component design method, comprising the steps of: reading out a plurality of three-dimensional homologous models corresponding to a plurality of foot shapes, each of which represents a set of three-dimensional coordinates indicating anatomical features of the foot; acquiring measurement data of the set of three-dimensional coordinates indicating anatomical features corresponding to the three-dimensional homologous models of a foot shape including at least the sole and lateral surfaces of a subject's foot; determining a three-dimensional homologous model that resembles the subject's foot shape from the plurality of three-dimensional homologous models by evaluating the similarity between the three-dimensional coordinates in the measurement data and the three-dimensional coordinates in the plurality of three-dimensional homologous models; determining basic design data corresponding to the determined three-dimensional homologous model from a plurality of basic design data that are three-dimensional design data used for additive manufacturing of footwear components corresponding to each of the foot shapes of the plurality of three-dimensional homologous models; and generating footwear component design data corresponding to the subject's foot shape based on the determined basic design data.

[0010] In addition, any combination of the above components, or mutual substitution of the components or expressions of the present disclosure between methods, devices, programs, temporary or non-temporary recording media on which programs are recorded, systems, etc., are also valid aspects of the present disclosure.

[0011] According to the present disclosure, it is possible to achieve both high design freedom and design validity for footwear components.

[0012] 1 is a diagram showing the basic configuration of a footwear component design system. FIG. 2 is a functional block diagram showing each component of a footwear component design device. FIG. 3 is a diagram showing an example of a three-dimensional homologous model. FIG. 4 is an upper perspective view showing the appearance of an insole manufactured by three-dimensional additive manufacturing. FIG. 5 is a lower perspective view showing the appearance of an insole manufactured by three-dimensional additive manufacturing. FIG. 6 is a diagram showing an insole without three-dimensional decoration added. FIG. 7 is a diagram showing an insole with three-dimensional decoration of text added to the side. FIG. 8 is a diagram showing an insole with three-dimensional decoration of marks added to the side and bottom. FIG. 9 is a diagram showing a schematic diagram of a sliced ​​cross-sectional image of insole design data. FIG. 10 is a functional block diagram showing each component of a shape analysis device. FIG. 11 is a diagram showing the relationship between the distribution of foot length and foot circumference by gender and Japanese JIS standards. FIG. 12 is a diagram showing the difference between a general cluster analysis method and a new cluster analysis method. FIG. 13 is a diagram showing a method combining different cluster analyses at an upper hierarchical level and a lower hierarchical level. FIG. 14 is a diagram showing the Euclidean distance and cosine similarity between points. FIG. 15 is a diagram showing the relationship between corresponding points in two foot print models. FIG. 16 is a diagram showing a comparison of Euclidean distance and cosine similarity when comparing foot print models. FIG. 17 is a diagram showing the difference in classification results of cluster analysis based on cosine similarity or Euclidean distance. Fig. 1 shows the classification results of cluster analysis based on cosine similarity, Fig. 2 shows the regression lines of foot length and foot circumference for the top three clusters, and the results of principal component analysis for the top three clusters, Fig. 3 shows the results of principal component analysis for the clusters,

[0013] The present disclosure will be described below based on preferred embodiments with reference to the accompanying drawings. In the embodiments and modifications, the same or equivalent components are designated by the same reference numerals, and redundant descriptions will be omitted as appropriate.

[0014] In this embodiment, footwear or footwear components customized for each wearer or prospective purchaser are provided. For example, not only are footwear or footwear components designed to best fit the wearer's foot size and foot shape, but the prospective purchaser's preferences are also reflected in the design and decoration. This allows for the design of one-of-a-kind footwear or footwear components that reflect the individual differences and preferences of each wearer or prospective purchaser.

[0015] Here, "footwear" includes shoes such as sports shoes that cover the entire foot, shoes such as sandals that do not cover the entire foot, socks, socks and tights that are integrated with socks. Furthermore, "footwear" includes not only cases where the entire footwear can be customized, but also cases where footwear components that are part of the footwear can be customized. The footwear component in this embodiment is an insole (commonly referred to as an "insole") that is inserted into the footwear and placed on the midsole. In the following, this embodiment will describe an example of designing an insole as a footwear component. However, as a modified example, this may be applied to designing footwear components other than insoles, such as soles and uppers, or may be applied to designing the entire footwear, such as sandals.

[0016] In this embodiment, unlike conventional manufacturing methods that mass-produce and sell footwear or footwear components of the same specifications, a three-dimensional additive manufacturing method (also called "additive manufacturing" or "3D printing") is adopted. The insole, which is a footwear component in this embodiment, is formed with a three-dimensional lattice structure.

[0017] However, when designing insoles customized for each customer, it takes a lot of time to design them from scratch for each order. Furthermore, when creating designs for each order, it is difficult to confirm in advance whether the manufacturing process will be appropriate for actual production, and whether the design is manufacturable. Thus, when customizing footwear, it is necessary to ensure both a high degree of design freedom and design validity.

[0018] Therefore, in this embodiment, a large amount of basic design data is prepared, which is a basic design that is premised on the addition of customization. Then, the design validity of this basic design data is confirmed in advance within the customizable range, thereby achieving both design freedom and design validity.

[0019] The large amount of basic design data is designed as follows: The three-dimensional shapes of the feet of a large number of test subjects are measured and stored as foot shape data, which is then pre-classified into multiple clusters using a cluster analysis technique described below. Furthermore, shoe lasts and insoles are designed based on the classified multiple clusters, providing footwear product shape variations based on standards that differ from conventional standards, thereby providing footwear or footwear components that fit the feet of many people. Demand for each shape variation is also forecasted according to the distribution of foot shapes.

[0020] The three-dimensional shape of the foot of a prospective insole purchaser is measured to obtain foot shape measurement data. The measurement data of the subject's foot shape is three-dimensional point cloud data indicating anatomical features corresponding to a three-dimensional homologous model of the subject's foot shape, which includes at least the sole and lateral surfaces of the subject's foot. The three-dimensional homologous model is a model that defines a coordinate group of the anatomical features of an average foot in three-dimensional space, and defines a predetermined number of contour points, for example, 295 points.

[0021] The degree of similarity is evaluated between the measurement data of the subject's foot shape and a large number of foot shape models classified by cluster analysis, and the design data corresponding to the three-dimensional homologous model that resembles the subject's foot shape is used as basic design data. Footwear component design data is generated by applying individual modifications specified by the subject to this basic design data. Individual modifications include modifications based on the prospective purchaser's preferences regarding arch height, and modifications that add freely entered text, patterns, and other decorative designs. A three-dimensional lattice-structure insole is manufactured using a three-dimensional additive manufacturing method based on the generated footwear component design data.

[0022] Below, we will first explain the footwear component design system 200 operated by a footwear manufacturer, and then explain the shape analysis system that classifies a large number of foot shape models or basic design data by cluster analysis.

[0023] 1 shows the basic configuration of a footwear component design system 200. The footwear component design system 200 includes measuring devices 15, such as a three-dimensional foot shape measuring device 18 and an information terminal 16, and a footwear component design device 250. When the three-dimensional foot shape measuring device 18 is used as the measuring device 15, the foot shape of the subject 10 is scanned by the three-dimensional foot shape measuring device 18. When the information terminal 16 is used as the measuring device 15, an image of the foot of the subject 10 is taken by the camera function of the information terminal 16.

[0024] Foot shape measurement data can be acquired either from a person considering purchasing an insole as the measurement subject 10 or from a large number of subjects for the purpose of acquiring foot shape data as the measurement subjects 10. Persons considering purchasing an insole include, for example, people considering purchasing an insole in-store at a footwear store, people hoping to purchase an insole through online sales, etc. A large number of subjects are subjects for the purpose of acquiring foot shape measurement data of various sizes or shapes, regardless of gender or age, at the stage of generating basic design data for footwear or footwear components.

[0025] When a three-dimensional foot shape measuring device 18 is used as the measuring device 15, the three-dimensional foot shape measuring device 18 acquires three-dimensional point cloud data related to the foot shape of the subject 10 by laser measurement. The "foot shape" here refers to a three-dimensional model that virtually reproduces the three-dimensional shape of the subject 10's foot. The three-dimensional foot shape measuring device 18 transmits the three-dimensional point cloud data of the foot shape to the footwear component design device 250. Note that the "footwear component design device" in the claims may refer to the entire footwear component design system 200 or may refer to the footwear component design device 250. Furthermore, in this embodiment, since the footwear component design device 250 is equipped with many of the characteristic functions included in the "footwear component design device" in the claims, the footwear component design device 250 essentially corresponds to the "footwear component design device."

[0026] When an information terminal 16 is used as the measuring device 15, the subject 10 captures images of their feet using a camera built into the information terminal 16, such as their mobile phone. For example, when capturing an image of the entire foot from directly above, even if the heel is captured sufficiently, the ankle is captured, making it difficult to capture the entire foot contour in a single image. Therefore, multiple images are captured from multiple angles to capture at least the contours of the toes and heel, and the contours of both the left and right sides of the forefoot to midfoot. The subject 10 may place their foot on a dedicated measurement mat 12 and capture an image of their foot using the camera function and foot shape capture application of the information terminal 16. The information terminal 16 transmits the multiple images of the foot to the footwear component design device 250.

[0027] If the information terminal 16 has a three-dimensional scanner function using technology such as LiDAR (Light Detection and Ranging), it may scan the area around the foot to acquire three-dimensional point cloud data of the foot shape. Even if the information terminal 16 does not have a three-dimensional scanner function using LiDAR or the like, it may acquire three-dimensional point cloud data of the foot shape by image synthesis processing such as photogrammetry. In these cases, the information terminal 16 transmits the three-dimensional point cloud data of the foot shape to the footwear component design device 250.

[0028] The footwear component design device 250 is a server computer connected to a plurality of three-dimensional foot shape measuring devices 18 and information terminals 16 via a network line such as the Internet or a LAN (Local Area Network) or a communication means such as wireless communication. The footwear component design device 250 may be configured as a single server computer or may be configured as a combination of a plurality of server computers.

[0029] FIG. 2 is a functional block diagram showing each component of the footwear component design device 250. The footwear component design device 250 can be realized using various hardware and software configurations. FIG. 2 depicts functional blocks of the footwear component design device 250 that are realized by combining various hardware and software configurations. Therefore, those skilled in the art will understand that these functional blocks can be realized in various forms using hardware alone, software alone, or a combination thereof. The footwear component design device 250 is configured using a combination of hardware, such as a microprocessor such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit), memory, a display, and a communication module. The footwear component design device 250 may be a server computer built and managed by a footwear manufacturer or a footwear retailer.

[0030] A program having the following functions runs on the footwear component design device 250. The functions of the footwear component design device 250 include a communication unit 210, a measurement data acquisition unit 212, a model conversion unit 214, an individual information acquisition unit 216, a footwear information acquisition unit 218, a model storage unit 220, a model determination unit 222, a design storage unit 230, a basic design determination unit 232, a design processing unit 234, and an output unit 236. The communication unit 210 of the footwear component design device 250 is connected to the three-dimensional foot shape measuring device 18 and the information terminal 16 via a network.

[0031] The measurement data acquisition unit 212 acquires three-dimensional point cloud data of the foot shape of the subject 10 from the three-dimensional foot shape measuring device 18 via the communication unit 210, or acquires photographed images of the foot of the subject 10 from the information terminal 16.

[0032] When acquiring three-dimensional point cloud data from the three-dimensional foot shape measuring device 18, the measurement data acquisition unit 212 acquires measurement data of a three-dimensional coordinate group by converting the three-dimensional point cloud data into three-dimensional point cloud data that indicates anatomical features corresponding to a three-dimensional homologous model. The conversion of the three-dimensional point cloud data is performed by the model conversion unit 214.

[0033] When acquiring a plurality of foot images of the subject 10 from the information terminal 16, the measurement data acquisition unit 212 acquires measurement data of a group of three-dimensional coordinates by generating three-dimensional point cloud data indicating anatomical features corresponding to a three-dimensional homologous model based on the plurality of foot images. The generation of the three-dimensional point cloud data based on the plurality of foot images and the conversion into three-dimensional point cloud data indicating anatomical features corresponding to a three-dimensional homologous model are performed by the model conversion unit 214.

[0034] In this way, even if the measurement data acquisition unit 212 acquires measurement data using various types of foot shape measuring devices as the measuring device 15, the model conversion unit 214 converts the measurement data into three-dimensional point cloud data corresponding to a three-dimensional homologous model. This allows the measurement data to be generalized into a data format that is easy to use. Meanwhile, in a modified example, the measurement data acquired by the measurement data acquisition unit 212 may be converted in advance by the three-dimensional foot shape measuring device 18 or the information terminal 16 into three-dimensional point cloud data corresponding to a three-dimensional homologous model of the foot shape.

[0035] The model storage unit 220 pre-stores three-dimensional coordinate groups indicating anatomical features of the foot as multiple three-dimensional models for multiple foot shapes. The three-dimensional foot shape models stored in the model storage unit 220 are, for example, three-dimensional homologous models in which coordinate groups of points indicating the anatomical features of an average foot in three-dimensional space are defined. The three-dimensional homologous foot shape models are models in which multiple foot shapes of various sizes and shapes are represented as polyhedrons in which they are anatomically associated using the same topological geometric structure with the same number of points. The points indicating the anatomical features of an average foot are obtained in advance based on foot samples obtained from a large number of subjects. In modified examples, the three-dimensional model may be various other models created in accordance with the same anatomical standards, such as a three-dimensional shape model in a spherical coordinate system or a model in which a template foot shape generated by three-dimensional CAD is deformed based on anatomical landmarks.

[0036] FIG. 3 shows an example of a three-dimensional homologous model. The three-dimensional homologous model 104 of the foot shape is composed of a point cloud representing the anatomical features of the foot. The three-dimensional homologous model 104 does not necessarily faithfully reproduce the foot shape of the subject 10, particularly the shape around the toes, but is constructed with an intentionally deformed shape to facilitate comparison with other foot shapes or three-dimensional homologous models. In other words, the three-dimensional homologous model 104 does not completely resemble the actual foot shape or foot contour of the subject 10, and has a shape that resembles a convex hull processing of the toe contours in particular. Furthermore, by detecting multiple characteristic points included in the foot contour of the three-dimensional homologous model 104 and using each point as a reference, the lengths of each part of the foot, such as foot length, foot circumference, arch circumference, heel circumference, and heel width, can be measured. The method for measuring the lengths of each part is known, so a description thereof is omitted.

[0037] Returning to FIG. 2 , the model storage unit 220 stores the multiple three-dimensional homology models 104 in a form in which the upper layer is divided into multiple clusters by cluster analysis based on cosine similarity. The cluster analysis based on cosine similarity is a cluster analysis based on the cosine similarity between position vectors from a predetermined reference point to each point on the three-dimensional homology model 104. The model storage unit 220 also stores the multiple three-dimensional homology models 104 in a form in which the lower layer is further divided into multiple clusters by a cluster analysis different from that for the upper layer. The cluster analysis for the lower layer is a cluster analysis based on principal component analysis in which the dimensions are reduced to the second principal component or less, excluding the first principal component, when principal component analysis is performed for each cluster in the upper layer.

[0038] The model determination unit 222 evaluates the similarity between three-dimensional coordinates in the measurement data acquired by the measurement data acquisition unit 212 and three-dimensional coordinates in the multiple three-dimensional homologous models 104. The model determination unit 222 determines a three-dimensional homologous model 104 that resembles the footprint of the subject 10 from the multiple three-dimensional homologous models 104 in the model storage unit 220 based on the similarity between the coordinates. More specifically, the model determination unit 222 first obtains position vectors from a predetermined reference point to each point in the multiple three-dimensional homologous models 104 stored in the model storage unit 220. Next, the model determination unit 222 obtains position vectors from a predetermined reference point to each point in the measurement data. Then, the model determination unit 222 determines a three-dimensional homologous model 104 that resembles the footprint of the subject 10 from the multiple three-dimensional homologous models 104 in the model storage unit 220 based on the distance between corresponding points of the three-dimensional homologous models or the similarity of the position vectors.

[0039] The distance between points is, for example, Euclidean distance, and the similarity between position vectors is, for example, cosine similarity. The model determination unit 222 adds up the selection results of three-dimensional homologous models that are similar based on Euclidean distance and the selection results of three-dimensional homologous models that are similar based on cosine similarity, and determines from among them the three-dimensional homologous model that is most similar to the footprint of the subject 10. Assume that approximately 6,000 three-dimensional footprint models are stored in the model storage unit 220. In this case, even if the model determination unit 222 calculates the similarity between all three-dimensional homologous models and determines the most similar three-dimensional homologous model, the calculation can be completed in about 10 seconds, making it possible to extract the footprint in a very short time. Euclidean distance and cosine similarity will be described later.

[0040] The model determination unit 222 scales the determined three-dimensional homologous model to various sizes according to the foot length of the subject 10, and calculates the similarity between each of the scaled three-dimensional homologous models and the three-dimensional homologous model of the foot shape of the subject 10. Based on the respective similarities, the model determination unit 222 determines the scale of the three-dimensional homologous model with the highest similarity.

[0041] The footwear information acquisition unit 218 acquires footwear information indicating the type of shoes specified by the subject. More specifically, the person wishing to purchase an insole inputs information such as the product model and size of the shoes into which the insole will be inserted as footwear information, and the footwear information acquisition unit 218 acquires the footwear information. The information acquired by the footwear information acquisition unit 218 is sent to the basic design determination unit 232. The information such as the product model and size of the shoes is referenced in the selection of basic design data, which will be described later.

[0042] In a modified example, if a person who wishes to purchase an insole does not input information such as the product model or size of the shoe, basic design data of a standard design may be selected regardless of the product model or size of the shoe. Alternatively, on the assumption that a person who wishes to purchase an insole does not input information such as the product model or size of the shoe, basic design data of a standard design may be selected regardless of the product model or size of the shoe.

[0043] The individual information acquisition unit 216 acquires individual information indicating individual modifications specified by the prospective purchaser as modifications to the basic design of the insole. The "individual modifications" here refer to changes in function or design according to the prospective purchaser's preferences for customization. For example, individual modifications include modifications related to preferences such as the height and hardness of the arch support, and modifications related to decorative preferences.

[0044] For example, suppose the arch height ratio in the basic design data is a predetermined value of 60 to 70%. The prospective purchaser can select "high," "standard," or "low" as the arch height ratio according to their own preference. If the individual information acquisition unit 216 acquires individual information indicating a "high" arch height ratio, the basic design data is modified so that the arch height ratio becomes a predetermined value of 70 to 80%. If the individual information acquisition unit 216 acquires individual information indicating a "low" arch height ratio, the basic design data is modified so that the arch height ratio becomes a predetermined value of 50 to 60%. Note that the numerical range of the arch height ratio is merely an example, and other numerical ranges may be used.

[0045] A prospective purchaser can select the hardness of the insole 300 from "hard," "standard," or "soft" according to their own preference. If the individual information acquisition unit 216 acquires individual information indicating "hard" or "soft" for the hardness, the partial density of the three-dimensional lattice structure may be changed, or the material of the insole 300 may be changed. In addition to the hardness, the material of the insole 300 may be selectable according to preference, and the surface treatment method, such as embossing, gloss, or matte, may be selectable according to preference. Furthermore, the type and form of the reinforcing member attached to the insole 300 may be selectable according to preference.

[0046] The prospective buyer can specify letters, symbols, marks, and patterns as variations based on their decorative preferences. A three-dimensional decoration (hereinafter referred to as "three-dimensional decoration") representing at least one of the specified letters, symbols, marks, and patterns can be added to at least one of the side, bottom, and top surfaces of the insole 300. The prospective buyer can freely enter text up to a specified character limit. The entered text is converted into design data for the three-dimensional decoration and then integrated into the basic design data. The individual information acquisition unit 216 stores a list of prohibited terms that are deemed inappropriate from a legal or public order and morals perspective. If the text entered by the prospective buyer contains prohibited terms, conversion to three-dimensional decoration is restricted. Discriminatory terms, prohibited terms, inappropriate terms, and terms reminiscent of these terms are constantly being generated. To automatically eliminate these terms, a glossary stored in advance as a list may be referenced, or a continuously updated learning model, generative artificial intelligence (AI), application programming interface (API), or other similar techniques may be used. In addition to text, potential buyers can select other elements, such as single alphabet letters, symbols, marks, and patterns, to add as three-dimensional decoration. As a variant, potential buyers may be able to select the surface color and type of surface pattern of the insole 300 as decoration other than the three-dimensional decoration. Potential buyers may also input handwritten drawings or simple drawings created by themselves as original drawings for the three-dimensional decoration and add them as three-dimensional decoration. Randomly generated patterns or character strings may be set and added as three-dimensional decoration. Information indicating the date and location of the insole 300 design, as well as two-dimensional codes or barcodes indicating this information, may be set and added as three-dimensional decoration. Potential buyers may specify the content of such decorations by having others select them through a predetermined online service. The serial number or individual identification number of the insole 300 may be added as three-dimensional decoration. Furthermore, factors affecting manufacturability may be taken into account during additive manufacturing to determine whether or not the three-dimensional decoration can be implemented. Furthermore, the shape of the three-dimensional decoration may be automatically modified to make it manufacturable.

[0047] The three-dimensional decoration of the text may be specified as desired, such as font type, size, character thickness, underline, character box, three-dimensional shape height, etc. Note that while the three-dimensional decoration in this embodiment is configured as a convex three-dimensional shape, in modified examples it may be configured as a concave three-dimensional shape, or the prospective purchaser may be able to select whether to configure it as a convex or concave shape.

[0048] The prospective purchaser can specify the placement location of the three-dimensional decoration. The placement location of the three-dimensional decoration is at least one of the side, top, and bottom surfaces around the insole 300. Information including the text of the three-dimensional decoration, i.e., the specification of the text, mark, pattern, etc., and individual information indicating the placement location, is sent to the design processing unit 234. The three-dimensional decoration and its placement location will be described later.

[0049] The basic design determination unit 232 determines, from the plurality of basic design data, basic design data corresponding to the three-dimensional homologous model 104 determined by the model determination unit 222. When footwear information is received from the footwear information acquisition unit 218, the basic design determination unit 232 determines basic design data using the model determination unit 222 based on the three-dimensional homologous model 104 and the footwear information.

[0050] The design storage unit 230 stores a plurality of basic design data as three-dimensional design data used for additive manufacturing of insoles corresponding to the respective foot shapes of the plurality of three-dimensional homologous models 104 stored in the model storage unit 220. The plurality of three-dimensional homologous models 104 are classified into a plurality of clusters, and the basic design data corresponding to these three-dimensional homologous models 104 are also substantially classified into a plurality of clusters. The plurality of basic design data stored in the design storage unit 230 may correspond one-to-one to the plurality of three-dimensional homologous models 104 classified into the plurality of clusters in the model storage unit 220, or may correspond in any of a many-to-one, one-to-many, or many-to-many manner.

[0051] In the case of a many-to-one correspondence, variations of the basic design data are prepared so that one piece of basic design data corresponds to a plurality of three-dimensional homologous models 104. For example, depending on the insole product model, it may be desirable to reduce the number of insole variations relative to the number of foot shape variations, and in such cases, a many-to-one correspondence may be used. In the case of a one-to-many correspondence, variations of the basic design data are prepared so that a plurality of pieces of basic design data corresponds to a single three-dimensional homologous model 104. For example, a plurality of types of insole product models may be prepared for a single foot shape. Alternatively, in the case where the bottom shape of the insole differs depending on the shoe last shape of the shoe product model into which the insole will be inserted, a plurality of pieces of basic design data corresponding to shoe product models may be prepared for a single foot shape. Furthermore, one piece of basic design data may correspond to a plurality of three-dimensional homologous models 104, and at the same time, such basic design data may be prepared for each insole product model or each shoe product model.

[0052] The plurality of basic design data stored in the design storage unit 230 are all designs that have been confirmed in advance to be valid, i.e., capable of being manufactured by additive manufacturing. Furthermore, it has been confirmed in advance that the design remains valid even if individual modifications are made to the basic design data within the range that can be specified as individual information.

[0053] The design processing unit 234 generates insole design data as footwear part design data corresponding to the foot shape of the subject 10 based on the basic design data determined by the basic design determination unit 232. When receiving individual information from the individual information acquisition unit 216, the design processing unit 234 generates insole design data as footwear part design data obtained by individually modifying the basic design based on the basic design data and the individual information. For example, when receiving specification information regarding arch height as the individual information, the design processing unit 234 generates insole design data by modifying the top surface shape of the basic design data including the arch shape to achieve the specified arch height. For example, when receiving specification information regarding hardness as the individual information, the design processing unit 234 generates insole design data by modifying a portion of the top surface shape of the basic design data to achieve the specified hardness. For example, when receiving information including text as the individual information, the design processing unit 234 generates insole design data by adding three-dimensional decoration modeled after the specified text to the basic design based on the basic design data and the individual information.

[0054] In the present embodiment, an example has been described in which the basic design data is individually modified according to the arch height ratio specified by the individual information. In a modified example, a plurality of basic design data each having a different arch height ratio may be prepared in advance as design variations based on one basic design data, and the design validity of each may be confirmed in advance.

[0055] As a result, it is possible to ensure both the degree of freedom in the design of the insole and the validity of the insole design. Furthermore, since insole design data based on designs other than the basic design data stored in the design storage unit 230 and specifications of three-dimensional decorations within a range that has not been confirmed in advance are excluded in advance, it is possible to prevent designs that have not been confirmed as being valid. In particular, it is possible to prevent manufacturing failures and the occurrence of defective products due to the creation of unexpected insole designs based on extremely non-standard foot shapes.

[0056] The output unit 236 transmits the insole design data generated by the design processing unit 234 to the additive manufacturing device 260, thereby executing additive manufacturing of the insole 300 based on the insole design data. The output unit 236 may also display the insole design data on a display screen of the footwear component design device 250, or may transmit the insole design data to the information terminal 16 via the communication unit 210.

[0057] FIG. 4 is an upper perspective view showing the appearance of an insole manufactured by three-dimensional additive manufacturing. FIG. 5 is a lower perspective view showing the appearance of an insole manufactured by three-dimensional additive manufacturing. The left side of FIG. 4 is the toe side 308, and the right side is the heel side 309. The right side of FIG. 5 is the toe side 308, and the left side is the heel side 309. The insole 300 is manufactured by a three-dimensional additive manufacturing method using a stereolithography system. The material of the insole 300 is a material that can be molded by a three-dimensional additive manufacturing method using a stereolithography system, and is preferably, for example, a resin material or a rubber material that is suitable for maintaining flexibility, elasticity, and durability. When the insole 300 is made of resin, for example, polyolefin resin, ethylene-vinyl acetate copolymer (EVA), polyamide-based thermoplastic elastomer (TPA, TPAE), thermoplastic polyurethane (TPU), or polyester-based thermoplastic elastomer (TPEE) can be used. When the insole 300 is made of rubber, it may be made of butadiene rubber (BR), for example. As a modified example, the insole 300 may be constructed by manufacturing mainly the upper layer of the insole 300 from resin or the like using an additive manufacturing method, mass-producing the lower layer from another lightweight material using a manufacturing method such as injection molding using a mold, and bonding the upper layer and lower layer together.

[0058] The base layer appearing on the underside 302 and side 303 of the insole 300 is constructed as a three-dimensional mesh structure in which unit structures, each of which is a three-dimensional lattice structure, are repeatedly arranged. These unit structures are repeatedly arranged in three orthogonal axial directions: front-to-back, left-to-right, and up-to-down. The side 303 is almost absent on the toe side 308, and is formed from the edge of the underside 302 extending from the midfoot to the heel side 309 in the form of a slightly upwardly slanted, cone-shaped peripheral wall. The side 303 has a larger surface area on the medial side 306 than on the lateral side 307 due to the height required for arch support. While the heel side 309 of the side 303 also has some surface area, it curves significantly to follow the contours of the heel, making it more difficult to add three-dimensional text decoration compared to the medial side 306 or lateral side 307.

[0059] A surface layer portion appearing on the upper surface 301 of the insole 300 is provided with a large number of holes. The curved shape of the upper surface 301 of the insole 300 is designed to correspond to the shape of the foot so as to fit the shape of the sole of the subject 10. The curved shape of the lower surface of the insole 300 is designed to correspond to the shoe designated by the subject 10 so as to fit the shape of the midsole and the shape of the interior upper of the shoe into which the insole 300 will be inserted. For example, a design may be determined to correspond to the shoe last designated by the subject 10 in the footwear information, a design may be determined to correspond to the designated shoe type, or a design may be determined to correspond to the foot length size and foot width type of the designated shoe. In a modified example, the lower surface shape of the insole 300 may be designed to have a standard shape regardless of whether the shoe is designated in the footwear information. Furthermore, a material suitable for the type of shoe designated in the footwear information may be selected as the material of the insole 300.

[0060] Fig. 6 is a side view showing the insole 300 without any three-dimensional decoration. Fig. 7 is a side view showing the insole 300 with three-dimensional text decoration added to the side. The text decoration 310 is mainly a decoration in the form of three-dimensional text or a character string. Fig. 8 is a bottom perspective view showing the insole 300 with three-dimensional mark decoration added to the side and bottom. The mark decorations 311, 312 may be decorations in the form of three-dimensional marks such as the logo or symbol mark of a footwear manufacturer.

[0061] The three-dimensional decoration is placed on at least one of the side surface 303, upper surface 301, and lower surface 302 around the insole 300. However, because the insole 300 is shaped like a sheet, the locations where the three-dimensional decoration can be placed are limited. For example, the edge of the side surface 303 around the insole 300 is slightly raised in a cone shape from the midfoot to the heel side 309, providing a relatively high and thick surface, thereby ensuring a sufficient installation area on a portion of the side surface 303. More specifically, the medial side 306, lateral side 307, and heel side 309 on the side surface 303, which can ensure sufficient installation area, can be selected. On the lower surface 302 of the insole 300, the heel or central portion can be selected. The upper surface 301 of the insole 300 can also be selected, but because it is in direct contact with the sole of the wearer's foot, it may be possible to avoid areas that affect kicking, or to limit the thickness of the three-dimensional decoration. Alternatively, if the product specifications require fabric to be attached to the upper surface 301 of the insole 300, the decoration may be arranged by printing text, marks, or other decorations on the fabric.

[0062] FIG. 9 is a schematic diagram of a sliced ​​cross-sectional image in the insole design data. The sliced ​​cross-sectional image 320 in the insole design data is data of each cross-section obtained by horizontally slicing the three-dimensional lattice structure across multiple layers when the insole 300 is placed horizontally. For example, when DLP (Digital Light Processing) is used as the stereolithography method, a cross-sectional pattern such as that shown in FIG. 9 is exposed and layered for each layer. Each cross-section is formed as a monochrome binary image as shown in FIG. 9. UV light is irradiated onto the irradiated areas 321, represented by white dots, to deposit resin, while UV light is not irradiated onto the non-irradiated areas 322, represented by black dots. In this way, the insole 300 is layer-by-layer manufactured by repeatedly irradiating the irradiated areas 321 based on the insole design data for each layer.

[0063] Since the slice cross-sectional image 320 is thus formed as a monochrome binary image, a monochrome binary image of the cross section of only the three-dimensional decorative portion may be generated, and then superimposed on a monochrome binary image of the basic design portion to generate the slice cross-sectional image 320 of the entire layer. Alternatively, partial numerical changes may be made to the basic design data as a three-dimensional model to match the shape of the three-dimensional decorative portion to generate insole design data as an overall three-dimensional model including the three-dimensional decorative portion, and then the slice cross-sectional image 320 of each layer may be generated.

[0064] The following describes a shape analysis device 50 that classifies a plurality of three-dimensional homologous models 104 stored in the model storage unit 220 by cluster analysis.

[0065] FIG. 10 is a functional block diagram showing each component of the shape analysis device 50. The shape analysis device 50 can be realized using various hardware and software configurations. FIG. 10 depicts functional blocks of the shape analysis device 50 that are realized by combining various hardware and software configurations. Therefore, those skilled in the art will understand that these functional blocks can be realized in various forms using only hardware, only software, or a combination of both. The shape analysis device 50 is composed of a combination of hardware, such as a microprocessor, memory, display, and communication module. The shape analysis device 50 may be a server computer built and managed by a shoe manufacturer. However, the shape analysis device 50 runs a program with the following functions: The shape analysis device 50 includes, as its functions, a communication unit 19, a measurement data acquisition unit 20, a model generation unit 21, a model storage unit 22, an output unit 26, an analysis processing unit 30, and a statistical processing unit 35. The communication unit 19 of the shape analysis device 50 and the 3D foot shape measurement device 18 are connected via a network. The foot shape of the subject 10 is scanned by the three-dimensional foot shape measuring device 18. Foot shape data acquisition may be performed, for example, at the shoe design stage, with a large number of subjects 10 being selected for the purpose of acquiring foot shape data of various sizes or shapes, regardless of gender or age.

[0066] The model storage unit 22 pre-stores three-dimensional coordinate groups representing anatomical features of at least some of the body as multiple three-dimensional models for multiple body shapes. The body shapes referred to here primarily refer to foot shapes. The three-dimensional foot shape models stored in the model storage unit 22 are, for example, three-dimensional homologous models in which coordinate groups of points representing the anatomical features of an average foot in three-dimensional space are defined. The three-dimensional homologous foot shape models are models in which multiple foot shapes of various sizes and shapes are represented as polyhedra in which they are anatomically associated using the same topological geometric structure with the same number of points. The points representing the anatomical features of an average foot are obtained in advance based on foot samples obtained from a large number of subjects. In modified examples, the three-dimensional model may be various other models created in accordance with the same anatomical standards, such as a three-dimensional shape model in a spherical coordinate system or a model in which a template foot shape generated by three-dimensional CAD is deformed based on anatomical landmarks.

[0067] The measurement data acquisition unit 20 acquires, as measurement data, a group of three-dimensional coordinates indicating anatomical features corresponding to a three-dimensional homologous model of the foot shape of the subject 10 from the three-dimensional foot shape measuring device 18 via the communication unit 19. The measurement data acquisition unit 20 may also acquire measurement data from an external terminal 17 other than the three-dimensional foot shape measuring device 18 via the communication unit 19. The model generation unit 21 generates a three-dimensional homologous model of the foot shape based on the measurement data acquired by the measurement data acquisition unit 20 and stores the new three-dimensional homologous model in the model storage unit 22. In this case, even if measurement data is acquired using various types of foot shape measuring devices as the three-dimensional foot shape measuring device 18 or the external terminal 17, the model generation unit 21 converts the data into a three-dimensional homologous model. This allows the measurement data to be generalized into a data format that is easy to use. Meanwhile, in a modified example, the measurement data acquired by the measurement data acquisition unit 20 may be converted into a three-dimensional homologous model of the foot shape in advance by the three-dimensional foot shape measuring device 18 or the external terminal 17. In this case, the model generation unit 21 does not need to generate a three-dimensional homologous model of the foot shape based on the measurement data, and the measurement data may be stored directly in the model storage unit 22 as a new three-dimensional homologous model.

[0068] The analysis processing unit 30 has a cluster analysis processing unit 31 and a principal component analysis processing unit 32. The cluster analysis processing unit 31 performs cluster analysis of upper and lower hierarchies, which will be described later, on the multiple three-dimensional homologous models stored in the model storage unit 22. The cluster analysis processing unit 31 implements a method that combines different cluster analyses for the upper and lower hierarchies. The principal component analysis processing unit 32 performs principal component analysis, which will be described later, on the multiple three-dimensional homologous models of footprints stored in the model storage unit 22. After performing cluster analysis of the upper hierarchies, the cluster analysis processing unit 31 performs cluster analysis of the lower hierarchies based on the results of the principal component analysis performed by the principal component analysis processing unit 32.

[0069] Information on each cluster as a classification result obtained by the cluster analysis performed by the cluster analysis processing unit 31 and the principal component analysis performed by the principal component analysis processing unit 32 is output to the outside by the output unit 26. Based on the classification result from the cluster analysis, the analysis processing unit 30 divides the three-dimensional homology models stored in the model storage unit 22 into a plurality of clusters and stores them.

[0070] The output unit 26 may display the classification results on a display screen of the shape analysis device 50 or the external terminal 17, or may transmit the classification results to the external terminal 17 via the communication unit 19. The output unit 26 outputs a group of three-dimensional coordinates of the three-dimensional homologous model for each cluster. The information on the classification results output by the output unit 26 may be used in designing a shoe last or an insole.

[0071] The statistical processing unit 35 executes a demand prediction for shoe lasts corresponding to each foot print model based on the distribution of foot print models for each cluster. The results of the demand prediction are output to the external terminal 17 by the output unit 26. The statistical processing unit 35 manages a database of shoe lasts, footwear, or insoles corresponding to the foot print models for each cluster stored in the model storage unit 22 according to the results of the demand prediction.

[0072] Figure 11 shows the relationship between the distribution of foot length and foot circumference by gender and the Japanese JIS standard. Shoe sizes in Japan are classified and standardized by the Japanese Industrial Standards (JIS), which is a standard standard, based on foot length and foot circumference / foot width. This standard takes into account differences in physique and specifies different sizes for men, women, and children.

[0073] Figure 11(a) shows the JIS standard for women and the distribution of measurements taken from a sample of many adult women. Figure 11(b) shows the JIS standard for men and the distribution of measurements taken from a sample of many adult men. The JIS standard for women specifies a foot length range of 19.5 cm to 27.0 cm, while the JIS standard for men specifies a foot length range of 20.0 cm to 30.0 cm. Foot circumference / foot width is divided into shoe widths A to F for women and shoe widths A to G for men, depending on foot length. However, different ranges are specified, such as shoe width A for women specifying a foot circumference of 183 mm and a foot width of 76 mm, and shoe width A for men with a foot length of 20.0 cm specifying a foot circumference of 189 mm and a foot width of 79 mm.

[0074] The diagonal lines in Figures 11(a) and 11(b) indicate the relationship between foot length (mm) and foot circumference (mm) for each foot length size according to the JIS standard. The diagonal lines in Figure 11(a) indicate women's shoe widths A to F, and in Figure 11(b) indicate men's shoe widths A to G. The difference in foot circumference between shoe widths is approximately 6 mm.

[0075] The distribution of female samples shown in Figure 11(a) is generally within the range of foot length and foot circumference specified by the JIS standard, but there are also samples that fall outside that range. The distribution of male samples shown in Figure 11(b) is generally within the range of foot length and foot circumference specified by the JIS standard, but there are also samples that fall outside that range, just like with women.

[0076] The bold diagonal lines in Figures 11(a) and 11(b) indicate the sample averages for men and women. For each foot length size, there is a ±3-4 width variance in foot circumference above and below the average line. In other words, even if the relationship between foot length and foot circumference shown in the figure is established and provided as a guideline, the actual variance or variation in foot circumference is large. Furthermore, while conventional shoe standards only specify the relationship between foot length and foot circumference, foot size and shape are also determined by other parameters. Therefore, even if only foot length and foot circumference are matched, the intended shoe does not necessarily fit the intended foot. To address this situation, we used the method described below to classify three-dimensionally measured foot shape models using cluster analysis and design size variations based on this classification.

[0077] FIG. 12 shows the difference between a general cluster analysis method and the new cluster analysis method. When clustering the foot shape data group X using a general cluster analysis method, cluster analysis based on Euclidean distance is considered. However, with this method, the clusters are classified in a manner that is significantly influenced by foot length. Furthermore, if classification is desired based on differences in foot shape as well as foot length, it is possible to divide the data by foot length and perform cluster analysis for each foot length. However, this method is also significantly influenced by foot length, and it is expected that it will be difficult to distinguish from conventional grading methods that classify based only on multiple foot circumferences for each foot length. Furthermore, there is a risk that the number of clusters classified will differ for each size.

[0078] Therefore, the cluster analysis processing unit 31 in this embodiment performs cluster analysis across all foot length sizes, as shown in the figure. The foot shape data group X is configured by combining large clusters W, which are divided by similar shapes (i.e., similar shapes) across foot length sizes, with small clusters H, which are divided by detailed features. The following figure illustrates the hierarchical state of such clusters.

[0079] 13 shows a technique that combines different cluster analyses at higher and lower levels. First, clusters 1 to n, which are obtained by dividing footprint data group X using cluster analysis based on cosine similarity, constitute the upper level, or first level 110. Clusters 1 to n are further divided by cluster analysis based on principal component analysis, which reduces the dimensions of each of the clusters 1 to n to the second principal component or lower, excluding the first principal component, and then clusters (1-1, 1-2, ... 1-m, 2-1, 2-2, ... 2-m, ... n-1, n-2, ... n-m) constitute the lower level, or second level 112. The cluster analysis techniques for the upper and lower levels will be described later.

[0080] FIG. 14 shows the Euclidean distance and cosine similarity between points. An example will be described in which four points, a first point 120, a second point 121, a third point 122, and a fourth point 123, exist in a vector space. A common method for measuring the proximity between two points is to determine the distance known as Euclidean distance. Euclidean distance is the linear distance between two points, and the smaller the value, the more similar the two points are. On the other hand, this embodiment primarily employs a method that determines the similarity based on cosine similarity. Cosine similarity is the cosine value (cos θ) of the angle θ formed by the position vectors of the two points from the origin. The smaller the angle θ, the closer the cosine value is to 1, and the closer the cosine value is to 1, the more similar the two points are. Generally, cosine similarity is used in situations such as measuring the similarity between sentences, but is rarely used in situations such as measuring the similarity between shapes.

[0081] Since the Euclidean distance (distance 124) from the first point 120 to the second point 121 is shorter than the Euclidean distance (distance 125) from the first point 120 to the third point 122, when the Euclidean distance is used as the standard, the second point 121 is closer to the first point 120 than the third point 122.

[0082] On the other hand, the angle θ formed by the position vectors from the origin of the first point 120 and the second point 121 is １ The angle θ formed by the position vectors from the origin of the first point 120 and the third point 122 is ２is smaller and has a cosine value closer to 1, so when the cosine similarity is used as a reference, the third point 122 is closer to the first point 120 than the second point 121.

[0083] Furthermore, the first point 120 and the fourth point 123 have almost the same position vector from the origin, the angle θ is almost zero, and the cosine similarity is almost 1. Therefore, the fourth point 123 is farther than the second point 121 when the Euclidean distance is used as the standard, but is extremely close to the first point 120 when the cosine similarity is used as the standard.

[0084] FIG. 15 shows the relationship between corresponding points in two footprint models. The footprint models are composed of point clouds, each point representing an anatomical feature of the footprint. Because each footprint model contains points representing the same anatomical feature, the similarity between the footprint models as a whole is determined based on the similarity in the positional relationships between the points. When comparing a relatively small footprint model 130 with a relatively large footprint model 131, cosine similarity is calculated based on the angle formed by the position vectors from the origin between corresponding points among the numerous point clouds contained in each footprint model. In the side view of FIG. 15( a), points 132 and 133, points 134 and 135, and points 136 and 137 correspond to each other, while in the top view of FIG. 15( b), points 138 and 139, points 140 and 141, and points 142 and 143 correspond to each other. In Figure 15(a), the angle formed by the position vectors of points 132 and 133, the angle formed by the position vectors of points 134 and 135, and the angle formed by the position vectors of points 136 and 137 are all almost zero. In Figure 15(b), the angle formed by the position vectors of points 138 and 139, the angle formed by the position vectors of points 140 and 141, and the angle formed by the position vectors of points 142 and 143 are all almost zero. Therefore, it can be said that the foot print model 130 and the foot print model 131 are almost similar in shape although they are different in size.

[0085] 16 shows a comparison of Euclidean distance and cosine similarity when comparing foot models. Foot models with similar foot lengths have a smaller Euclidean distance value, making them more likely to be judged as "similar." Conversely, foot models with a larger Euclidean distance are more likely to be judged as "dissimilar." Foot models with similar shapes have a cosine similarity closer to 1, making them more likely to be judged as "similar." Conversely, foot models with a cosine similarity closer to 0 are more likely to be judged as "dissimilar."

[0086] In the example of Figure 16(a), the two footprint models nearly overlap. The Euclidean distance is relatively small, and the cosine similarity is nearly 1. Therefore, it is easy to determine that they are "similar" whether based on the Euclidean distance or the cosine similarity.

[0087] On the other hand, in the example of Figure 16(b), the two footprint models are clearly different in size. The Euclidean distance is relatively large, and when Euclidean distance is used as a criterion, it is easy to judge that they are "dissimilar." On the other hand, the cosine similarity is almost 1, which is almost the same as the example of Figure 16(a), and when cosine similarity is used as a criterion, it is easy to judge that they are "similar." In other words, when comparing two footprint models, even if there is a large difference when Euclidean distance is used as a criterion, there may be almost no difference when cosine similarity is used as a criterion, indicating that cosine similarity is not affected by size. In the example of Figure 16(b), the two footprint models are different in size but are similar enough to be said to be almost similar in shape.

[0088] Figure 17 shows the difference in classification results between cluster analysis based on cosine similarity and Euclidean distance. As classification results obtained by actual cluster analysis of footprint data groups, the right shows a hierarchical cluster based on Euclidean distance, and the left shows a hierarchical cluster based on cosine similarity.

[0089] In the cluster analysis based on Euclidean distance, in the subdivided clusters at the bottom level, foot length gradually increases from left to right, with foot length decreasing toward the left and increasing toward the right. This also shows that the influence of foot length is dominant in the cluster analysis based on Euclidean distance. In addition, with regard to gender, there is a gradation of gradually increasing numbers of men from left to right, with more women toward the left and more men toward the right. This is because women tend to have smaller feet than men, and it can be seen that the cluster analysis based on Euclidean distance is also influenced by gender.

[0090] On the other hand, cluster analysis based on cosine similarity shows that various foot lengths are randomly mixed from left to right in the lowest level of the subdivided clusters, and that foot length size has almost no effect.Similarly, gender is also randomly mixed, and that gender has almost no effect.

[0091] Figure 18 shows the classification results of cluster analysis based on cosine similarity. The footprint data samples used for cluster analysis were from approximately 5,000 adult men and women. A hierarchical clustering method based on the Ward method was used, based on the cosine similarity between points contained in each footprint data.

[0092] The top three clusters (first cluster 150, second cluster 151, third cluster 152) are considered to be the upper hierarchical level, and the foot shape based on the coordinates of each point obtained by normalizing the foot length size in each of the first cluster 150, second cluster 151, and third cluster 152 is considered to be the average foot.

[0093] By overlaying the foot shape models of the average feet of each cluster, the characteristics of each cluster and the areas where these characteristics are likely to appear can be determined based on the variations in local shapes. For example, one of the characteristics may appear in the difference in instep height, such as the instep of the first cluster 150 being the lowest, the instep of the second cluster 151 being intermediate, and the instep of the third cluster 152 being the highest. For example, one of the characteristics may appear in the difference in the angle of the first toe, such as the angle of the first cluster 150 being most protruding toward the medial side, the angle of the second cluster 151 being intermediate, and the angle of the third cluster 152 being slightly tilted toward the second toe. For example, one of the characteristics may appear in the inclination around the ankle, such as the inclination of the first cluster 150 being most protruding toward the medial side, the inclination of the second cluster 151 being intermediate, and the inclination of the third cluster 152 being the smallest.

[0094] FIG. 19 shows the regression lines of foot length and foot circumference for the top three clusters and the results of principal component analysis of the top three clusters. First, FIG. 19 shows the regression lines showing the relationship between foot length and foot circumference for each of the first cluster 150, the second cluster 151, and the third cluster 152, superimposed on a JIS standard graph similar to that of FIG. 11. As can be seen from this graph, the regression lines are roughly parallel to each other, but their slopes are significantly different from the slope of the regression line showing the relationship between foot length and foot circumference in the JIS standard. Therefore, by designing shoe last variations along the regression lines of each cluster, it is possible to design variations that differ from the gender-specific grading according to the conventional JIS standard, i.e., variations that do not distinguish between men and women and are divided by foot shape characteristics other than foot length size.

[0095] The principal component analysis processor 32 performs principal component analysis on the foot shape data for each cluster, calculating data by adding the first principal component PC1, which represents the variance, to the average foot, which represents the center of gravity for each cluster. The relationship between foot length and foot circumference for this data is plotted on the graph shown in FIG. 19. Each cluster is primarily clustered based on cosine similarity, and although the influence of foot length size is negligible, the data itself contains information about foot length. Therefore, when the principal component analysis processor 32 performs principal component analysis, the first principal component PC1 extracts principal components that are significantly influenced by foot length. However, because the average foot shape is roughly divided into clusters with different shapes based on cosine similarity as a broad classification at higher levels, it is easier to classify the lower levels of clusters based on principal components. In this regard, as plotted by open circles in the graph, the plot showing the relationship between foot length and foot circumference for the average foot + first principal component PC1 for each cluster generally follows the regression line for each cluster, so the results of the cluster analysis and the principal component analysis are consistent.

[0096] FIG. 20 shows the results of principal component analysis on clusters. When the principal component analysis processing unit 32 performs principal component analysis on a cluster, it becomes clear which parameters have a significant influence on the variance of the data in that cluster. In FIG. 20, the first principal component PC1 is set on the vertical axis in the upper part, the second principal component PC2 is set on the vertical axis in the middle part, and the third principal component PC3 is set on the vertical axis in the lower part. On the left side, foot length is set on the horizontal axis, and on the right side, foot circumference is set on the horizontal axis. In each graph, the group of plots surrounded on the left side represents women, and the group of plots surrounded on the right side represents men.

[0097] As shown in the two graphs on the left and right in the top row, the first principal component PC1 not only has a strong correlation with foot length, but also a strong correlation with foot circumference. On the other hand, as shown in the graphs in the middle and bottom rows, the regression lines for the second principal component PC2 and below are nearly horizontal, indicating that they are largely unaffected by foot length or foot circumference. In this embodiment, these characteristics of the results of principal component analysis are utilized to exclude the first principal component PC1, which is affected by foot length and foot circumference, and to use only the second principal component PC2 and below, which are unaffected by foot length and foot circumference, for sub-clustering. Utilizing these characteristics is considered advantageous for constructing clustered variations that focus on detailed shape differences, unlike conventional size-based grading.

[0098] As shown in Figure 13, in the first hierarchical layer 110, which is the upper hierarchical layer, the cluster analysis processing unit 31 performs cluster analysis based on cosine similarity to separate the data into multiple clusters. In the second hierarchical layer 112, which is the lower hierarchical layer, the cluster analysis processing unit 31 performs cluster analysis based on principal component analysis in which the dimensions of each cluster separated in the first hierarchical layer 110 are reduced to two or less principal components, thereby separating the data into many smaller clusters. In this case, the principal component analysis is a principal component analysis in which the dimensions of the principal component analysis results by the principal component analysis processing unit 32 are reduced to two or less principal components excluding the first principal component. In the cluster analysis in the second hierarchical layer 112, hierarchical clustering by Ward's method based on Euclidean distance is used based on the second principal component PC2 and subsequent principal components with a cumulative contribution rate of up to 99%.

[0099] The group of foot shapes divided into many clusters based on detailed features can be used not only for designing variations of shoe lasts and footwear, but also for designing insoles. For example, variations in the shape of details that are not dependent on foot length or circumference can be absorbed by variations in the design of insoles, thereby reducing the number of variations that must be absorbed in the design of shoe lasts and footwear. In other words, by combining variations of shoe lasts and footwear with variations of insoles, an overall range of variations can be designed to cover the entire group of foot shapes.

[0100] In addition, each of the multiple foot shape models included in the foot shape model group may be assigned to a size range in accordance with the JIS standard based on which shoe last in the size range designed in accordance with the JIS standard it fits.

[0101] Since the number of people included in each cluster is different, the proportion of people in each cluster to the entire population of footprint data varies. Meanwhile, if a database holds footprint data groups collected and accumulated for various purposes other than the footprint data group used in the cluster analysis in the first embodiment, the distribution of each foot length size in the population can be calculated based on that data. Therefore, the statistical processing unit 35 can quantify the proportion of each cluster and foot length combination by multiplying the proportion of people in each cluster by the distribution of each foot length size. The proportion of each cluster and foot length combination is quantified when foot length sizes are divided into a range of 22 cm to 31 cm in 1 cm increments.

[0102] When the statistical processing unit 35 designs variations of combination patterns of sizes and shapes of footwear products and insoles based on the proportions of each combination of clusters and foot lengths, it can predict demand for initial inventory according to the proportions of each combination of clusters and foot lengths. In this case, it is possible to exclude combination patterns of clusters and foot lengths predicted to have extremely low demand from the variations of footwear products from the beginning. The statistical processing unit 35 can reflect the demand prediction not only in the demand prediction for initial inventory but also in production plans when inventory decreases due to actual sales. The statistical processing unit 35 may output information regarding the proportions of each combination of clusters and foot lengths to the outside via the output unit 26.

[0103] The present disclosure has been described above based on the embodiments. These embodiments are merely examples, and it will be understood by those skilled in the art that various modifications are possible in the combination of the respective components and processing steps, and that such modifications are also within the scope of the present disclosure. Furthermore, the above-described embodiments can be generalized to provide the following aspects.

[0104] In the above embodiment, an example has been described in which the footwear component design system 200 is applied to customizing and designing an insole. In a modified example, the system may be applied to customizing and designing shoes such as sandals that can be manufactured entirely, including the upper, by additive manufacturing. Alternatively, the system may be applied to customizing and designing some footwear components, such as the sole or upper of a shoe. Alternatively, the system may be applied to customizing and designing products other than footwear and footwear components, such as seamless clothing items such as gloves, hats, and accessories; accessories such as glasses, rings, necklaces, and earrings; accessories such as headphones, hearing aids, and ear monitors; computer mice, mouthpieces, and bedding such as pillows.

[0105] a design storage unit that stores a plurality of basic design data as three-dimensional design data used for additive manufacturing of footwear parts corresponding to the respective foot shapes of the plurality of three-dimensional homologous models; a measurement data acquisition unit that acquires measurement data of the three-dimensional coordinate groups that indicate anatomical features corresponding to the three-dimensional homologous models of a foot shape including at least the sole and lateral surfaces of a subject's foot; a model determination unit that determines a three-dimensional homologous model that resembles the subject's foot shape from the plurality of three-dimensional homologous models by evaluating the similarity between coordinates between the three-dimensional coordinate groups in the measurement data and the three-dimensional coordinate groups in the plurality of three-dimensional homologous models; a basic design determination unit that determines basic design data corresponding to the determined three-dimensional homologous model from the plurality of basic design data; and a design processing unit that generates footwear part design data corresponding to the subject's foot shape based on the determined basic design data.

[0106] As a result, the footwear component design device of aspect 1 can design footwear or footwear components that are optimal for the foot size and foot shape of each wearer or prospective purchaser. Since the basic design data is determined from multiple three-dimensional homologous models prepared in advance, design validity can be ensured. Therefore, it is possible to achieve both high design freedom and design validity for footwear or footwear components.

[0107] [Aspect 2] The footwear component design device according to Aspect 1, wherein the model determination unit obtains position vectors from a predetermined reference point to each point in the plurality of three-dimensional homologous models, obtains position vectors from the predetermined reference point to each point of the three-dimensional homologous models in the measurement data, and determines a three-dimensional homologous model that resembles the body shape of the subject from the plurality of three-dimensional homologous models based on the distance between corresponding points of the three-dimensional homologous models or the similarity of the position vectors.

[0108] This allows the footwear component design device of aspect 2 to offer a wider range of sizes that fit individual feet more precisely than the conventional size range that is standardized based only on foot length and foot width, thereby enabling the design of footwear or footwear components that are optimal for the foot size and foot shape of each wearer or prospective purchaser.

[0109] [Aspect 3] The footwear component design device according to Aspect 1 or 2, wherein the model storage unit stores the plurality of three-dimensional homologous models in a form in which the plurality of three-dimensional homologous models are divided into a plurality of clusters by cluster analysis based on cosine similarities between position vectors from a predetermined reference point to each point of the three-dimensional homologous models.

[0110] As a result, the footwear component design device of aspect 3 can classify based on similarity, which is less affected by size, and can offer a range of sizes that precisely fit the feet of each individual, regardless of foot size. This makes it possible to design footwear or footwear components that are optimal for the foot size and foot shape of each wearer or prospective purchaser.

[0111] Aspect 4: The footwear component design device according to Aspect 3, wherein the model storage unit stores the plurality of three-dimensional homologous models in a form in which an upper level is divided into a plurality of clusters by cluster analysis based on cosine similarity, and a lower level is further divided into a plurality of clusters by cluster analysis based on principal component analysis in which dimensions are reduced to a second principal component or less excluding a first principal component when principal component analysis is performed on each cluster in the upper level.

[0112] This allows the footwear component design device of aspect 4 to classify items based on factors that are less affected by gender and size, and also allows for a range of sizes that precisely fit the feet of each individual, regardless of gender or size. This makes it possible to design footwear or footwear components that are optimal for the foot size and shape of each wearer or prospective purchaser.

[0113] [Aspect 5] The footwear component design device according to any one of Aspects 1 to 4, further comprising an individual information acquisition unit that acquires individual information indicating individual modifications specified in accordance with the subject's wishes as modifications to the basic design of the footwear component, wherein the design processing unit generates the footwear component design data by adding the individual modifications to the basic design based on the basic design data and the individual information.

[0114] As a result, the footwear component design device of aspect 5 can not only design footwear or footwear components that are optimal for the foot size and foot shape of each wearer or prospective purchaser, but can also reflect the prospective purchaser's preferences in the design and decoration. In this way, it is possible to customize the design of one-of-a-kind footwear or footwear components that reflect individual differences and preferences. Since the basic design data is determined from multiple three-dimensional homologous models prepared in advance, design validity can be ensured. Therefore, it is possible to achieve both high design freedom and design validity for footwear or footwear components.

[0115] [Aspect 6] The footwear component design device according to Aspect 5, wherein the individual information acquisition unit acquires, as the individual information, information including text designated in accordance with the subject's wishes as a decorative shape to be added to the outer surface of the footwear component, and the design processing unit generates the footwear component design data by adding a decorative shape modeled on the designated text to the basic design based on the basic design data and the individual information.

[0116] As a result, the footwear component design device of aspect 6 can not only design footwear or footwear components that are optimal for the foot size and foot shape of each wearer or prospective purchaser, but can also reflect the prospective purchaser's preferences in the design and decoration. In this way, it is possible to design a unique, customized piece of footwear or footwear component that reflects individual differences and preferences. Since the basic design data is determined from multiple three-dimensional homologous models prepared in advance, design validity can be ensured. Therefore, it is possible to achieve both high design freedom and design validity for footwear or footwear components.

[0117] [Aspect 7] The footwear component design device according to any one of Aspects 1 to 6, further comprising a footwear information acquisition unit that acquires footwear information indicating the type of footwear designated by the subject, wherein the plurality of basic design data are designed to correspond to the shoe last shape of at least one of a plurality of types of footwear, and the basic design determination unit determines the basic design data based on a three-dimensional homologous model that resembles the foot shape of the subject and the footwear information.

[0118] As a result, the footwear component design device of aspect 7 can not only design footwear components that are optimal for the foot size and foot shape of each wearer or prospective purchaser, but also design footwear components of sizes and shapes that are suitable for the footwear preferences of each wearer or prospective purchaser.

[0119] [Aspect 8] The footwear component design device according to any one of Aspects 1 to 7, wherein the footwear component is a footwear insole.

[0120] As a result, the footwear component design device of aspect 8 can not only design an insole that is optimal for the foot size and foot shape of each wearer or prospective purchaser, but can also reflect the prospective purchaser's preferences in the design and decoration. In this way, it is possible to design a unique, customized insole that reflects individual differences and preferences. Since basic design data is determined from multiple three-dimensional homologous models prepared in advance, design validity can be ensured. Therefore, it is possible to achieve both high design freedom and design validity for the insole.

[0121] [Aspect 9] A footwear component design method comprising the steps of: reading out a plurality of three-dimensional homologous models corresponding to a plurality of foot shapes, each of which represents a group of three-dimensional coordinates indicating anatomical features of the foot; acquiring measurement data of the group of three-dimensional coordinates indicating anatomical features corresponding to the three-dimensional homologous models of a foot shape including at least the sole and lateral surface of a subject's foot; determining a three-dimensional homologous model that resembles the subject's foot shape from the plurality of three-dimensional homologous models by evaluating the similarity between coordinates between the group of three-dimensional coordinates in the measurement data and the group of three-dimensional coordinates in the plurality of three-dimensional homologous models; determining basic design data corresponding to the determined three-dimensional homologous model from a plurality of basic design data, which is three-dimensional design data used for additive manufacturing of footwear components corresponding to each foot shape of the plurality of three-dimensional homologous models; and generating footwear component design data corresponding to the subject's foot shape based on the determined basic design data.

[0122] As a result, the footwear component design method of aspect 9 makes it possible to design footwear or footwear components that are optimal for the foot size and foot shape of each wearer or prospective purchaser. Since the basic design data is determined from multiple three-dimensional homologous models prepared in advance, design validity can be ensured. Therefore, it is possible to achieve both high design freedom and design validity for footwear or footwear components.

[0123] The present disclosure relates to a footwear component design device and a footwear component design method.

[0124] 104 Three-dimensional homologous model, 212 Measurement data acquisition unit, 216 Individual information acquisition unit, 218 Footwear information acquisition unit, 220 Model storage unit, 222 Model determination unit, 230 Design storage unit, 232 Basic design determination unit, 234 Design processing unit, 250 Footwear part design device, 300 Insole, 303 Side.

Claims

1. A footwear component design device comprising: a model memory unit that pre-stores three-dimensional coordinate groups indicating anatomical features of the foot as multiple three-dimensional homologous models for multiple foot shapes; a design memory unit that stores multiple basic design data as three-dimensional design data used for additive manufacturing of footwear components corresponding to each of the foot shapes of the multiple three-dimensional homologous models; a measurement data acquisition unit that acquires measurement data of three-dimensional coordinate groups indicating anatomical features corresponding to the three-dimensional homologous models in a foot shape including at least the sole and lateral surfaces of a subject's foot; a model determination unit that determines a three-dimensional homologous model that resembles the subject's foot shape from the multiple three-dimensional homologous models by evaluating the similarity between coordinates between the three-dimensional coordinate groups in the measurement data and the three-dimensional coordinate groups in the multiple three-dimensional homologous models; a basic design determination unit that determines basic design data corresponding to the determined three-dimensional homologous model from the multiple basic design data; and a design processing unit that generates footwear component design data corresponding to the subject's foot shape based on the determined basic design data.

2. The footwear component design device of claim 1, characterized in that the model determination unit obtains position vectors from a predetermined reference point to each point in the multiple three-dimensional homology models, obtains position vectors from the predetermined reference point to each point of the three-dimensional homology models in the measurement data, and determines a three-dimensional homology model that is similar to the subject's body shape from the multiple three-dimensional homology models based on the distance between points of corresponding three-dimensional homology models or the similarity of the position vectors.

3. The footwear part design device of claim 1 or 2, characterized in that the model memory unit stores the multiple three-dimensional homology models in a form divided into multiple clusters by cluster analysis based on the cosine similarity between position vectors from a specified reference point to each point of the three-dimensional homology models.

4. The footwear part design device according to claim 3, characterized in that the model memory unit stores the multiple three-dimensional homology models in a form in which the upper hierarchy is divided into multiple clusters by cluster analysis based on cosine similarity, and the lower hierarchy is further divided into multiple clusters by cluster analysis based on principal component analysis in which the dimensions are reduced to the second principal component or less excluding the first principal component when principal component analysis is performed for each cluster in the upper hierarchy.

5. A footwear part design device as described in any one of claims 1 to 4, further comprising an individual information acquisition unit that acquires individual information indicating individual modifications specified in accordance with the wishes of the subject as modifications to the basic design of the footwear part, and the design processing unit generates the footwear part design data by adding the individual modifications to the basic design based on the basic design data and the individual information.

6. The footwear part designing device of claim 5, characterized in that the individual information acquisition unit acquires as the individual information information including text specified in accordance with the subject's wishes as a decorative shape to be added to the outer surface of the footwear part, and the design processing unit generates the footwear part design data by adding a decorative shape modeled on the specified text to the basic design based on the basic design data and the individual information.

7. A footwear component design device as described in any of claims 1 to 6, further comprising a footwear information acquisition unit that acquires footwear information indicating the type of footwear specified by the subject, the multiple basic design data being designed to correspond to the shoe shape of at least one of multiple types of footwear, and the basic design determination unit determining the basic design data based on a three-dimensional homologous model similar to the subject's foot shape and the footwear information.

8. A footwear part designing device according to any one of claims 1 to 7, wherein the footwear part is an insole of footwear.

9. A footwear component design method comprising the steps of: reading out a plurality of three-dimensional homologous models for a plurality of foot shapes, the three-dimensional coordinate groups indicating anatomical features of the foot; acquiring measurement data of the three-dimensional coordinate groups indicating anatomical features corresponding to the three-dimensional homologous models in a foot shape including at least the sole and sides of a subject's foot; determining a three-dimensional homologous model similar to the subject's foot shape from the plurality of three-dimensional homologous models by evaluating the similarity between coordinates between the three-dimensional coordinate groups in the measurement data and the three-dimensional coordinate groups in the plurality of three-dimensional homologous models; determining basic design data corresponding to the determined three-dimensional homologous model from a plurality of basic design data, which is three-dimensional design data used for additive manufacturing of footwear components corresponding to each foot shape of the plurality of three-dimensional homologous models; and generating footwear component design data corresponding to the subject's foot shape based on the determined basic design data.

Images