JP2018085086A - Information providing device and information providing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information providing device and an information providing method that can quickly provide at least one useful and accurate information on an objective image, styling, and coordination suitable for an object using an evaluation result obtained on an article attribute of the object.SOLUTION: An information providing device and an information providing method selects S1 at least one article attribute constituting an article, compares S5 at least one article attribute measurement result acquired for each of the article attributes with a reference value, then, based on the comparison result and at least one tendency prescribed for the article attribute, specifies S6 an adaptation information category, selects S7 an article adaptation information included in the adaptation information category from a plurality of article adaptation information groups, and outputs S8 it.SELECTED DRAWING: Figure 12

Description

  The present invention relates to an information providing apparatus and an information providing information method, and more specifically, systematically classifying based on article attribute information including measurement data such as the type and name, color, shape / size, and material of the article. The present invention relates to an information providing apparatus and an information providing method for providing information relating to styling, information relating to a person who suits the article, and the like from a plurality of stored article conformity information groups.

  Recently, not only for movie actors and TV program performers, but also for stylists who want to improve their image in situations such as shooting personal photos such as PR photos. Coordinating costumes, clothes, accessories, accessories, etc. (hereinafter referred to as articles) that fit the scene. After observing the customer as a whole and interviewing the stylist, the stylist coordinates the customer and the goods suitable for the styling from the information and knowledge about the styling classified and arranged based on the experience of the stylist. Yes. However, in reality, such coordination is largely dependent on the stylist's subjectivity, and it also matches the impression that the customer wants to give to goods and third parties that match the image that the customer wants to dress due to the observation ability of the stylist, lack of experience and knowledge, etc. There is a problem that it is difficult for a stylist to coordinate the stuff.

  Moreover, it is preferable to construct a system that solves this problem and performs coordination by a stylist through search by a computer. By the way, in recent years, proposals relating to entertainment such as fortune telling using a computer (for example, see Patent Document 1) and proposals regarding makeup in the beauty field (for example, see Patent Document 2) have been made. However, a certain standard is required to analyze the image of an article, search for and judge an article with the same attribute / image, and improve the consistency / unification of the article search. There are still no devices, systems or methods that can be used for coordination. If such an apparatus, system, and method can be constructed, the present invention is not limited to coordination by a stylist, and can be effectively applied to, for example, recommending a product to a purchaser on an EC site.

  By the way, the present inventors have already provided at least one useful and accurate information such as styling suitable for the operator to the operator quickly using the evaluation result obtained for the face part attribute of the operator. It proposes an information providing apparatus and information providing method that can be used. In the present specification, an article that is used for measurement of an attribute or the like in the information providing apparatus of the present invention is particularly referred to as an “object”.

JP 2005-352803 A JP 2014-149841 A

  As a result of intensive studies on the expansion of the further application range of the information providing apparatus and the information providing method according to the proposal, the present inventors have made an examination based on the evaluation result on the attribute of the article. The present inventors have found that information on styling and coordination suitable for the article and information such as a person image suitable for the article can be extracted, and the present invention has been completed. That is, the present invention uses useful and accurate information (hereinafter referred to as article conformity information) regarding an objective image, styling, coordination, etc. suitable for an object using article attribute information such as an evaluation result obtained for the object. It is an object to provide an information providing apparatus and an information providing method capable of promptly providing at least one.

  According to one aspect of the present invention, the object includes at least a storage unit, an attribute information acquisition unit, an evaluation unit, a compatible information category identification unit, an information selection unit, and an output unit, and the storage unit includes the object. A plurality of item attribute information, a plurality of tendencies, a plurality of matching information categories, and a plurality of item matching information groups belonging to each of the matching information categories, each of the plurality of item attribute information, The type / name of the article, at least one article attribute defined in advance for each of the articles, and at least one article attribute reference value that serves as a reference for each article attribute. It is represented as a linear axis including the article attribute reference value in relation to the article attribute information, and the plurality of articles suitable for the at least one article attribute reference value on the linear axis. The information group is stipulated to be classified into at least two compatible information categories, and the attribute information acquisition unit is selected for the object together with its type / name and attribute name as article attribute information for the object. Each of the at least one article attribute measurement result is acquired as an object attribute measurement result, and the evaluation unit acquires article attribute information of the object through the attribute information acquisition unit, and the type / name and attribute of the object The article attribute reference value called from the storage unit based on the name is compared with the article attribute measurement value of the object, and each comparison result is sent to the matching information category specifying unit, and the matching information category specifying unit Is based on the comparison result received for each tendency related to the attribute of each article called from the storage unit. In this case, one overlapping information category is identified from each corresponding information category classified by each tendency, or if it cannot be narrowed down to one corresponding information category, the duplication number for each corresponding information category The information selection unit selects at least one item conforming information belonging to the identified conforming information category from among the item conforming information group in the storage unit. The output unit is achieved by an information providing apparatus configured to output each selected article matching information.

  According to another aspect of the present invention, the object is also attribute information for acquiring at least one item attribute selected for the object and the measurement result thereof together with the type and name as the item attribute information of the object. An article attribute reference value that is input from the storage unit based on the type / name and attribute name of the object, and the attribute measurement value of the object; , Each of the comparison results is input to the evaluation step of sending each comparison result to the matching information category specifying unit, and the three tendencies of the plurality of items related to the item attribute information of the object called from the storage unit. Identifying one matching information category from each matching information category classified according to each tendency based on the comparison result received, or 1 If it is not possible to narrow down to the conformance information category, the conformity information category identifying step of identifying a plurality of conformance information categories by ranking according to the overlap for each conformance information category, and at least belonging to the identified conformance information category An article conformity information selection step for selecting one article conformity information from the article conformity information group stored in the storage unit; and an output step for outputting each selected article conformity information. This is achieved by the information providing method.

  The object is also achieved by a program for causing a computer to execute the information providing method of the present invention.

  According to the information providing apparatus and the information providing method of the present invention, by selecting at least one attribute of an object and inputting at least one article attribute measurement result for the selected attribute to the information providing apparatus, Based on the comparison result between the article attribute measurement result of the object and the article attribute reference value corresponding to the measurement result, it is associated with at least one tendency among the plurality of conformance information categories for classifying the plurality of article conformity information groups. At least one matching information category can be specified, and article matching information belonging to this matching information category can be output. This makes it possible to quickly provide at least one useful and accurate information related to an objective image and styling adapted to an object using a computer, and further to coordination thereof.

It is a block diagram which shows roughly the hardware structural example of one Embodiment of the information provision apparatus of this invention. FIG. 2 is a block diagram schematically showing a functional configuration example of an arithmetic control unit shown in FIG. 1. It is a block diagram which shows roughly an example of the memory | storage part shown in FIG. It is a table | surface which shows an example of the relationship between goods attribute information in the case of using T-shirt as an article, and this and tendency. FIG. 5 is a diagram illustrating an example of a relationship between article attribute information and tendency shown in FIG. 4. It is a table | surface which shows an example of the article | item attribute information in the case of using a pencil as an article | item, and the relationship between this and tendency. FIG. 7 is a diagram illustrating an example of a relationship between article attribute information and tendency shown in FIG. 6. FIG. 8 is a diagram illustrating an example of a relationship between colors and categories of various systems related to FIGS. It is a figure which shows an example of the orthogonal coordinate system (styling map) which showed the relationship between an article attribute and tendency, and a compatible information category. It is a table | surface for demonstrating an example of the relationship between a tendency and a compatible information category. It is a figure which shows an example of the relationship between a conformity information category and an article conformity information group. It is a flowchart which shows an example of the procedure of the information provision method using the information provision apparatus shown in FIG.

An embodiment of the information providing apparatus of the present invention will be described in detail with reference to FIGS. 1 to 3 attached to the information providing apparatus of the present invention.
[Hardware configuration example]
1 is a block diagram schematically showing a hardware configuration example of the present embodiment, FIG. 2 is a block diagram schematically showing a functional configuration example of an arithmetic control unit shown in FIG. 1, and FIG. 3 is shown in FIG. It is a figure which shows typically an example of the information memorize | stored in the memory | storage part. The information providing apparatus 1 according to the present embodiment uses useful information of an article statistically accumulated in advance (hereinafter, this information is referred to as “article conformance information”, and a plurality of article conformance information is referred to as an “article conformance information group”. Measurement data and input data (hereinafter referred to as “article attribute values”) regarding attributes (hereinafter referred to as “article attributes”) such as color, shape / size, and material of the articles. ) To extract at least one item conforming information suitable for styling and an operator from a plurality of item conforming information groups.

  The information providing apparatus 1 includes an arithmetic control unit 2, a storage unit 3, an input unit 5, an output unit 7, and an internal bus 4. Here, examples of the information providing apparatus 1 include a general-purpose personal computer, a tablet terminal, and a smartphone. These may be stand-alone terminals, or may be client terminals connected to a server via a telecommunication means (not shown) in a client / server system. In the latter case, the storage unit 3 includes a storage unit (not shown) provided on the server side. In addition, at least some of the functions of the respective units in the information providing apparatus 1 may be configured so that artificial intelligence (AI) is responsible. The internal bus 4 has a function of mutually connecting the above-described units, that is, the arithmetic control unit 2, the storage unit 3, the input unit 5, and the output unit 7.

The arithmetic control unit 2 performs various controls on each unit by a CPU or the like. The arithmetic control unit 2 is configured by an arithmetic control device including a CPU, MPU, ROM, and the like in hardware. The arithmetic control unit 2 executes a startup program stored in the ROM after the information providing apparatus of the present invention is turned on, and stores an operating system (OS), various processing drivers, and the present invention stored in the storage unit 3. The program for executing the information providing method and various data are read into the RAM which is the main storage device, and the display information expanded in the RAM is output to the display 10 of the output unit 7 or to the printer 11. Or output. As shown in FIG. 2, the arithmetic control unit 2 in the present embodiment includes functional units such as an attribute information acquisition unit 21 including an image data analysis unit 25, an evaluation unit 22, a matching information category identification unit 23, and an information selection unit 24. It is configured. Each of these functional units will be described later.
Note that the program and various data for executing the information providing method of the present invention may be stored in a recording medium such as a CD-R or a DVD-R. Accordingly, the recording medium can be stored in the storage unit 3 as necessary.

  The storage unit 3 temporarily or permanently stores and stores programs executed by the arithmetic control unit 2 and various data. The storage unit 3 includes a volatile memory such as various RAMs, a magnetic disk such as a hard disk drive (HDD), an optical disk, and a secondary storage device such as a nonvolatile memory. As shown in FIG. 3, a plurality of item attribute information, a plurality of (three to four or more) tendencies, a plurality of matching information categories, and a plurality of item matching information groups are stored in the storage unit 3 in advance. It is remembered. Information stored in the storage unit 3 of this embodiment illustrated in FIG. 3 will be described below with reference to FIGS.

(Article attribute information)
The article attribute information is the name / type of the article, at least one article attribute for the article, and at least one attribute reference value (hereinafter referred to as “article attribute reference value”) defined in advance for each article attribute. ). Among these, in this embodiment, the article attribute is defined by the three types of article shape / size, material, and color as described above, but is not limited thereto, and is added due to future advancement of measurement technology, etc. Can be defined automatically. Moreover, since the article attribute reference value may be changed due to future advancement of measurement technology or the like, it can be updated as appropriate. The article attribute reference value is mainly expressed quantitatively with a numerical value, but may be expressed qualitatively other than the numerical value.

  As shown in FIGS. 4 to 7, each item attribute further has two attributes in principle at both poles. For example, when the article attribute is a shape / size, an attribute related to “straight-curve”, an attribute related to “small, thin, short—large, thick, long” and an attribute related to “familiar-conspicuous” Is prescribed. The attribute “straight-curved” is that when the article is a T-shirt (see FIG. 4), the collar opening is a round shape, the length-to-width ratio is an ellipse with a length of 3: 3 and a width of 20 cm. If the article is a pencil (see FIG. 6), the cross section is a regular hexagon with a side of 4 mm. It is related to whether the number of strokes is relatively smaller than that, or the number of corners is increased to approach a circle. The attribute “small, thin, short-large, thick, long” indicates that the article is a T-shirt (see FIG. 4), the height is 65 cm, the width is 45 cm, and the sleeve length is 25 cm, and the article is a pencil ( 6), the cross section is a regular hexagon having a side of 4 mm, and the length is 175 mm as a reference, which is relatively small (thin, short) or larger (thick, long). In addition, the attribute “familiar-conspicuous” indicates that the item is plain and has no pattern, whether the article is a T-shirt (see FIG. 4) or the article is a pencil (see FIG. 6). It is related to whether the standout is conspicuous (the outline of the pattern becomes clear) or inconspicuous (the outline of the pattern is blurred and becomes gradation).

  In addition, when the article attribute is a material, an attribute relating to “flat-roughness”, an attribute relating to “thin, light, soft-thick, heavy, hard”, and “matte (matte) whether or not there is gloss. )-Glossy (present) "attribute. When the article is a T-shirt (see FIG. 4), the attribute “flat-concave / concave” is based on 28 gauge, the gauge number is relatively larger and the mesh is smaller and flat, or the gauge number is smaller. If the article is a pencil (see FIG. 6), the surface is embossed with letters or the like on the basis that the surface is unprocessed, or the surface is polished by polishing or the like. It is about whether it is smooth. Also, the attribute “thin, light, soft-thick, heavy, hard” is an article in which the article is a T-shirt (see FIG. 4) and is relatively larger than the 28th and the thread is thin. Plastics and rubbers based on the fact that they are thin (light and soft), thin and thick and heavy (heavy and hard), and if the article is a pencil (see Figure 6), it is woody Depending on whether it is light and soft, or heavy and hard by metallic objects such as metal. In addition, the attribute “matte-shiny” has a gloss that is higher than this standard, based on the fact that it is unprocessed, whether the article is a T-shirt (see FIG. 4) or the article is a pencil (see FIG. 6). It is about whether or not it exists.

  On the other hand, when the article attribute is a color, an attribute relating to “blue base-yellow base”, an attribute relating to “bright-dark” relating to lightness, and an attribute relating to “moderate-bright” relating to saturation are defined. ing. The attribute "Blue Base-Yellow Base" is based on an intermediate color that does not belong to either blue or yellow (article attribute reference value), and the color is more bluish or yellowish than the standard. It is about. The attribute “bright-dark” relates to whether the brightness is relatively brighter or darker than the medium level that is the reference value (article attribute reference value). Further, the attribute “Odaka-Azaka” is related to whether the saturation is relatively larger (bright color) or smaller (bright color with no buzz) than the medium value that is the reference value (item attribute reference value). Is.

(Tendency)
4 to 7 also show the relationship between the article attribute information and the tendency. As shown in these figures, the article attributes are classified into three tendencies. That is, each attribute relating to “blue base-yellow base”, “straight line-curve”, and “flat-convexity” has a cool-warm tendency, “bright-dark”, “small, thin, short-large, thick”. "Long" and "thin, light, soft-thick, heavy, hard" attributes are light-deep, "moderate-bright", "familiar-conspicuous" and "matte-matte" Each of the attributes relating to “Yes” corresponds to gradation-contrast tendency. In this way, each tendency is related to the item attribute information, and is named by an impression or image received from two attributes of each pole in each item attribute, and is a linear axis (scale axis) serving as an index of the attributes of these poles. ) (See FIGS. 5 and 7). The article attribute reference value is included in any intermediate position on the straight line. In addition, the attribute in each item attribute is not limited to these, and if further analysis / examination discovers an attribute that has a statistically significant relationship with the item conformity information group, It is possible to replace the attribute with the current attribute or additionally employ the attribute. In the present invention, the present invention is not limited to these tendencies, and further statistical processing is repeated to obtain knowledge about tendencies that are more relevant between the article attributes and the article conformity information group. It goes without saying that new trends can be added or changed if possible.

(Product conformity information group)
The article conformity information group is collected by associating information that is recognized to be very highly related to past article attribute information by statistical processing in association with the respective attributes of the respective polarities. This product suitability information group includes information on people who look good on the product and other information and advice. Other information includes information that matches the image and impression of the article, such as information about tableware; information about interior such as furniture and interior (color of floors and walls, etc.); information about graphic design such as logo color; Information on promotions, product planning (package colors, etc.); information on hair and makeup, fashion and other styling; teaching materials for styling certification. The article conformity information group also includes information that further coordinates the information. These article conformity information groups can be arranged one-dimensionally depending on the strength of their attributes for each tendency.

(Compliance information category)
The article conformity information group is classified into at least two categories related to the attributes of the two poles at the boundary of at least one article attribute reference value by any one tendency. Hereinafter, this category is referred to as a “conforming information category”. When there is one article attribute reference value, the article conformity information group is divided into two conformance information categories related to bipolar attributes with the article attribute reference value as a boundary. For example, in the case of a cool-warm tendency, the product conformity information group is linear in one pole, a conformity information category (cool category) for an attribute that gives a cool impression such as blue, and a curvilinear in the other pole. It can be divided into conformity information categories (warm categories) for attributes that give a warm impression such as yellow. In the light-deep tendency, the matching information category (light category) for attributes that give a light (lightness) impression such as bright, small, and light, and deep (such as dark, large, and heavy) Applicable information category (deep category) for attributes that give an impression) If the gradation-contrast tendency, it is gentle, familiar, matte (matte) or other gradation (soft color impression) The range of information about the attribute (gradation category) and the matching information category (contrast category) for the attribute with contrast (giving a vivid impression) such as vivid, conspicuous, or glossy Each item conformity information group is divided. In addition, when there are two or more article attribute reference values, the compliance information category that is the largest article attribute reference value or more, the second article attribute reference value or more, and the conformance information that is less than the maximum article attribute reference value. The category is classified as a conformity information category that is greater than or equal to the third item attribute reference value and less than the second item attribute reference value, and so on.

Articles can take various colors due to the influence of the kind of light, the difference in unique color, gradation, etc., and how light is reflected. Therefore, some categories will be described as to which category the appropriate color of the PCCS tone belongs. For example, for the red line, PCCS tone V24 (L = + 45.4, a = + 49.2, b = −8.0) is in the cool category, and PCCS tone dp2 (L− + 42.3, a = + 43. 5, b = + 10.7) is in the warm category, PCCS tone It2 (L = + 73.3, a = + 25.7, b = + 8.9) is in the light category, PCCS tone dk2 (L = + 26.5, a = + 34.1, b = + 8.9) is in the deep category, PCCS tone sf2 (L = + 65.3, a = + 29.9, b = + 7.8) is in the gradation category, and PCCS tone 2 (L = +49.4, a = + 70.0, b = + 14.5) correspond to the contrast category, respectively.

  For the yellow system, the PCCS tone b8 (L = + 75.7, a = + 2.3, b = + 61.9) is in the cool category, and the PCCS tone V7 (L− + 76.9, a = + 14.3). , B = + 70.1) is in the warm category, PCCS tone It8 (L = + 87.9, a = + 2.3, b = + 38.2) is in the light category, and PCCS tone dp6 (L = + 56.4, a = + 18.9, b = + 38.6) is in the deep category, PCCS tone sf8 (L = + 73.6, a = + 2.0, b = + 40.2) is in the gradation category, and PCCS tone 8 (L = + 84) 0.0, a = + 5.2, b = + 82.9) correspond to the contrast category, respectively.

  For the green line, the PCCS tone dp14 (L = + 41.5, a = −28.8, b = + 0.6) is in the cool category, and the PCCS tone dp10 (L− + 53.6, a = −13). .3, b = + 43.2) is in the warm category, PCCS tone It10 (L = + 80.3, a = −6.6, b = + 27.3) is in the light category, and PCCS tone dk12 (L = + 40. 1, a = −11.5, b = + 8.7) is in the deep category, PCCS tone d12 (L = + 57.5, a = −12.8, b = + 11.3) is in the gradation category, PCCS tone 12 (L = + 80.5, a = -21.9, b = + 17.9) correspond to the contrast category.

  For the blue system, PCCS tone V19 (L = + 42.2, a = + 2.7, b = −36.9) is in the cool category, and PCCS tone dp16 (L− + 42.5, a = −18). .9, b = −17.3) is in the warm category, PCCS tone It16 (L = + 70.1, a = −19.0, b = −17.4) is in the light category, and PCCS tone dk18 (L = +29.7, a = −1.2, b = −19.0) are in the deep category, and PCCS tone d16 (L = + 50.3, a = −11.2, b = −14.2) is the gradation category. PCCS tone 16 (L = + 47.4, a = -20.0, b = -32.8) corresponds to the contrast category.

  Next, when at least two different tendencies are obtained from the three tendencies, such as when a plurality of article attributes are selected by the customer, the relationship between the at least two tendencies and the matching information category will be described. To do. In addition, in order to simplify the description, each tendency has a single item attribute reference value included therein. FIG. 9 shows the relationship between the article attribute and tendency and the matching information category. Of the three tendencies, the cool-warm tendency is plotted on the horizontal axis, and the light-deep trend is plotted on the vertical axis. An example of an orthogonal coordinate system is shown. In addition, a virtual axis of gradation-contrast tendency is provided in a diagonally 45 ° downward and rightward direction (the same applies to FIGS. 5 and 7). In this Cartesian coordinate system, the cool-warm tendency on the horizontal axis and the light-deep tendency on the vertical axis are provided so as to intersect each other at the origin with the respective article attribute reference values as the origin. In the orthogonal coordinate system, although not shown, the article matching information group is two-dimensionally arranged according to the strength of the attribute from the origin, for example. In the following, this orthogonal coordinate system is referred to as a “styling map”, and in particular, the first quadrant (I) is “bright taste”, the second quadrant (II) is “aqua taste”, and the third quadrant (III) is “ The “crystal taste” and the fourth quadrant (IV) will be called “artist”.

  In this styling map, the first quadrant (bright taste) and the fourth quadrant (artist) become the warm category, and the second quadrant (aqua taste) and third quadrant (crystal taste) due to the cool-warm tendency on the horizontal axis. Is a cool category. Also, due to the light-deep tendency on the vertical axis, the first quadrant (bright taste) and the second quadrant (aqua taste) are in the light category, and the third quadrant (crystal taste) and the fourth quadrant (artist) are in the deep category. It becomes. In addition, as shown in FIG. 6, the first quadrant (bright taste) and the third quadrant (crystal taste) that include a 45-degree obliquely rising virtual axis passing through the origin are contrast categories due to the gradation-contrast tendency. The second quadrant (aqua taste) and the fourth quadrant (artist), which include a hypothetical axis that is 45 degrees diagonally downward and passes through the origin, are gradation categories. In addition, since this styling map is suitable for explaining the relationship between the respective tendencies and the matching information category in the present embodiment, the relationship between the propensity and the matching information category has been described. Instead of using such a styling map, it can also be shown using other methods and structures, and is not limited to using a styling map.

  FIG. 10 summarizes the relationship between the matching information categories classified by the three tendencies shown in FIGS. 4 to 7 and the first to fourth quadrants shown in FIG. 9 in a table format. As shown in this table, each trend information category of each tendency is composed of a combination of two quadrants of the first to fourth quadrants. In other words, each of the first to fourth quadrants is a part of three of the six compatible information categories. That is, the “bright taste” in the first quadrant is a part of each of the matching information categories of light, warm and contrast. Further, the “aqua taste” in the second quadrant is a part of each matching information category of light, cool and gradation. Furthermore, the “Crystal taste” in the third quadrant is part of the cool, deep and contrast fit information categories. Furthermore, the “artist” in the fourth quadrant is a part of the deep, warm, and gradation matching information categories.

  FIG. 11 shows the relationship between the quadrants and the article matching information group. In this figure, information relating to an object (styling) and information relating to a person image that the object suits are posted as an article conformity information group classified by the conformance information category, but the present invention is not limited thereto. As information on the object (styling), for example, when the matching information category is a bright taste, the object can be objectively classified as “an object with a casual feeling”. Further, when the conformity information category is an aqua taste, the object can be objectively classified as “good to touch”. Furthermore, when the conformity information category is crystal taste, the object can be objectively classified as a “metallic object”. Furthermore, when the relevant information category is an artist, the object can be objectively classified as “an object with depth”.

  In addition, for example, when the relevant information category is a bright taste, the person has a “bright and cheerful” personality, and the service type is “to speak early” (to other people). How to deal with). Further, when the matching information category is an aqua taste, the person has a “gentle and gentle” personality and can be classified into a customer service type of “listening to talk”. Furthermore, when the matching information category is a crystal taste, the person can be classified into a customer service type of “energetic and quick action” and “quickly respond with quick action”. Furthermore, when the relevant information category is an artist, the person has a personality of “there is calmness and thinks carefully” and can be classified into a customer service type that “follows the procedure”.

  The input unit 5 in the present embodiment illustrated in FIG. 1 includes an input terminal 9 and a device driver 6. Information input from the input terminal 9 or the device driver 6 is stored in the storage unit 3 via the internal bus 4. The input terminal 9 is a device for the operator to input the selection result of the article attribute. Here, the operator selects an article attribute with reference to the combination of article attributes shown in FIG. By combining at least two sets in this way, it is possible to further narrow down the matching information category to suit the operator as described later, and to provide more appropriate article matching information to the operator. This is because it can. The input unit 5 mainly functions as the attribute information acquisition unit 21 (see FIG. 2).

  Using this, the input terminal 9 is mainly used to input a selection instruction and necessary data. Examples of the input terminal 9 include a mouse, a pointing device, a keyboard, and a touch panel. When the input terminal 9 is a mouse or a pointing device, the input terminal 9 can be used for a selection instruction or data input using a mouse cursor or a point on the selection screen displayed on the display 10 of the output unit 7. Can be used to input various data. When the input terminal 9 is a touch panel, a selection instruction or data can be input using the operation input button on the display 10 or a display screen prepared separately.

  The device driver 6 receives various data such as an article attribute measurement result, article image data, and 3D scanning data, which will be described later, from an external device such as a measurement device, a digital camera, a scanner device, or a 3D scanner device connected thereto. It is a driver to receive. Then, based on the instruction signal from the arithmetic control unit 2, the driver 6 can detect, for example, image data of an article photographed by a digital camera (not shown), 3D scanning data from a 3D scanner, and a measurement result for a predetermined attribute. Execute read processing.

  The output unit 7 in this embodiment includes a display 10 and a printer 11. The output unit 7 displays various formats and data on the display 10 based on the instruction signal from the arithmetic control unit 2 or causes the printer 11 to print. In various formats, a selection screen for an article to select its article attributes, an article suitability information group selected by the information providing apparatus of this embodiment, and a plurality of article suitability information corresponding to the information group are output. An output screen is included. Examples of the display 10 include a liquid crystal monitor and a projector.

[Calculation control unit 2]
Next, functions of the arithmetic control unit 2 will be described. As shown in FIG. 2, the arithmetic control unit 2 processes the program and various data for executing the information providing method of the present invention, thereby obtaining the attribute information acquisition unit 21, the evaluation unit 22, the conforming information category. Processing of each functional unit of the specifying unit 23 and the information selection unit 24 is executed. These processes are performed according to various command signals from the arithmetic control unit 2. In addition, each said parts 21-24 attach | subject and classify | categorize according to the function for convenience of explanation, and do not limit a software structure. In addition, the present invention includes a form in which a part of the processing by these is executed by hardware mounted on the information providing apparatus of the present embodiment.

[Attribute information acquisition unit 21]
The attribute information acquisition unit 21 selects a target by selecting the object by operating the input device of the input unit 5 by the operator, or an external device (keyboard, touch panel, scanner, etc.) (not shown) via the device driver 6. It has a function of acquiring at least one article attribute measurement result (measurement value) sent.

  Specifically, for example, when the selected article attribute is a shape / size, the attribute information acquisition unit 21 acquires a predetermined dimension in each article attribute as an article attribute measurement result and sends it to the evaluation unit. When the selected article attribute is a material, the attribute information acquisition unit 21 acquires a measurement value in the selected article attribute as an article attribute measurement result and sends it to the evaluation unit 22. Further, when the selected article attribute is a color, the attribute information acquisition unit 21 acquires the color value in each selected article attribute as an article attribute measurement result and sends it to the evaluation unit 22.

  The attribute information acquisition unit 21 can also acquire image data and 3D scanning data. Therefore, the attribute information acquisition unit 21 can include an image data analysis unit 25 to analyze the acquired image data and 3D scanning data. The image data or the like may be the object itself or an object other than the object. In accordance with the selection result of the article attribute by the operator or the operator of the information providing apparatus 1 of the present invention, the image data analyzing unit 25 measures a predetermined attribute such as a dimension from the acquired image data by a conventionally known method. The result is defined as an article attribute measurement result. In the end, the information providing apparatus 1 of the present invention provides information about the object itself in the former case, and provides information about an object other than the object in the latter case. become.

  Hereinafter, the case of two-dimensional image data will be described. For example, if the selected article attribute is shape / size, the image data analysis unit 25 will process image data such as an object sent from the attribute information acquisition unit 21. A predetermined dimension related to the shape / size of the selected article attribute is obtained and used as an article attribute measurement result.

  Further, when the selected article attribute is a color, the image data analysis unit 25 measures the color value from the image data to obtain an article attribute measurement result. Color measurement can be performed by a known color measurement method. In addition, the color measurement position at the time of color measurement can be set by a known method so as to represent each of these areas. As long as a representative value can be obtained, the color measurement may be performed for only one point, or multiple points may be measured to obtain an average value thereof.

Furthermore, when the selected article attribute is a material, the image data analysis unit 25 confirms and measures the material in the selected article attribute in the image data (in this case, 3D scanning data is preferable), and this is measured. As a result. Furthermore, in this case, the image data analysis unit 25 acquires a material feeling using a known image analysis method, and uses this as an article attribute measurement result.
These article attribute measurement results are sent to the evaluation unit 22.

[Evaluation unit 22]
The evaluation unit 22 compares the article attribute measurement result sent from the attribute information acquisition unit 21 with the article attribute reference value called from the storage unit 3 to be described later, and the comparison result is a matching information category specifying unit 23 (described later). Configured to send to. For example, when the selected article attribute is a shape / size, the evaluation unit 22 calls each dimension sent from the attribute information acquisition unit 21 (including the measurement value in the image data analysis unit 25) and the storage unit 3. Contrast with the standard value of the dimension of

  Hereinafter, the evaluation contents in the evaluation unit 22 according to each of the article attributes will be specifically described. As the article attribute “shape / size”, there are three attributes of “straight (target) -curve (target)” and “small, thin, short—large, thick, long” and “familiar-conspicuous” (FIG. 4 and FIG. 4). (See FIG. 6). In addition, the article attribute “material” includes three attributes of “flat-roughness”, “light, soft, etc.—heavy, hard, etc.” and “mat-gloss”. Further, the item attribute “color” has three attributes of “blue-yellow”, “bright-dark” (lightness), and “bright-slow” (saturation).

  When there are a plurality of items attributed by the operator or the number of both items selected, the evaluation unit 22 sends the selected item attribute and the comparison result of the item attribute measurement result to the matching information category specifying unit 23, respectively.

[Compliance information category identification unit 23]
The conformity information category specifying unit 23 calls at least one predetermined tendency from each of the selection results of the selected article attributes sent from the evaluation unit 22, and sets each tendency based on the comparison result of the article attribute measurement results. It is configured to identify at least one relevant information category associated therewith and send the identification result to the information selection unit 24. When there is one item attribute measurement result from the evaluation unit 22, the matching information category specifying unit 23 calls the tendency corresponding to the item attribute from the storage unit 3. Then, based on the article attribute measurement result, one of the two matching information categories classified by the tendency is specified, and the article matching information group belonging to the matching information category specified to the information selection unit 24 Issue a command to select.

  For example, when the attribute “shape / size” of the product attribute “shape / size” is “straight line (linear) −curve (curve)” and the evaluation result in the evaluation unit 22 is “straight line”, the matching information category specifying unit 23 The cool-warm tendency is called from the storage unit 3, and the second quadrant (aqua taste) and the third quadrant (crystal taste) partitioned by the cool-warm tendency on the horizontal axis are matched on the styling map of FIG. Identify as a category. In addition, when the article attribute is the color (light and dark) and the evaluation result in the evaluation unit 22 is “bright”, the matching information category specifying unit 23 performs the styling of the light-deep tendency figure 6 from the storage unit 3. On the map, the first quadrant (bright taste) and the second quadrant (aqua taste) partitioned by the light-deep tendency on the vertical axis are specified as the matching information category. Furthermore, when the article attribute is the color (daunty-bright) and the evaluation result in the evaluation unit 22 is “daunty”, the matching information category specifying unit 23 calls the gradation-contrast tendency from the storage unit 3. On the styling map of FIG. 6, the second quadrant (aqua taste) and the fourth quadrant (artist) that include a hypothetical right-down 45 degree axis passing through the origin are specified as the matching information category.

  In addition, when the operator of the information providing apparatus of the present invention selects the article attribute as in the following (a) to (b), the conforming information category specifying unit 23 causes at least two of the three tendencies. If it is necessary to specify the relevant information category for each tendency, the relevant information categories are superimposed on the styling map shown in FIG. 6 and the relevant relevant information categories are identified. The category can be narrowed down. As a result, it becomes possible to provide a more appropriate (applicable to the operator) article conformity information group by narrowing down than those included in the conformance information category including any two quadrants in the case of one tendency. .

(A) When selecting one set of at least two attributes among one item attribute for one item For example, in FIG. 4 and FIG. 6, two attributes of the item attribute “color” are set as one set. This is the case to select. In this case, if the comparison results in the evaluation unit 22 are “yellow” and “bright”, respectively, the matching information category specifying unit 23 is the first quadrant (bright taste) and the fourth quadrant (the former matching information category). The first quadrant (bright taste) and the second quadrant (aqua taste), which are the latter category, are overlapped to narrow down to the first quadrant (bright taste) in which both overlap, and are specified as the matching information category.

(B) When selecting at least one attribute that is included in two or more item attributes for one object and that corresponds to different tendencies, for example, the attribute in the item attribute “color” in FIGS. 4 and 6 For example, in addition to “blue-yellow” and “bright-dark”, the attribute “matt (matte) -shiny” in the item attribute “material” is further selected. In this case, if the comparison results in the evaluation unit 22 are “yellow”, “bright”, and “shiny”, respectively, the matching information category specifying unit 23 is the first category (bright taste) and the second quadrant that are the former categories. (Aqua taste), 4th quadrant (Artist), and the latter category, 1st quadrant (Bright taste), 3rd quadrant (Crystal taste) are superimposed, and the 1st quadrant (Bright taste) that overlaps is compatible Is identified as a category with a relatively high conformity, and other categories are also identified as categories for which conformance cannot be denied although the conformity is relatively lower than the first quadrant.

  In this way, by selecting a plurality of article attributes so that different tendencies are included, it is possible to further narrow down the matching information category, and it is possible to provide more accurate article matching information to the operator. . In addition, if the article attribute measurement results for multiple articles belong to one or two tender bipolar attributes and cannot be narrowed down to a single conforming information category, they are ranked and matched. Information categories can be identified. In this case, the conformity information category identification unit 23 sends the plurality of identified conformance information categories to the information selection unit 24 in that order, so that the most suitable article conformity information (group), and the next conformity article conformity information (group) ), Etc., and the like can be provided to the operator with an order.

  Next, if each of the three tendencies has two or more article attribute reference values, the matching information category is specified as follows. In order to simplify the explanation, in the case of two tendencies of cool-warm tendency (horizontal axis of styling map) and light-deep tendency (vertical axis of styling map), for each of these tendencies, the relevant information category Is divided into the largest article attribute reference value or more, the second article attribute reference value or more, less than the largest article attribute reference value, the third article attribute reference value or more, the second article attribute reference value or less, and so on. . When the styling maps are shared and projected onto the styling map, they are partitioned into a grid pattern. Since multiple item attribute measurement results are included (plotted) in one of the grids, the matching information category is specified by ranking according to the number of plots in the section unit, and this specification is performed. The results are sent to the information selection unit 24 in this order. Next, when the gradation-contrast tendency is further included, since the plot of the measurement result on the oblique virtual axis on the styling map is included in one of the grid-like sections of the board, similarly, the section unit In order to identify the matching information category by ranking according to the number of plots, the identified result is sent to the information selection unit 24 in the ranking. When each of the three tendencies has two or more article attribute reference values, the article suitability information is based on the strength of the attribute on the styling map as described above, that is, the degree of attribute according to the distance from the origin. (Group) can be provided.

[Information Selection Unit 24 to Output Unit 25]
The information selection unit 24 selects at least one item suitability information belonging to the relevant information category specified based on the specification result from the match information category specifying unit 23 from the item match information group in the storage unit 3 and outputs it. The unit 25 is configured to output each selected article matching information. In the present embodiment, the image is displayed on the display 10 and printed by the printer 11. However, the output method is not limited to these, and other methods may be used.

Information Providing Method According to the Present Invention Next, an example of an information providing method using the information providing apparatus 1 according to the present invention will be described with reference to FIG. FIG. 12 is a flowchart showing an overall procedure of an information providing method using the information providing apparatus shown in FIG. Artificial intelligence (AI) can also bear at least a part of each step in the information providing method of the present invention described below.

  As shown in FIG. 12, the information providing apparatus 1 of the present invention uses the function of the arithmetic control unit 2 to display on the display 10 the article attribute information (object type / name, And the shape / size, color, and material) are displayed so that they can be selected, and the operator inputs the type / name and article attribute of the article. The article attributes may be called from the storage unit 3 in advance, displayed on the display 10, and selected from among them. Further, when the type / name of the article is limited, the type / name of the article stored in the storage unit 3 in advance may be displayed on the display 10 and selected from the display / selection. After acquiring the type / name (step S1) and the article attribute (step S2) of the article input or selected as described above, the information providing apparatus 1 of the present invention is one of the following (1) and (2). The article attribute measurement result for the object is acquired by any method (step S3). Thereafter, the attribute information acquisition unit 21 sends the selection result of the selected article attribute acquired for the object and the article attribute measurement result to the evaluation unit 22.

(1) Method for obtaining an article attribute measurement result from image data of an object or 3D scanning data After the image providing apparatus 1 according to the present invention obtains image data etc., the article attribute selected from the data is defined in advance. The article attribute measurement result is obtained by the method used. Specifically, in FIG. 1, the arithmetic control unit 2 acquires an image of an object and image data obtained by capturing the object from a predetermined angle when necessary using a digital camera (not shown) or a 3D scanner device. . The image data is stored in the storage unit 3 via the device driver 6. The image data need only be acquired and processed as digital data, and is not limited to that captured by a digital camera or a 3D scanner device. For example, even if an analog photograph obtained by capturing an object is read by a scanner. Good.

  The information providing apparatus 1 of the present invention obtains each dimension (article attribute measurement result) defined for the object from the image data, and acquires the article attribute measurement result (step S3). At this time, an article attribute measurement result for another article attribute can be directly input from the input terminal 5 such as a touch panel. Although not shown, 3D scanning data is processed in substantially the same manner as the above-described two-dimensional image data.

(2) Method by inputting operator's article attribute measurement result The operator inputs the attribute measurement result obtained for the article to the information providing apparatus 1 of the present invention, and this is used as the article attribute measurement result. For example, when the article is the T-shirt shown in FIG. 4 and has the attribute “straight-curve” in the article attribute “shape / size”, the measured value such as the width of the collar opening is directly input from the input terminal 5. This is the attribute attribute measurement result.

  Next, in order to obtain the color value as the article attribute measurement result directly from the image data of the object or directly from the object (step S3), a conventionally known method can be used. Similarly, the color value can be obtained from the 3D scanning data (step S3).

  Returning to FIG. 12, it is confirmed by an operator's input operation whether or not another attribute of the article is to be acquired (step S4). When acquiring another article attribute, returning to step S2 again, a plurality of article attributes can be selected as a result. If another article attribute is not acquired, the process proceeds to step S5.

  The evaluation unit 22 performs the following evaluation process (step S5). That is, the selection result of the selected article attribute, the article attribute measurement result, and the article attribute reference value are respectively called from the storage unit 3, and the article attribute measurement result and the article attribute reference value are compared. Thus, one or more article attribute evaluation results are obtained. The evaluation unit 22 sends the comparison result to the matching information category specifying unit 23.

  In the conformity information category specifying unit 23, when there is one item attribute selected from the evaluation unit 22, at least one item attribute reference value is used as a boundary depending on the tendency defined for the item attribute. One of the two categories related to the attribute (the matching information category including two of the four quadrants in the styling map shown in FIG. 6) is specified. Also, when two or more (preferably three or more) article attributes defined for one article are selected by the operator, or at least two (preferably belonging to different tendencies from a plurality of article attributes) If at least three attributes are selected, the matching information category specifying unit 23 selects the matching information category on the styling map shown in FIG. 6 from the tendency for each item attribute and the comparison result in the evaluation unit 22. Those overlapping and overlapping are specified as a fitness information category with high fitness (step S6). If three or more attributes are selected and it is not possible to narrow down to a single conforming information category even if the conforming information categories are overlapped in the same manner, ranking is performed according to the duplication number for each conforming information category. A matching information category is specified (step S6). When at least two item attributes are selected in this way, the matching information category with the largest number of overlapping items includes useful item matching information (group) that matches the object. Next, the article selection information is selected from the matching information category by the information selection unit 24 (step S7), and the selected article matching information is output to the output unit 7 (step S8). The read article conformity information group is sent to the output unit 7 and can be displayed on the display 10 or printed by the printer 12.

  As described above, the information providing apparatus and the information providing method according to the present invention, as described above, include a plurality of article attributes in which an article or the like (including an article attribute measurement result or a provision of image data) constitutes the article. Based on at least one item attribute measurement result relating to at least one selected item attribute selected from the above, it is possible to provide article conformity information regarding a person who suits the article, other information, advice, and the like. Other information and advice include information about items that match the image you want to wear and items that fit the impression you want to give, such as information about tableware; information about interiors such as furniture and interior (color of floors and walls, etc.); color of logos, etc. Information on graphic design; information on advertisements, promotions, product planning (package colors, etc.); information on hair and makeup, fashion and other styling; teaching materials for styling, and information that further coordinates these information. The information providing apparatus and the information providing method of the present invention having such a function can be used to match an image of the product objectively to the user of the online shop who selects and purchases one product in an online shop, for example. Or it can fully apply to the evaluation function in recommendation engines (including those using AI) such as collaborative filtering and content-based filtering for recommending more products.

  Further, according to the information providing apparatus and the information providing method of the present invention, the image of the article is measured by the analysis axis, and the object having the same attribute as that of the image is searched and determined, so that the consistency of the search of the article is improved. Can be increased or shared. In addition, when an information group showing statistically high relevance to the article attribute is obtained, the information group is classified and included in the article conformity information group for each conforming information category, and the article conformity information is provided for the article thereafter. Can be used. In addition, when an article attribute that shows a statistically high degree of association with the article conformity information group stored in the storage unit in the information providing device is confirmed, the article attribute is associated with the tendency or the conformance information category. In addition, the information can be additionally stored in the article attribute information of the storage unit in the information providing apparatus of the present invention, and can be utilized for subsequent information provision.

  In addition, this invention is not limited to the said embodiment, In an implementation stage, a component can be deform | transformed and embodied in the range which does not deviate from the summary. In addition, various inventions can be formed by appropriately combining a plurality of components disclosed in the embodiment. For example, some constituent elements can be removed from all the constituent elements shown in the embodiments, or constituent elements over different embodiments can be appropriately combined.

DESCRIPTION OF SYMBOLS 1 Information provision apparatus 2 Arithmetic control part 3 Memory | storage part 4 Internal bus 5 Input part 6 Device driver 7 Output part 9 Input terminal 10 Display 11 Printer 21 Attribute information acquisition part 22 Evaluation part 23 Compliant information category specification part 24 Information selection part 25 Image Data analysis department

According to one aspect of the present invention, the object includes at least a storage unit, an attribute information acquisition unit, an evaluation unit, a conforming information category identification unit, an information selection unit, and an output unit, and the storage unit stores an object. A plurality of article attribute information, a plurality of tendencies, a plurality of matching information categories, and a plurality of article matching information groups belonging to each of the matching information categories. Type, name, at least one article attribute defined in advance for each of the articles, and at least one article attribute reference value that serves as a reference for each article attribute. Expressed as a linear axis including the article attribute reference value in relation to the article attribute information, and the plurality of article conformity information on the linear axis with the at least one article attribute reference value as a boundary. It is defined that the group is classified into at least two compatible information categories, and the attribute information acquisition unit, as article attribute information for the object, together with its type / name and attribute name, at least selected for the object Each measurement result of one article attribute is acquired as an object attribute measurement result, and the evaluation unit acquires article attribute information of the object through the attribute information acquisition unit, and the type / name and attribute name of the object The article attribute reference value called from the storage unit and the article attribute measurement value of the object are respectively compared, and the comparison result is sent to the matching information category specifying unit, and the matching information category specifying unit is , Based on the comparison results received for the respective tendencies related to the attributes of the articles called from the storage unit Specify one overlapping matching information category from each matching information category classified according to each tendency, or if it is not possible to narrow down to one matching information category, the duplication number for each matching information category A plurality of matching information categories are identified by ranking, and the information selection unit selects at least one item matching information belonging to the specified matching information category from among the item matching information group in the storage unit, The output unit is achieved by an information providing apparatus configured to output each selected article matching information.

According to another aspect of the present invention, the object is also attribute information for acquiring at least one item attribute selected for the object and the measurement result thereof together with the type and name as the item attribute information of the object. An article attribute reference value and an attribute measurement value of the object that are received from the storage unit based on the type, name, and attribute name of the object upon receipt of the article attribute information through the obtaining step and the attribute information obtaining unit Each of the comparison results is input with respect to an evaluation step of sending each comparison result to the matching information category specifying unit and a plurality of tendencies of a plurality of items related to each item attribute information of the object called from the storage unit. Based on the comparison result, one overlapping information category is identified from the corresponding information categories classified by the respective tendencies, or 1 If it is not possible to narrow down to the conformance information category, the conformity information category identifying step of identifying a plurality of conformance information categories by ranking according to the overlap for each conformance information category, and at least belonging to the identified conformance information category includes an article adapted information selection step of selecting one article adapted information from the article adapted information group stored in the storage unit, and an output step of outputting each article adapted information the selected computer is running It is achieved by an information providing method for.

Claims (5)

Including at least a storage unit, an attribute information acquisition unit, an evaluation unit, a matching information category identification unit, an information selection unit, and an output unit,
The storage unit stores a plurality of article attribute information including the object, a plurality of tendencies, a plurality of matching information categories, and a plurality of article matching information groups belonging to each of the matching information categories,
Each of the plurality of article attribute information includes an article type / name, at least one article attribute defined in advance for each article, and at least one article attribute reference value serving as a reference for each article attribute. Including
Each of the plurality of tendencies is represented as a linear axis including the article attribute reference value in relation to the article attribute information, and the plurality of articles is delimited by the at least one article attribute reference value on the linear axis. It stipulates that the conformity information group is classified into at least two conformance information categories,
The attribute information acquisition unit acquires the measurement result of at least one item attribute selected for the object as the object attribute measurement result as the object attribute information for the object, together with the type / name and the attribute name,
The evaluation unit acquires article attribute information of the object through the attribute information acquisition unit, and the article attribute reference value and the object that are called from the storage unit based on the type / name and attribute name of the object Each of the product attribute measurement values, and send the comparison results to the relevant information category identification unit,
The conformity information category specifying unit is configured such that each conformity information classified by the respective tendencies based on the comparison results received for the respective tendencies related to the attributes of the respective articles called from the storage unit. Identifies one applicable information category that overlaps from one category, or if you cannot narrow down to one applicable information category, specify multiple applicable information categories by ranking according to the overlap for each applicable information category And
The information selection unit selects at least one item conforming information belonging to the specified conforming information category from among the item conforming information group in the storage unit,
The information providing apparatus, wherein the output unit is configured to output the selected article suitability information.   The attribute information acquisition unit further includes an image data analysis unit. The image data analysis unit acquires image data of the target object instead of the attribute measurement result of the target object or together with the attribute measurement result of the target object. The information providing apparatus according to claim 1, wherein a predetermined analysis is performed. Attribute information acquisition step for acquiring at least one item attribute selected for the object and its measurement result, together with its type and name as the item attribute information of the object,
The article attribute information is input through the attribute information acquisition unit, and the article attribute reference value called from the storage unit based on the type / name and attribute name of the object is compared with the attribute measurement value of the object. , An evaluation step of sending each comparison result to the relevant information category identification unit;
Relevant information categories classified by the respective tendencies based on the comparison results received for the three tendencies related to the respective article attribute information of the object called from the storage unit Specify one overlapping conformance information category from the list, or if it is not possible to narrow down to one conformance information category, identify multiple conformance information categories by ranking according to the overlap for each conformance information category A conformity information category identifying step, and an article conformity information selection step for selecting at least one item conformance information belonging to the identified conformity information category from among the article conformity information groups stored in the storage unit,
And an output step of outputting each selected article conformity information.   The attribute information acquisition step further includes an image data analysis step, wherein the image data analysis step is performed by acquiring image data of the target object instead of the attribute measurement result of the target object or together with the attribute measurement result of the target object. The information providing method according to claim 3, wherein the predetermined analysis is performed. The program for making a computer perform the information provision method of any one of Claims 3-4.