TW201535320A - Aging analyzing method and aging analyzing device - Google Patents

Abstract

In the present invention, an aging analysis method includes: acquiring a feature amount group for a subject, the feature amount group pertaining to a plurality of morphological characteristic types that correlate with age; using a plurality of common factors of aging extracted by factor analysis of a set of the feature amount group pertaining to a population of a plurality of people, and also using the feature amount group of the subject, to determine a factor expression pattern indicative of the circumstances of expression of that plurality of common factors of aging; and acquiring aging information on the subject on the basis of the factor expression pattern of the subject.

Description

Ageing analysis method and ageing analysis device

The present invention relates to an ageing analysis technique.

Ageing (aging) repair or ageing prevention is not only for women's concerns, but also for men's concerns, and there are various products for this purpose, such as cosmetics or food. Attention should be paid to the age of the appearance or the age of each part, and so on. Patent Document 1 proposes a method of (1) changing direction of a face shape, (2) a depression of an upper eyelid, and (3) a wrinkle of a corner of a mouth obtained from a two-dimensional facial image, ( 4) The characteristics of the nasolabial fold and (5) the shape of the lower jaw are indicators, and the age-increasing pattern of the face is identified. Further, Patent Document 2 proposes a method of calculating a weighting coefficient for a base vector of a face of a subject using a basis vector obtained by multivariate analysis of three-dimensional shape information of a face surface, based on the weighting coefficient. The degree of the tendency of the appearance of the face of the subject is obtained.

Prior technical literature Patent literature

Patent Document 1: Japanese Patent Laid-Open Publication No. 2001-331791

Patent Document 2: Japanese Patent No. 5231685

The ageing analysis method of the first aspect includes the following steps: obtaining knowledge about age a feature quantity group of a plurality of shape feature species of the test group; a plurality of common age common factors selected by factor analysis of a set of feature quantity groups of the parent group of the plural individual, and characteristics of the test subject The quantity group determines the factor expression pattern indicating the performance status of the plurality of age-related common factors of the test subject; and obtains the age-increasing information of the subject based on the factor expression pattern of the subject.

The age-increasing analysis apparatus according to the second aspect includes: an acquisition unit that acquires a feature quantity group of the subject of a plurality of shape feature types associated with age; and a determination mechanism that uses characteristics of the parent group regarding the plurality of individuals The set of the quantity group is subjected to factor analysis to select a plurality of common age common factors and the characteristic quantity group of the test subject, and the factor expression pattern indicating the performance status of the plurality of common age common factors is determined by the test subject; And an output mechanism that outputs the age-increasing information of the subject based on the factor expression pattern of the subject.

Furthermore, as another aspect of the present invention, there may be an auxiliary method for aging repair including the aging analysis method according to the first aspect, and an aging analysis method for causing at least one computer to execute the first aspect. A program, or a computer readable memory medium on which such a program is recorded. The recording medium contains non-transitory tangible media.

10‧‧‧ Ageing analysis device

11‧‧‧CPU

12‧‧‧ memory

13‧‧‧Input and output interface (I/F)

14‧‧‧Communication unit

15‧‧‧ display device

16‧‧‧ Input device

21‧‧‧Acquisition Department

22‧‧‧Decision Department

23‧‧‧Output Processing Department

30‧‧‧Analytical device

31‧‧‧Acquisition Department

32‧‧‧Decision Department

33‧‧‧Output Processing Department

34‧‧‧ Calculation Department

35‧‧‧Selection Department

37‧‧‧Analysis and Processing Department

S11~S75‧‧‧Steps

The above and other objects, features, and advantages of the invention will be apparent from the appended claims appended claims

Fig. 1 is a view showing a method of aging analysis in the first embodiment.

Fig. 2 is a view conceptually showing an example of the hardware configuration of the aging analyzer of the first embodiment.

Fig. 3 is a view conceptually showing an example of the processing configuration of the aging analyzer of the first embodiment.

Fig. 4 is a view showing an example (first pattern determining method) of an external age analysis method (analysis method) in the second embodiment.

Fig. 5 is a conceptual view showing an external age analysis device (analytical package) in the second embodiment The processing of the configuration is shown in the figure.

Fig. 6 is a view showing another example of the processing configuration of the external age measuring device (analytical device) in the second embodiment.

Fig. 7 is a view showing an example (the second pattern determining method) of the external age analysis method (analysis method) in the third embodiment.

Fig. 8 is a view showing a method of analyzing the age of the outside of the table in the variation (analysis method).

Fig. 9 is a view showing an example of a processing configuration of an external age measuring device (analytical device) in a modified example.

Fig. 10A is a view showing an example of a shape characteristic species and an ageing common factor.

Fig. 10B is a view for explaining the shape characteristic species illustrated in Fig. 10A.

Fig. 10C is a diagram showing the relationship between the factors shown in Fig. 10A and the appearance age.

Fig. 11 is a view showing another example of the shape characteristic species and the age-increasing common factor.

Fig. 12 is a view showing an example of a complex regression formula for obtaining a factor score from a feature quantity group.

Fig. 13 is a graph showing the performance of the factor score and the age-related common factor of each subject obtained from the complex regression formula shown in Fig. 12.

Fig. 14 is a graph showing the performance of the factor score and the age-related common factor of each subject obtained by factor analysis of a set of feature quantity groups including the parent group (497) of the subject.

Fig. 15 is a view showing the determination method of the factor expression state in the present embodiment.

Figure 16 is a graph showing the classification of individual samples for each age of the apparent age.

Fig. 17 is a view showing an example of classification of a factor expression pattern.

Fig. 18 is a view showing an example of ageing information based on the classification of the factor expression pattern.

Hereinafter, embodiments of the present invention will be described. Furthermore, each of the following The embodiments are exemplified, and the present invention is not limited to the configurations of the following embodiments.

[First Embodiment]

Fig. 1 is a view showing a method of aging analysis in the first embodiment. As shown in Fig. 1, the aging analysis method according to the first embodiment includes the steps of: obtaining a feature quantity group of a subject having a plurality of shape feature types associated with age (S11); using the characteristics of the subject a quantity group, and determining a factor expression indicating a performance state of the test subject by a factor analysis of a set of feature quantity groups of the parent group of the plurality of shape feature species The pattern (S13); and the age-information information of the subject is obtained based on the factor expression pattern of the subject (S15).

The above age refers to the apparent age or actual age. Each feature amount included in the feature quantity group obtained by (S11) is information on the shape of one part of the body associated with such age. That is, the feature quantity group obtained by (S11) is shape information of a plurality of parts of the body.

The "feature amount" and the "shape feature type" are used in the following manner. The "shape feature type" is information common to each body of a shape of a specific part of a human being, and the "feature amount" is a physical quantity reflecting a feature of each of the shape feature types. For example, the "shape feature" is the length under the nose, the relative size of the eye relative to the width of the cheek, and the inclination (angle) of the eye in the up and down direction, indicating that the "feature amount" of the "shape feature" is, for example, 20 mm (mm). , 0.32, 2 degrees of the measured value of each body. Such a "shape feature" is not limited to a part of a body having the shape as long as it is information related to the shape of the body. For example, the "shape feature" is a shape of a part of a face and a head (including a neck) other than the face. Further, the "shape feature" may be a shape other than the head, the back of the hand, the foot, and the like. Further, each feature amount can be expressed by various units such as the length, the angle, the curvature of the curved surface or the curved line, and the ratio. Here, the term "head" means the neck of the human body and the portion above the neck.

In (S13), the collection of feature quantity groups for the parent group of plural individuals is used. A plurality of common factors obtained by line factor analysis. This common factor is recorded as an ageing common factor. Further, the feature quantity group of each sample (each body) included in the set of the feature quantity groups corresponds to a plurality of shape feature types which are the same as the feature quantity group of the subject obtained in (S11). In (S13), the characteristic quantity group of the subject obtained in (S11) is used, and the factor expression pattern indicating the performance status of the plurality of age-related common factors is determined for the subject. For example, the factor expression pattern may indicate the presence or absence of each of the ageing common factors. However, the embodiment of the factor expression pattern is not limited. The factor expression pattern can indicate the degree of performance for each of the age-related common factors, or only the age-related common factor or the non-expressing age-related common factor.

In (S15), based on the factor expression pattern of the subject determined in (S13), the age-increasing information of the subject is obtained. The obtained age-increasing information can indicate the age impression (appearance age) of the subject, and can also indicate that the shape characteristics indicated by the feature quantity group are comprehensively related to the age of the younger age (hereinafter, there are also Recorded as the age of the shape). Here, the acquisition of the ageing information in (S15) can be performed in various aspects. For example, ageing information can be pre-allocated for each factor representation pattern. The factor expression pattern may be present in an amount equivalent to the combined number of ageing common factors. In this case, the ageing information for the factor expression pattern assignment of the subject can be obtained. Further, (S15) also includes a state in which the age-increasing information of the subject is graspable. For example, (S15), a table indicating ageing information for each factor expression pattern obtained may be provided together with a factor expression pattern of the subject determined in (S13). In this case, the person can compare the factor expression pattern of the subject with the table, and grasp the age-increasing information of the subject. Furthermore, the specific aspect of the acquisition of the ageing information in (S15) is described below.

The aging analysis method in the first embodiment can be executed in at least one computer like the aging analysis device described below. However, in the above aging analysis method, at least a part of the steps performed by a person may also be included. For example, in (S13), an intermediate process can also be performed by a computer, and based on the information calculated by the computer, only the final factor performance is performed by the person. The decision of the pattern. Further, in (S15), the factor expression pattern of the subject may be compared with the table, and the age-increasing information of the subject is obtained, and only the final age information of the subject is obtained ( Master) is performed by people. The aging analysis method according to the first embodiment is a method of repeatedly and continuously achieving the following effects, that is, using the correlation between the age and the shape characteristic of the human, and the common factor between the species and the age of the feature. The correlation between the two can easily and objectively obtain the age-increasing information of the subject without actually comparing the information of a plurality of other people or actually asking a plurality of judges. Therefore, for example, even if the steps performed by a person are included, the age-increasing analysis method in the first embodiment can be said to be a creation of a technical idea using the laws of nature as a whole.

FIG. 2 is a view conceptually showing an example of the hardware configuration of the aging analyzer 10 in the first embodiment. The aging analyzer 10 is a so-called computer, and has, for example, a CPU (Central Processing Unit) 11, a memory 12, an input/output interface (I/F) 13, and a communication unit 14 that are connected to each other by a bus bar. . The memory 12 is a RAM (Random Access Memory), a ROM (Read Only Memory), a hard disk, or the like. The communication unit 14 exchanges signals with other computers or machines. A portable recording medium or the like may be connected to the communication unit 14.

The input/output I/F 13 can be connected to a user interface device such as the display device 15 or the input device 16. The display device 15 is displayed in the same manner as an LCD (Liquid Crystal Display) or a CRT (Cathode Ray Tube) display, and is performed by a CPU 11 or a GPU (Graphics Processing Unit) (not shown). A device for processing a picture corresponding to the drawing data. The input device 16 is a device that accepts input of a user operation such as a keyboard, a mouse, or the like. The display device 15 and the input device 16 can also be integrated and implemented in the form of a touch panel. The hardware configuration of the ageing analysis device 10 is not limited.

FIG. 3 is a view conceptually showing an example of the processing configuration of the aging analysis apparatus 10 in the first embodiment. As shown in FIG. 3, the aging analysis apparatus 10 includes an acquisition unit 21 that acquires a feature quantity group of a subject having a plurality of shape feature types associated with age; The portion 22, which uses the feature quantity group of the test subject, determines the representation of the plurality of common age common factors selected by performing factor analysis on the set of the feature quantity groups of the parent group of the plurality of shape feature types. The factor expression pattern of the performance status of the tester; and the output processing unit 23 outputs the age-increasing information of the subject based on the factor expression pattern of the subject. These processing units are realized, for example, by the CPU 11 executing a program stored in the memory 12. Moreover, the program can be installed, for example, from a portable recording medium such as a CD (Compact Disc) or a memory card, or from another computer on the network via the I/F 13 or the communication unit 14, and stored in a memory. 12.

The acquisition unit 21 executes the above (S11). The acquisition unit 21 may acquire the feature quantity group of the subject by the information input by the user by operating the input device 16 based on the input screen or the like, or may be via the communication unit 14 or the input/output I from the portable recording medium or another computer. /F13 to obtain the characteristic quantity group of the subject.

Further, the acquisition unit 21 may generate the feature quantity group of the subject by itself from the information related to the subject. For example, the aging analyzer 10 may further include at least one of a contact type measuring unit and a non-contact type measuring unit (both not shown), and the obtaining unit 21 may be specific to the subject obtained by the measuring unit. The feature quantity group of the subject of the plurality of shape feature types is automatically calculated in the three-dimensional coordinate information of the part. As the contact type measuring unit, a contact type three-dimensional coordinate reading device (digitizer) is exemplified. As the non-contact type measuring unit, a three-dimensional laser scanner, a distance image sensor, or the like is exemplified. Further, the acquisition unit 21 may directly acquire the feature amount group from the two-dimensional image captured by the subject, or may use a plurality of two-dimensional images obtained by imaging the subject from different directions, using a well-known method. The three-dimensional coordinate information of the specific part of the subject is calculated, and the feature quantity group of the subject is automatically calculated using the three-dimensional coordinate information. In the calculation of the feature quantity group based on the three-dimensional coordinate information, standardization based on the same model can also be used.

In order to directly acquire the feature quantity group from the two-dimensional image, it is only necessary to use a well-known technique such as a method of utilizing the luminance distribution of the periphery of each pixel or a method of arranging the face portion. can. By using such a well-known technique, the subject quantity group of the subject can be automatically calculated approximately from the two-dimensional image. Depending on the shape feature, there may be situations where it is difficult to detect feature points. In this case, the feature point can be predicted based on the average value of each feature point automatically acquired from the parent group.

For example, in the following reference, a method is proposed in which the coordinates of a point that cannot be directly measured for an unknown individual in the feature point measurement of the skeleton shape of the face are based on the coordinates of the point at which the point can be measured around the point Inference (interpolation) is performed by the correspondence of statistical averages of a plurality of known shape measurement values. In this method, the coordinates of the morphological feature points on the soft tissue (6 points) are inferred from the coordinates of the surrounding measurable points (25 points) and the coordinates of the parent group of the sample.

References: Aoki Yoshito et al., Medical Imaging Technology, 22(5), 250-258 (2004).

As described above, in the present embodiment, the method of acquiring the feature quantity group of the subject by the acquisition unit 21 is not limited.

The determination unit 22 executes the above (S13).

The output processing unit 23 executes the above (S15), and outputs the acquired ageing information. However, the output aspect of the ageing information by the output processing unit 23 is not limited. The output processing unit 23 generates an output data indicating the ageing information, and outputs the output data to the display device 15 or another output device such as a printing device via the input/output I/F 13. Further, the output processing unit 23 may transmit the output data to another device via the communication unit 14, or may record the output data on the portable recording medium.

(Operation and effect of the first embodiment)

As in the method proposed in the above-mentioned Patent Document 1, the method of determining the age-increasing condition based on the shape change of each part of the head is used to determine the age-increasing tendency by directly using the value obtained from the shape information of each part. As a result, there may be cases where it is impossible to accurately analyze the ageing status of each individual. The reason is that the original shape characteristics of each body directly reflect In the age of growth. Furthermore, in these methods, it is impossible to determine the age impression (appearance age) of each person.

The inventors of the present invention screen a plurality of shape feature species associated with age from all of the shape feature species possessed by humans, and find a plurality of common factor factors from the feature quantity group of the selected shape feature species (age age) Common factor), verifying that each of these common factors are associated with age. Further, the inventors of the present invention investigated the relationship between the expression pattern and the age of each of the age-related common factors selected as described above, and found that the factor expression pattern exhibited the tendency to age.

According to the first embodiment, the factor expression pattern can be determined based on the characteristic quantity group of the subject, and as a result, the age-increasing information of the subject can be obtained. Here, the factor expression pattern can be determined not only according to the feature quantity of each specific part, but also according to the characteristic quantity group of the subject of the plurality of shape feature types, and thus the age information obtained from the factor expression pattern is obtained. It is not susceptible to the natural shape characteristics of each body. Further, according to the first embodiment, information on the age of the subject other than the subject can be obtained. As described above, according to the first embodiment, the degree of aging of the subject can be objectively analyzed using the feature amount of the shape of the subject.

Moreover, the inventors of the present invention classify a plurality of shape feature species for each age-age common factor with strong dependence, based on the factor load of the age-related common factor, and find that each age-age common factor is respectively The parts of the person are correspondingly matched, and the specific changes of each part have a greater impact on the age. Therefore, according to the expression pattern of the factor, the ageing information for each part of the subject can be obtained.

Hereinafter, the first embodiment will be described in detail. In the following, the external age analysis method (hereinafter, simply referred to as an analysis method) and the external age analysis device (hereinafter, simply referred to as an analysis device) in the second embodiment will be exemplified. That is, in the second embodiment below, information on the age of the appearance can be obtained as the age-increasing information. However, in the second embodiment below, the same as in the first embodiment described above. Land, obtain the information of the shape age as the ageing information. In the following description, contents that are different from the above-described first embodiment will be mainly described, and the same contents as those of the above-described first embodiment will be appropriately omitted.

[Second Embodiment]

The analysis method in the second embodiment also includes the same steps as the aging analysis method in the first embodiment shown in Fig. 1. In the second embodiment, each step will be described in further detail. Hereinafter, the analysis method in the second embodiment will be described with reference to Figs. 1 and 4 .

The feature quantity group obtained in (S11) is information on the individual of the subject who has a plurality of shape feature types associated with the appearance age. Examples of such a shape characteristic include, for example, at least one of the length of the nose and the thickness of the lip, the size of the eye, the inclination of the eye, the inclination of the eyebrow, and the bulging of the lower jaw, and particularly including at least 2 of them. Kind. The shape characteristic species are shape information of each part of the head which is selected according to the study of the inventors of the present invention and which has a large correlation with the appearance age. However, in the present embodiment, as long as there are a plurality of shape feature types associated with the appearance age, there is no limitation on the position of the body embossing the shape feature types.

Here, the plurality of shape feature types corresponding to the feature quantity group of the subject and the feature quantity group of each sample included in the parent group are a plurality of shape features from the head surface of the person in the following manner Screened out in the species. That is, the plurality of shape feature types associated with the age are obtained by complex regression analysis in which the feature quantity group related to the plurality of shape feature types is set as a description variable group and the age is set as the target variable. The screening is performed for the complex correlation coefficient of each shape feature. For example, a shape characteristic type in which the complex correlation coefficient for each shape feature species is higher than the limit value of the significant level of the sample number of the parent group of the complex regression analysis is selected. Furthermore, in the present embodiment, the age system uses the appearance age.

The plurality of common ageing common factors are obtained by screening a plurality of shapes in this manner The set of the feature quantity groups of the parent group related to the feature type is selected and analyzed by factor analysis. The number of factors used in the factor analysis can be determined, for example, based on the selected shape characteristics. Further, it is preferable that among the selected aging common factors, the correlation coefficient between the factor score (for example, average) and the age of the ageing common factor of the individual samples included in the parent group of the factor analysis is greater than that of the parent group, respectively. The 1% of the sample size is the limit of the significant level. However, the basis of the magnitude of the correlation coefficient is not limited to this reference, and can be appropriately set.

The method of determining the factor expression pattern of the subject in (S13) includes a method of using a factor score (hereinafter, referred to as a first pattern determination method), and a method of not using a factor score (hereinafter, referred to as a second pattern determination method) ). The second pattern determining method will be described in the third embodiment.

Fig. 4 is a view showing an example of the analysis method (first pattern determination method) in the second embodiment. When the first pattern determining method is used, the analysis method in the second embodiment includes (S41), (S43), and (S45) instead of (S13) in the first embodiment. . In other words, in the second embodiment, the step (S13) of determining the factor expression pattern of the subject is detailed (S41), (S43), and (S45).

(S41) The factor scores of the subjects of the plurality of common age common factors are calculated using the feature amount group of the subject obtained in (S11).

(S43) determining the performance of the age-integrating factor of the subject by comparing the threshold value of the specific position in the factor score distribution of the parent group with the factor score of the subject with respect to the plurality of common age common factors situation.

(S45) A factor expression pattern indicating at least one age-age common factor that is determined to be expressed as a performance state in (S43) is determined for the subject.

In the method of calculating the factor score of the subject in (S41), there are various methods. First, there is a method of using a complex regression analysis to infer the factor score of each age-integrating factor of the subject according to the characteristic group of the subject. Specifically, by applying the feature quantity group of the subject to each complex regression formula, this can calculate each of the subjects a factor score of the age-integrating factor, wherein each of the complex regression formulas applies a feature quantity group related to the plurality of shape feature species to the explanatory variable group by using a set of feature quantity groups for the parent group The factor score is set as the target variable and is obtained separately for the complex regression analysis of each ageing common factor. This technique is recorded as the first factor score calculation method.

In the factor analysis of the lottery common factor, the factor load amount of each ageing common factor is calculated for each shape feature corresponding to each feature amount. Further, with respect to each sample of the parent group used as a sample of the factor analysis, the factor score of each ageing common factor is calculated. The factor load of each ageing common factor as a result of the factor analysis, and the factor score of each ageing common factor for each sample of the parent group are stored together with the set of feature quantity groups in the parent group. In the computer.

In the first factor score calculation method, a common factor for each ageing common factor is obtained in advance by using a factor score for each age-integrating factor of each sample of the parent group and a complex regression analysis of the feature quantity group. The complex regression formula is further stored in a computer such as the analysis device 30 described below. Each complex regression formula is based on the individual's characteristic quantity group, which respectively describes the formula of the factor score of each individual's ageing common factor. The first factor score calculation method calculates the factor scores of the subjects of each of the plurality of age-related common factors by substituting the characteristic quantity groups of the test subjects into the complex regression formulas stored in the computer, respectively. . In this way, the first factor score calculation method can quickly calculate the factor score of the subject by using a complex regression formula obtained in advance by complex regression analysis using a set of feature quantity groups that do not include the test subject's parent group. .

Further, after the factor score calculation method of the first factor is used to calculate the factor score of the subject as described above, the feature amount group of the subject is added to the set of the feature group of the original parent group. The set of new feature quantity groups is re-applied to the complex regression analysis for each ageing common factor, and the complex regression formulas obtained from the original parent group are replaced (updated) by using the complex regression formulas obtained. In this way, the number of parent groups increases as the number of testees increases, which increases storage on computers. The accuracy of the complex regression formula.

As another method of calculating the factor score of the subject, there is also a method of adding the subject to the parent group and applying factor analysis to the new parent group. According to this method, the factor load amount of each ageing common factor is calculated in a state in which the subject quantity group of the subject is added, and the test subject who is an individual sample of the person included in the parent group may Calculate the factor score for each ageing common factor. In this method, a set of feature quantity groups (matrix A) including a new parent group of the subject, a correlation coefficient group (matrix B) between a plurality of shape feature species, and a new factor analysis can also be used. The obtained factor load group (matrix C) is obtained by multiplying the inverse matrix of the matrix A and the matrix B by the matrix C, and calculating the factor scores of the age-related common factors of the subjects included in the parent group. According to this method, the factor score of the subject can be calculated.

The threshold value indicating the specific position in the factor score distribution of the parent group used in (S43) is determined for each ageing common factor. The threshold of each ageing common factor can be set to one or multiple. For example, for the threshold, the average of the factor score distribution may be utilized, or the value obtained by adding or subtracting a value from the average value may be utilized. (S43) It is possible to determine the presence or absence of each of the age-related common factors as a performance status. For example, when the factor score of a certain ageing common factor of the subject is greater than the threshold for the ageing common factor, the subject may be determined to exhibit the ageing common factor, and when the threshold is below the threshold, It can be determined that the subject does not exhibit the age-related common factor. Further, (S43), the performance score indicating the degree of performance may be determined for each of the ageing common factors. For example, when the factor score of a certain ageing common factor of the subject is greater than the first threshold value of the ageing common factor, the performance score of the ageing common factor of the subject is set to +1. When the factor score is less than the second threshold, the performance score is set to -1, and when the factor score is equal to or less than the first threshold and the second threshold is equal to or greater than the second threshold, the performance score is determined to be zero.

In (S45), the subject is determined according to the performance status determined in (S43) Factor expression pattern. In the second embodiment, the factor expression pattern is presented in a manner indicating at least one age-age common factor determined to be expressed. In this case, the total number of patterns of the factor expression pattern and the total number of combinations of ageing common factors (the values obtained by multiplying the ageing common factor number by 2) are equal. Furthermore, as described above, there is no limitation on the appearance of the factor expression pattern.

In the second embodiment, (S15), the subject age information of the subject is obtained based on the factor expression pattern of the subject determined in (S45). Here, the appearance age information can be predetermined for each factor expression pattern. For example, by the expert's faculty assessment, the individual ages of the parent group are separately assessed. Then, by the same method as described above, the factor expression pattern is separately obtained for each sample of the parent group, and the parent group is classified by the factor expression pattern, and the average of the appearance ages is calculated for each factor expression pattern. If so, the average of the apparent ages can be prepared in advance for each factor representation pattern. In this case, (S15) obtains the average age of the outside of the factor expression pattern of the subject as the appearance age information. However, the method of obtaining the appearance of the age information is not limited to such an example.

As another example, all the factor expression patterns may be classified into a plurality of groups according to the commonality of the age-increasing information (appearance age information), and the age-related information (appearance age information) is assigned to each group. For example, all factor expression patterns can be classified into a plurality of groups based on at least one of actual age or appearance age, factor score, and factor performance number of individual samples having the same factor representation pattern.

According to the study by the present inventors, it is generally found that there is a tendency that the smaller the number of expressions of the age-related common factor, the lower the appearance age, and the higher the number of expressions of the age-related common factor, the higher the appearance age. Therefore, for example, if all the factor expression patterns are classified into a plurality of groups according to the performance number of the age-increasing common factor, the age layer of the appearance can be assigned to each group. Specifically, only one factor expression pattern of the ageing common factor is displayed, which can be classified into the youngest group of the plurality of groups, and the factors of all the ageing common factors are displayed. The expression pattern and the number of factors that are one less than the age-related common factor are the factor expression pattern of the age-related common factor, which can be classified into the oldest group of the plurality of groups.

In the case where the factor expression pattern is pre-classified as such, (S15), a group corresponding to the factor expression pattern of the subject may be selected from the plurality of groups, so that the acquired group can be assigned to the selected group. Outside the age information. However, the classification of factor expression patterns is not limited to classification based on the number of performances of age-related common factors. For example, as described above, each group obtained by using the classification based on the performance number of the ageing common factor is classified into subgroups by cluster analysis based on the factor scores of the respective samples belonging to the group. According to the further classification, according to the tendency of the factor score, the factor group that is classified into the same age layer can be grouped and grouped, so that not only the appearance of the age layer but also the tendency corresponding to the factor score can be obtained. Outside the age information.

(device configuration)

Next, the analysis device according to the second embodiment will be described focusing on contents different from the first embodiment. The same contents as those of the first embodiment are omitted as appropriate. The analysis device according to the second embodiment has the same hardware configuration as the aging analysis device 10 of the first embodiment shown in FIG. 2, and the second embodiment is executed using the hardware configuration and the following processing configuration. Analytical methods in morphology. The above-described contents of the analysis method in the second embodiment are also applied to the analysis device in the second embodiment.

FIG. 5 is a view conceptually showing an example of the processing configuration of the analysis device 30 in the second embodiment. As shown in FIG. 5, the analysis device 30 includes an acquisition unit 31, a determination unit 32, an output processing unit 33, and the like. The determination unit 32 includes a calculation unit 34. The acquisition unit 31, the determination unit 32, and the output processing unit 33 correspond to the acquisition unit 21, the determination unit 22, and the output processing unit 23 in the first embodiment, respectively. Each of the processing units is realized by the CPU 11 executing the program stored in the memory 12 in the same manner as the processing units in the first embodiment.

The acquisition unit 31 executes (S11). Further, the acquisition unit 31 may acquire a set of feature quantity groups related to the parent group and a correlation coefficient group between the plurality of shape feature types, as needed Factor load group obtained by factor analysis, factor score group of each sample included in the parent group, and the like. The analysis device 30 can maintain the information by itself, or can be obtained from a portable recording medium or other computer.

As described above, the determination unit 32 includes the calculation unit 34, and determines the factor expression pattern of the subject based on the factor score of each ageing common factor calculated by the calculation unit 34, and the calculation unit 34 The factor scores of the subjects of each of the plurality of age-related common factors are calculated using the feature amount group of the subject acquired by the acquisition unit 31. In other words, the determination unit 32 executes (S41), (S43), and (S45). The above (S41) is executed by the calculation unit 34.

As described above, the calculation unit 34 can calculate the factor score of the subject by various methods. For example, the calculation unit 34 can calculate the first factor score using the above method. In this case, the calculation unit 34 calculates the factor score of each ageing common factor of the subject by applying the characteristic quantity group of the subject to each complex regression formula, and each of the complex regression formulas is performed by Regarding the set of the feature quantity groups of the parent group, the feature quantity group related to the plurality of shape feature types is respectively set as a complex regression analysis in which the variable group is described and the factor score is set as the target variable.

The calculation unit 34 may use a complex regression formula of each of the ageing common factors held in advance, or may use the complex regression formula obtained from another device. Further, the calculation unit 34 may apply a complex regression analysis for each of the ageing common factors to the set of the feature amount groups of the parent group and the feature quantity group of the subject, and use the respective complex regression formulas obtained by the respective methods. , replacing the original complex regression formula obtained from the original parent group.

The calculation unit 34 can also calculate the factor score of each age-related common factor of the subject by adding the subject to the original parent group and performing factor analysis on the new parent group. In the factor analysis, the calculation unit 34 updates the factor load amount group, and further calculates a factor score for each ageing common factor for each sample of the new parent group (including the subject). In this case, the calculation unit 34 can also be based on the utilization acquisition mechanism. The multiplication of the obtained set of feature quantity groups, the correlation coefficient group, and the above-mentioned matrix of the factor load group is used to calculate a factor score for each age-related common factor of the subject included in the parent group.

The determining unit 32 determines the common age of the testee by comparing the threshold value of the specific position in the factor score distribution of the parent group with the factor score of the subject with respect to the plurality of common age common factors. The factor determines the factor expression pattern of at least one age-related common factor determined to be expressed for the subject. The determination unit 32 may hold the threshold value of each ageing common factor in advance, or may be obtained from another computer. The decision method regarding the specific factor expression pattern is as described above.

The output processing unit 33 executes (S15), and further outputs the acquired age information of the subject as the age-increasing information. For example, the output processing unit 33 outputs the name given to each of the plurality of ageing common factors and the ageing information indicating the factor expression pattern of the subject. The name of each of the age-related common factors can be given, for example, in such a manner that the characteristics of the common feature types of the age-increasing common factor are large. As a result, according to the age-increasing information that is output, the characteristics of the subjects whose ages are related to the appearance of the appearance may be preferred. Further, the output processing unit 33 may also include the average age of each of the factor expression patterns prepared in advance as described above in the appearance age information. The output form of the output processing unit 33 is the same as that of the output processing unit 23.

Further, the output processing unit 33 may output the ageing information given to the group selected based on the factor expression pattern of the subject. In this case, as described above, all the factor expression patterns are classified into a plurality of groups based on at least one of the age, the factor score, and the factor performance number of the respective samples having the same factor expression pattern, and for each group, in advance Give information on the age-related tendency related to the age of the appearance corresponding to the classification method. The output processing unit 33 outputs the age information related to the subject by displaying the group assignment information of the factor expression pattern and assigning the outside age information (age-age information) to each group. The group distribution information of the factor expression pattern and the age information assigned to each group can be obtained by the following selection section. 35 and pre-maintained, can also be obtained from other computers.

In the case where the factor expression pattern is classified as described above, as shown in FIG. 6, the analysis device 30 has the selection unit 35 in addition to the configuration shown in FIG. 5.

Fig. 6 is a view showing another example of the processing configuration of the analyzing device 30 in the second embodiment.

The selection unit 35 selects a factor representative corresponding to the subject from a plurality of groups obtained by classifying all factor expression patterns corresponding to all combinations of the expression states of the common age common factors according to the commonality of ageing. A group of patterns. The output processing unit 33 outputs the age information that is given to the group other than the group selected by the selection unit 35 as the age information of the subject.

(Operation and effect of the second embodiment)

As described above, in the second embodiment, the factor scores of the subjects of the plurality of age-related common factors are calculated using the feature amount group of the subject, and the determined factor score is determined based on the calculated factor score. The factor expression pattern. Here, the scores of the factors of the subjects respectively indicate the degree of expression of the corresponding age-related common factors of the subjects, and each age-related common factor has a large correlation with the apparent age, and is therefore determined according to the factor score. The factor expression pattern of the subject indicates the occurrence of some of the main reasons for the influence of the subject's external age. As a result, according to the second embodiment in which the age information of the subject is obtained based on the expression pattern of the factor, the age of the subject can be accurately analyzed.

In the first factor score calculation method which is one of the methods for calculating the factor score of the subject, the factor score of the subject can be quickly obtained by substituting the complex regression formula obtained in advance into the subject quantity group of the subject. . Moreover, the performance of the subject's age-related common factor is determined according to the comparison between the factor score of the subject and the threshold of the specific position in the factor score distribution of the parent group, and the subject is determined according to the performance status. The factor expression pattern. In this way, the factor score distribution of the parent group is added to determine the subject The performance of the age-related common factor, thereby improving the objectivity of the factor expression pattern of the subject, and further improving the objectivity and reliability of the ageing information (appearance age information) obtained from the factor expression pattern .

Further, in the second embodiment, each factor expression pattern is classified into a plurality of groups based on at least one of the age, the factor score, and the factor expression number of the respective samples having the same factor expression pattern, and the selection corresponds to the A group of factor expression patterns of the tester. Further, information on the age of the selected group is obtained as the information of the subject. In this way, it is possible to obtain the age information of the subject, such as the age level or the factor expression tendency given to each group as the appearance of the appearance age information.

[Third embodiment]

In the second embodiment described above, the factor expression pattern (first pattern determination method) of the subject is determined based on the factor score of each age-related common factor of the subject. In the third embodiment, the factor expression pattern of the subject (the second pattern determination method) is determined based on the subject quantity group of the subject without using the factor score of the subject. In the third embodiment, the contents different from the above-described respective embodiments will be mainly described, and the same content will be appropriately omitted.

The analysis method in the third embodiment also includes the same steps as the aging analysis method in the first embodiment shown in Fig. 1. In the third embodiment, each step will be described in further detail. Hereinafter, the analysis method in the third embodiment will be described with reference to Fig. 7 .

Fig. 7 is a view showing an example of the analysis method (second pattern determination method) in the third embodiment. When the second pattern determining method is used, the analysis method in the third embodiment is as shown in FIG. 7 and includes (S71), (S73), and (S75) instead of (S13) in the first embodiment. In other words, in the third embodiment, the step (S13) of determining the factor expression pattern of the subject is detailed (S71), (S73), and (S75).

(S71) A representative value of the feature quantity group for each of the plurality of factor expression patterns is obtained. In the third embodiment, individual samples of the parent group are based on individual samples. Each of the ageing common factor factor scores determines a factor expression pattern, and each of the factor expression patterns determines a representative value of the feature quantity group of each sample having the same factor expression pattern. The representative value of each factor expression pattern is a data indicating a representative position in the feature amount space (a vector having the same number of elements as the feature quantity group) determined according to the feature quantity group of each sample having the same factor expression pattern. For example, it is information indicating the center of gravity of the feature quantity groups. However, the representative value is not limited to the center of gravity.

(S73) The distance between the feature amount group of the subject obtained in (S11) and the representative value of each factor expression pattern obtained in (S71) is calculated. The representative value is a vector having the same number of elements as the feature quantity group as described above, and in (S73), the Euclid distance is calculated.

(S75) The above-described factor expression pattern of the subject is determined based on the distance between the representative value of each factor expression pattern calculated in (S73) and the subject quantity group of the subject. Specifically, the factor expression pattern indicating the minimum distance is determined as the above-described factor expression pattern of the subject.

(device configuration)

Next, the analysis device according to the third embodiment will be described focusing on contents different from the first embodiment. The same contents as those of the first embodiment are omitted as appropriate. The analysis device according to the third embodiment has the same hardware configuration as the aging analysis device 10 of the first embodiment shown in Fig. 2, and the third embodiment is executed using the above-described hardware configuration and the following processing configuration. The analytical method in the middle. The above-described contents relating to the analysis method in the third embodiment are also applied to the analysis device in the third embodiment.

The analysis device 30 in the third embodiment has the same processing configuration as that of the second embodiment shown in Fig. 5 or Fig. 6. However, each of the following processing units performs processing different from that of the second embodiment.

The acquisition unit 31 executes the above (S71) in addition to the execution (S11). The acquisition unit 31 can The representative value of the characteristic quantity group of each factor expression pattern can be obtained by itself, and can also be obtained from a portable recording medium or other computer.

The calculation unit 34 executes the above (S73).

The determination unit 32 executes the above (S75).

(Operation and effect in the third embodiment)

As described above, in the third embodiment, the distance between the feature amount group of the subject and the representative value group of each of the factor expression patterns calculated in advance is calculated, and the factor expression pattern of the subject is determined based on the distance. Therefore, according to the third embodiment, the factor expression pattern of the subject can be quickly determined without calculating the factor score of the subject, and only by the distance calculation.

[variation]

The factor analysis based on the above-described ageing common factor can be performed by the ageing analysis device 10 of the first embodiment, the analysis device 30 of the second embodiment and the third embodiment, or can be executed by another computer. Further, the parent group of the factor analysis can be updated by taking in the subjects to be analyzed by the ageing analysis device 10 and the analysis device 30 one by one. In this case, as shown in FIG. 8, (S17), (S18), and (S19) are re-executed. In addition, FIG. 8 shows an example in which a new processing procedure is added to the analysis method in the second embodiment shown in FIG. 4. However, the new processing step may be added to the first embodiment shown in FIG. The age analysis method or the analysis method in the third embodiment shown in Fig. 7 .

Fig. 8 is a view showing an analysis method in a variation. The analysis method further includes (S19), the feature quantity group of the new individual sample not included in the parent group related to the plurality of shape feature types, the feature quantity group of the new individual sample and the original mother The set of the feature quantity groups of the group performs a factor analysis using the number of the age-related common factors (S17), and uses the factor load group regained by the factor analysis of (S17) to separately calculate and calculate the original parent group. The factor score of each ageing common factor (S18) related to each individual sample of the new parent group added to the new individual sample, using the regained The factor load group and the factor scores for the individual samples of the new parent group are updated with the original information obtained from the original parent group. In Fig. 8, the feature quantity group of the subject is used as the feature quantity group of the new individual sample (S11). However, new individual samples may also not include the subject.

Fig. 9 is a view showing an example of the processing configuration of the analyzing device 30 in the modification. In this case, the analysis device 30 further includes an analysis processing unit 37 that analyzes the feature quantity group of the plurality of individual samples related to the plurality of shape feature types using a factor of the number of the age-related common factors. . The acquisition unit 31 further acquires a feature quantity group of a new individual sample that is not included in the original parent group related to the plurality of shape feature types. The analysis processing unit 37 performs factor analysis on the set of the feature quantity group of the new individual sample and the feature quantity group of the original parent group, and uses the factor load group re-acquired by the factor analysis to separately calculate and correct the original The factor score of each ageing common factor related to each individual sample of the new parent group added to the new individual sample, based on the regained factor load group and the factor scores for each sample of the new parent group. And update the original information obtained from the original parent group. The factor load group and factor score group obtained from the original parent group and the new parent group can be maintained by the analysis device 30 or by other computers. Further, the acquisition unit 31 may use the feature amount group of the subject as the feature quantity group of the new individual sample.

According to the variation, the feature quantity group of the new person (including the subject) not included in the original parent group is added to the set of the feature quantity group of the original parent group, and the factor analysis is re-executed, and the re-acquisition is utilized. The factor load group and the factor score group update the original information, so when a new person's feature quantity group is obtained, the learning can be advanced and the analysis accuracy can be improved.

Hereinafter, the above embodiments will be described in more detail by way of examples. The invention is not limited by the following examples.

Example

Fig. 10A is a view showing an example of a shape characteristic species and an ageing common factor. Fig. 10A shows an example of five common factors selected by factor analysis of the feature quantities of 13 shape feature species of 497 female Japanese groups of actual ages from 20th generation to 60th generation. In Fig. 10A, five common factors are recorded as factor 1 to factor 5. Further, 13 shape feature types are represented by numbers from 1 to 13. The lower mark y in Fig. 10A means a component in the up and down direction (the direction orthogonal to the ground). For example, "(brow-brow) y" means the drop (vertical distance) from the brow to the top and bottom of the brow.

Fig. 10B is a view for explaining the shape characteristic species illustrated in Fig. 10A.

The longitudinal width y of the eye indicates the length of the longest part of the eye in the upper and lower directions, and the area of the eye is the exposed area of the mucosa of the eye. The cheek width is different from the width of the cheek bow, and is the lower cheek of the outer corner of the line perpendicular to the ground and the surface of the face of the subject through the front view. The width of the bone position.

The degree of lifting is the angle between the inner corner of the eye (the corner of the eye) and the outer corner of the eye (the slightly curved eye) and the straight line extending horizontally from the inner corner of the eye relative to the ground. The degree of eye lifting is the angle between the outer eye and the inner corner of the eye (eye corner) and the straight line extending horizontally from the inner corner of the eye. The so-called external eye slightly means that the eye on the appearance is slightly different, for example, it is equivalent to the end of the eye corner outside the eyelid more drooping than the outer corner of the eye.

(brow-eyebrow) y is the drop from the brow to the top and bottom of the eyebrow (vertical distance), and the ray is the drop from the brow to the top and bottom of the brow (vertical distance).

The degree of fisheye indicates the inclination of the line connecting the inner corner of the eye to the outer corner of the eye in the front and rear directions. The degree of fisheye can also be said to be slightly retracted.

"tragion-lower horn point/lower horn point-颚 点 point" is the value obtained by dividing the distance from the eardrop point to the lower corner point by the distance from the lower corner point to the apex point.

The width of the lower corner point is the width between the re-protrusions near the corner point.

(Upper lip [left and right] - lower end) y / mouth split width means the aspect ratio of the lip, specifically, Indicates the value obtained by dividing the drop (vertical distance) from the upper end of the lip to the lower end of the lip by the width of the mouth. (Nose-lower point - upper lip [left and right] / lower nose - underarm) y is the length below the nose, specifically, the difference from the lower point of the nose to the upper and lower ends of the lip is divided by the nose. The value obtained from the difference from the lower point to the lower direction of the lower arm.

Although not shown in FIG. 10B, the left cheek angle is a line connecting the left ear point to the left cheek width point and the line connecting the left cheek width point and the left nose point. The minimum curvature of the left cheek is the position of the cheek bone that passes below the outer corner of the eye from the plane of the nose through the front view and the plane perpendicular to the ground and the surface of the face of the subject. The minimum curvature of the curve. In other words, the minimum curvature of the left cheek is the minimum curvature of the cheek portion of the skull facial section through the left cheek width point, the sub-nasal point, and the right cheek width point.

The inventors of the present invention selected such 13 shape feature species in the following manner. The inventors of the present invention derive the knowledge based on the analysis of the change amount of the feature points associated with the appearance age and the interpretation relationship of the feature points of the shape expert, and use the data points (feature points) of 4000 or more in the same model to derive It is presumed to be the relationship between the 60 feature points related to the apparent age (shape feature species). In this process, the shape feature species are excluded in a manner that does not have a plurality of shape features that are associated with a large number. Moreover, the inventors systematically analyzed the relationship between the feature quantity of 60 shape feature species and the apparent age, and selected the above-mentioned 13 which has a large correlation with the appearance age (the complex correlation coefficient is greater than 0.200). Shape features. As the statistical analysis, a complex regression analysis was performed in which the feature amount group related to the 60 shape feature species was set as the explanatory variable group and the appearance age was set as the target variable.

Fig. 10C is a graph showing the correlation of the factors shown in Fig. 10A with the apparent age. Further, the inventors of the present invention calculated the correlation coefficient between the factor scores of the individual samples included in the parent group and the apparent age for each common factor, and confirmed that the correlation coefficient of each common factor is as shown in FIG. 10C. A limit value greater than 1% of the sample size of the parent group. Thus, it is found that the age-related common cause related to the appearance age as shown in FIG. 10A child.

On the other hand, if the factor load of each shape feature is compared with respect to each common factor, as shown in FIG. 10A, 13 shape features can be classified according to the age-related common factor that is greatly affected. . Specifically, the shape feature types 1 to 3 (the degree of eye lifting, the degree of eye lifting and the degree of fish eye) depend to a large extent on the factor 1, and the shape feature species 4 and 5 largely depend on the factor 2, the shape feature. Species 6 and 7 depend to a large extent on factor 3, shape feature species 8 and 9 depend to a greater extent on factor 4, and shape feature species 10 and 11 depend to a greater extent on factor 5. According to this relationship, factor 1 is considered to be a factor corresponding to the inclination of the eye, factor 2 is a factor corresponding to the inclination of the eyebrow, factor 3 is a factor corresponding to the size of the eye, and factor 4 is a bulging corresponding to the lower jaw. Factor, factor 5 is a factor corresponding to the length of the nose and the thickness of the lips.

However, the number and shape characteristics of the age-increasing common factor are not limited to those shown in FIG. 10A. For example, the feature quantity group of the plurality of shape feature types associated with the appearance age may include the following feature quantities, that is, (a) a plurality of feature amounts indicating the length of the nose and the thickness of the lips, and (b) indicating the eyes. a plurality of feature quantities relative to the relative size of the cheek width, (c) a plurality of feature quantities indicating the inclination of the eye in the up and down direction and the front and rear direction, and (d) a plurality of features representing the eyelid sagging state with respect to the cheek width The amount and (e) indicate a plurality of feature amounts of the width of the lower corner of the cheek relative to the width of the cheek and the positional relationship between the tip of the beak, the ear and the lower jaw.

Fig. 11 is a view showing another example of the shape characteristic species and the age-increasing common factor. In the example of Fig. 11, an example in which five age-old common factors are selected by using 12 shape feature types different from those in Fig. 10A is shown. In the example of Fig. 11, it is also confirmed that the correlation coefficient between the age-added common factor and the apparent age is greater than the limit value of the 1% significant level of the sample number of the parent group.

Next, a method of calculating the factor score of each of the age-integrating factors of the subject by applying the characteristic quantity group of the subject to the complex regression formula for each ageing common factor will be described (see the second embodiment). Specific examples. In this specific example, the description is obtained. An example of a complex regression formula and an evaluation of the factor score of the subject obtained using the complex regression formula. Here, the shape feature type and the age-increasing common factor shown in FIG. 10A are utilized.

In this specific example, first, from the parent group of 497 Japanese women whose actual age is 20 to 60 generations, 5 persons were randomly selected for each age as the subjects, and 472 except for the testee. Factor analysis of the set of characteristic quantity groups of the tentative parent group of the person. Thereby, the factor loading amount for each of the 13 age feature types is calculated, and the factor score of each ageing common factor is calculated for each of the tentative parent group samples. Secondly, by using the factor score of each ageing common factor related to each sample of the tentative parent group and the complex regression analysis of the feature quantity group, the common factor for each ageing shown in FIG. 12 is calculated. Complex regression formula.

Fig. 12 is a view showing an example of a complex regression formula for obtaining a factor score from a feature quantity group. In Fig. 12, with respect to each ageing common factor (factor 1 to factor 5), the partial regression coefficients for each shape feature species are respectively indicated.

Fig. 13 is a graph showing the performance of the factor score and the age-related common factor of each subject obtained from the complex regression formula shown in Fig. 12. In Fig. 13, the factor scores of each of the ageing common factors of each of the 25 subjects are shown. Moreover, the performance status of each ageing common factor is represented by the performance scores of -1, 0, and +1, and the scores of the performance scores of -1 or +1 are represented by hatching. In the leftmost row of Fig. 13, the age of the subject is expressed in the form of the age (20'S, etc.), and the number of the subject is indicated in the second row from the left, and the performance score is expressed in the far right row. The number of the ageing common factor of -1 or +1.

Fig. 14 is a graph showing the performance of the factor score and the age-related common factor of each subject obtained by factor analysis of a set of feature quantity groups including the parent group (497) of the subject. In FIG. 14, the factor score obtained by factor analysis of the parent group including the subject is shown, so that the factor scores of the subjects and the ageing common factors of the subjects shown in FIG. 14 are original. The situation that should be obtained (positive solution). Therefore, when the performance status of each ageing common factor is compared using FIG. 13 and FIG. 14, FIG. 13 and FIG. In Fig. 14, only the factor 1 of the subject (309) is different in performance. The performance score of the factor 1 of the subject (309) estimated by FIG. 13 is +1, whereas the performance score of the positive solution shown in FIG. 14 becomes zero. However, since the performance of the age-related common factors of the other 24 subjects was consistent, the results confirmed that the factor score inferred by the method of FIG. 13 represents a higher accuracy of 24/25.

Further, the inventors of the present invention performed a factor analysis in which the number of common factor groups of the nine shape feature species of the 248-person parent group was set to 4, and determined the factor expression of each sample of the parent group. pattern. On the other hand, the inventors of the present invention classify the respective samples by using the factor expression pattern (16), calculate the center of gravity in the feature amount space (9-dimensional) for each factor expression pattern, and calculate the respective samples for each sample. The distance from the center of gravity of each factor expression pattern is determined, and the factor expression pattern that minimizes the distance is determined as the factor expression pattern of each sample. As a result, it was confirmed that the factor expression pattern determined using the distance was consistent with the factor expression pattern determined by factor analysis at a ratio of 81.1%.

Fig. 15 is a view showing the determination method of the factor expression state in the present embodiment. In the present embodiment, the threshold for obtaining the performance state of the factor according to the factor score is the first threshold value (+0.25σ) obtained by adding the average of the factor scores of the ageing common factor by 0.25 σ (σ is the standard deviation). And a second threshold (-0.25σ) obtained by subtracting 0.25σ from the average. Since the standard deviation σ is calculated for each of the ageing common factors, the first threshold and the second threshold are respectively set for each ageing common factor. In this embodiment, when the factor score of a certain ageing common factor of the subject is greater than the first threshold value of the ageing common factor, the performance score of the ageing common factor of the subject is set to +1. When the factor score is less than the second threshold, the performance score is set to -1, and when the factor score is equal to or less than the first threshold and the second threshold is equal to or greater than the second threshold, the performance score is determined to be zero.

Further, in order to verify the correlation between the performance status of the age-related common factor and the apparent age, the inventors and the like are based on the performance score of the age-related common factor, and are obtained by factor analysis of the parent group (497). The individual samples are classified into the following four types (S Group, N group, X group, Y group). The performance score of S group +1 is more than 2 types of performance scores of -1, the performance score of N group +1 is 1 or less, and the performance score of -1 is 1 or less, and the performance of +1 is The difference between the number of scores and the number of performance scores of -1 is one or less. The score of X group +1 is more than 2, and the performance score of -1 is more than 2, and the performance score of +1 The difference between the number of the number and the number of performance scores of -1 is one or less, and the performance score of the Y group-1 is more than two types of performance scores of +1.

The present inventors summarized the classification status of individual samples for each age of appearance age, as shown in Fig. 16, and confirmed that the performance status of the age-related common factor is correlated with the appearance age. Figure 16 is a graph showing the classification of individual samples for each age of the external age. According to Fig. 16, the higher the apparent age, the higher the ratio of the individual samples belonging to the S group, the lower the apparent age, and the higher the ratio of the individual samples belonging to the Y group. Therefore, it is proved that the performance status of the age-related common factor is related to the appearance age, and the more common age-age common factors are, the higher the appearance age.

Fig. 17 is a view showing an example of classification of a factor expression pattern. The present inventors have used the factor expression pattern of each sample determined in the example of FIG. 10A, and for each factor expression pattern, calculated the average of the ages of the individual samples having the pattern of the factor expression, by using the average The age, and the factor expression pattern is classified into three groups (denoted as area 1, area 2, and area 3). In Fig. 17, the circled number indicates the number of the ageing common factor determined to be represented, and the square surrounding at least one circled number indicates a factor expression pattern, and the number of bracketed numbers placed below the square. Represents the average age, and the number to the right of the parenthesized number indicates the number of individual samples belonging to the factor's performance pattern.

As shown in FIG. 17, it can be seen that each group corresponds to an age group of 30 generations, 40 generations, and 50 generations. Furthermore, the factor expression pattern showing only one age-age common factor is classified into the youngest group in the plurality of groups, and the factor expression pattern and the performance of all the age-related common factors (5) are displayed. The factor expression pattern of the number of all ageing common factors (4) of the ageing common factor is classified into the oldest group among the plurality of groups.

The present inventors have further clustered the factor scores of the respective samples by using the factor expression patterns belonging to the groups recorded as the regions 2 and 3, and classifying the factor expression patterns into further details. Group of. In Fig. 17, the factor representation patterns surrounded by circles represent the same group.

Fig. 18 is a view showing an example of ageing information based on the classification of the factor expression pattern. For each group generated by the classification of the expression pattern of the factors as described above, the name, the factor expression pattern, the performance number of the age-related common factor, the average of the appearance age, and the number of people are respectively indicated. The subject can be provided with information on the age of the subject by prompting the subject with the factor expression pattern determined by the subject. For example, in a subject who exhibits only one common factor of ageing, it may be suggested that the part corresponding to the age-related common factor (eye, eyebrow, chin, nose, lips, etc.) has an increase in appearance age. The main reason is the analysis result. In this way, as long as the part of the main cause of the increase in the appearance of the appearance can be specified, a cosmetic surgery for the specific part can be proposed. Further, it is also possible to present an average of the ages assigned to the outside of the corresponding group as the outside age of the subject.

Thus, the method and apparatus of each of the above embodiments can be used for the assistance of aging repair. In this case, as long as each group of the performance pattern or the factor expression pattern is classified for each factor, and the information of the corresponding cosmetic surgery is associated in advance, the age information of the subject can be provided together with the ageing information. Cosmetic surgery information.

Further, in the plurality of flowcharts used in the above description, a plurality of steps (processing) are sequentially described, but the order of execution of the steps executed in the respective embodiments is not limited to the order of the description. In each of the embodiments, the order of the steps shown in the drawings can be changed within a range that does not affect the content. Further, each of the above embodiments and each modification can be combined without departing from the scope of the content.

Some or all of the above embodiments and each of the modifications may be specified as follows. However, the above embodiments and modifications are not limited to the following description.

<1> An ageing analysis method comprising the steps of: obtaining a feature quantity group of a subject of a plurality of shape feature types associated with age; using a set of feature quantity groups for a parent group of plural individuals Factor-analyzed and selected a plurality of age-related common factors and the characteristic quantity group of the test subject, and determining a factor expression pattern indicating the performance status of the plurality of age-related common factors of the test subject; and based on the above The above-mentioned factor expression pattern of the subject obtained the age-increasing information of the subject.

<2> The method of aging analysis according to <1>, wherein the determining of the factor expression pattern comprises using the above-described feature quantity group of the subject to calculate the test subject of each of the plurality of age-related common factors The factor score is determined based on the above-described factor score of each of the above-mentioned age-integrating factors of the subject, and the factor expression pattern of the subject is determined.

<3> The method of aging analysis according to <2>, wherein the determining of the factor expression pattern comprises: by using a threshold value indicating a specific position in a factor score distribution of the parent group, and the above-mentioned subject or the parent group The factor scores of the individual samples are compared with the plurality of common age common factors, and the performance status of the age-related common factors of the testee or the individual samples is determined, and the testee or the individual sample is determined to be It is determined that the above-mentioned factor representing at least one age-integrating factor expressed in the above-described performance state represents a pattern.

<4> An ageing analysis method such as <2> or <3>, wherein The factor score of the test subject is calculated by applying the feature amount group of the test subject to each complex regression formula, and each of the complex regression formulas is applied by using the set of the feature quantity groups with respect to the parent group. The feature quantity group related to the plurality of shape feature types described above is obtained as a complex regression analysis for each of the above-described respective ageing common factors, which is a variable group and a factor score is set as a target variable.

<5> The age-age analysis method according to <4>, wherein the parent group does not include the test subject, and the age-increasing analysis method further comprises: a set of the feature quantity group regarding the parent group, and the test subject The above feature quantity group is applied to the complex regression analysis of each of the above-mentioned common ageing common factors, and each complex regression formula obtained by complex regression analysis of each of the above-mentioned ageing common factors is replaced by the original obtained by the parent group. Each complex regression formula.

<6> The ageing analysis method according to <2> or <3>, wherein the parent group includes the subject, and the factor score of the subject is calculated by using the set of the feature group regarding the parent group And a correlation coefficient group between the plurality of shape feature types and a factor load group obtained by the factor analysis, and calculating a factor score of each age-integrating factor for each sample of the parent group.

<7> The ageing analysis method according to <1>, wherein for each of the parent group, a factor expression pattern is determined according to a factor score of each of the above-mentioned respective ageing common factors of the individual sample, and the factor expression is performed The decision of the pattern includes: For each of the plurality of factor expression patterns determined by the parent group, representative values of the feature quantity groups of the individual samples having the same factor expression pattern are respectively obtained, and the above-mentioned feature quantity group of the subject and the above factors are respectively calculated. The distance of the representative value of the representation pattern is expressed, and the factor expression pattern of the subject is determined based on the distance.

<8> The age-age analysis method according to any one of <1> to <7>, wherein a correlation coefficient between a factor score and an age of each of the individual samples included in the parent group related to each of the above-mentioned common ageing factors is greater than The 1% of the sample size of the above parent group is the limit of the significant level.

<9> The age-age analysis method according to any one of <1> to <8>, wherein the obtaining of the age-increasing information includes: from the commonality of the age-related information, the performance status of the common-age common factor is Among the plurality of groups in which all the factor expression patterns of all combinations are classified, a group corresponding to the factor expression pattern of the above-mentioned subject is selected.

<10> The ageing analysis method according to <9>, wherein for each of the parent group, a factor expression pattern is determined according to a factor score of each of the above-mentioned respective ageing common factors of the individual sample, and all of the above factors The performance pattern is classified into the above plurality of groups based on at least one of an age, a factor score, and a factor performance number of respective samples having the same factor expression pattern.

<11> A method for aging analysis according to <9> or <10>, wherein the plurality of groups are formed according to an age group, and the factor showing no age-related common factor or only one age-related common factor is expressed. The pattern is classified into the youngest group of the above plurality of groups, showing the factor expression pattern of all the age-related common factors and the factor performance showing the number of age-related common factors one less than the total age-related common factor The pattern is classified into the oldest group among the above plurality of groups.

The aging analysis method according to any one of <9> to <11> wherein the acquisition of the ageing information further includes obtaining an ageing group selected for a factor expression pattern according to the subject. News.

The method of aging analysis according to any one of <1> to <12>, further comprising: obtaining a feature quantity group of a new individual sample not included in the parent group of the plurality of shape feature types; Performing a factor analysis using the number of the age-related common factors on the feature quantity group of the new individual sample and the feature quantity group of the parent group, and using the factor load group re-acquired by the above factor analysis to separately calculate The factor score of each ageing common factor of each individual sample of the new parent group obtained by adding the above-mentioned new individual sample to the parent group, based on the above-mentioned retrieved factor load group and each sample of the above-mentioned new parent group Factor score, and update the original information obtained from the above parent group.

The aging analysis method according to any one of <1> to <13> wherein the plurality of shape feature types associated with age include at least one of a length under the nose and a thickness of the lips, and an eye size At least two of the inclination of the eye, the inclination of the eyebrows, and the bulging of the lower jaw.

<15> The method of aging analysis according to any one of <1> to <14>, wherein The above-mentioned age is the appearance age, and the number of the age-related common factors is five, and the above-mentioned feature quantity group having the above-mentioned plurality of shape feature species associated with the above-mentioned external appearance age includes a plurality of features indicating the length of the nose and the thickness of the lip. a plurality of feature quantities indicating a relative size of the eye relative to the width of the cheek, a plurality of feature amounts indicating the inclination of the eye in the up and down direction and the front and rear direction, and a plurality of feature amounts indicating a state of the eyebrow sag relative to the width of the cheek, And a plurality of feature quantities indicating the magnitude of the corner width below the cheek width and the positional relationship between the tip, the ear and the lower jaw angle.

The method of aging analysis according to any one of <1> to <15, wherein the plurality of shape feature types associated with age are based on feature amounts related to the plurality of shape feature species The group is set as a complex correlation coefficient for each shape feature obtained by complex regression analysis which describes the variable group and the age is set as the target variable, and is selected from a plurality of shape feature types on the surface of the human head.

<17> An ageing analysis apparatus comprising: an acquisition unit that acquires a feature quantity group of a subject having a plurality of shape feature types associated with age; and a determination unit that uses the parent group for the plurality of individuals The set of the feature quantity group is subjected to factor analysis, and the plurality of common age common factors and the characteristic quantity group of the test subject are selected, and the factors representing the performance status of the plurality of common age common factors are determined. And an output mechanism that outputs the age-increasing information of the subject based on the factor expression pattern of the subject.

<18> An ageing analysis device such as <17>, wherein The determination means includes a calculation means for calculating a factor score of the subject of each of the plurality of age-related common factors using the feature amount group of the subject, and based on the above-mentioned The above factors of each of the aforementioned ageing common factors of the tester are scored, and the above factor expression pattern of the subject is determined.

<19> The ageing analysis device according to <18>, wherein the determining means is configured to indicate a threshold value of a specific position in the factor score distribution of the parent group and a factor score of the subject to be determined by the plurality of common age common factors respectively In comparison, the age-integrating factor expressed by the subject is determined, and the above-described factor expression pattern indicating at least one age-related common factor determined to be expressed is determined for the subject.

<20> The ageing analysis device according to <18> or <19>, wherein the calculation means applies the feature amount group of the subject to each complex regression formula, wherein each of the complex regression formulas is The set of the feature quantity groups of the parent group is applied to a complex regression group of each of the above-mentioned common age common factors, which is a feature quantity group related to the plurality of shape feature types, which is a variable group and a factor score is set as a target variable. Obtained separately by analysis.

<21> The ageing analysis device according to <20>, wherein the parent group does not include the subject, and the obtaining unit further acquires a set of the feature groups of the parent group. The calculation means applies a complex regression analysis for each of the above-mentioned respective age-age common factors to the set of the feature quantity groups of the parent group and the feature quantity group of the test subject, and uses the common age of each of the above-mentioned age groups. The complex regression formulas obtained by the complex regression analysis of the factors are replaced by the original complex regression formulas obtained from the above parent group.

<22> The ageing analysis device according to <18> or <19>, wherein the parent group includes the subject, and the acquisition unit further acquires a set of the feature amount group and the plurality of shape features of the parent group a correlation coefficient group between the species and a factor load group obtained by the factor analysis, wherein the calculation means uses the set of the feature quantity groups obtained by the acquisition means, the correlation coefficient group, and the factor load amount The group calculates the factor scores of the respective ageing common factors for each of the above-mentioned parent groups.

<23> The ageing analysis device according to <17>, wherein the individual expression samples of the respective ageing common factors of the individual samples are determined for each of the parent groups, and the factor obtaining pattern is determined And each of the plurality of factor expression patterns determined by the parent group further obtains a representative value of the feature quantity group of the individual samples having the same factor expression pattern, wherein the determining means includes a calculating means, and the calculating means calculates The distance between the feature amount group of the subject and the representative value of each of the factor expression patterns, and the factor expression pattern of the subject is determined based on the distance.

<24> The aging analysis apparatus according to any one of <17> to <23>, wherein a correlation coefficient between a factor score and an age of an individual sample included in the parent group related to each of the above-mentioned common ageing common factors is greater than that of the parent group, respectively The 1% of the sample size is the limit of the significant level.

The aging analysis apparatus according to any one of <17>, wherein the output means outputs a name given to each of the plurality of age-related common factors and a factor expression indicating the subject The above ageing information of the pattern.

The aging analysis apparatus according to any one of <17> to <25>, further comprising: a selection means for performing a performance condition common to each of the above-mentioned ageing factors according to the commonality of ageing Among the plurality of groups in which all the factor expression patterns corresponding to all combinations are classified, a group corresponding to the factor expression pattern of the above-mentioned subject is selected.

<27> The ageing analysis device according to <26>, wherein for each of the parent group, a factor expression pattern is determined according to a factor score of each of the above-mentioned respective ageing common factors of the individual sample, and all of the above factors The performance pattern is classified into the plurality of groups based on at least one of an age, a factor score, and a factor performance number of respective samples having the same factor expression pattern.

<28> The age-analyzing device according to <26> or <27>, wherein the plurality of groups are formed according to an age group, and the factor of the age-inducing common factor or the factor of only one age-age common factor is not exhibited. The pattern, which is classified into the youngest group of the above plurality of groups, exhibits a factor expression pattern of all age-related common factors and exhibits a smaller number of common age-age common factors A factor expression pattern of an ageing common factor, classified into the oldest group of the above plurality of groups.

The aging analysis apparatus according to any one of <26> to <28> wherein the output means outputs the ageing information given to the group selected based on the factor expression pattern of the subject.

The aging analysis apparatus according to any one of <17> to <29> further comprising an analysis processing means for the feature quantity group of the plurality of individual samples of the plurality of shape feature types Performing a factor analysis using the number of the age-related common factors, the obtaining means further obtaining a feature quantity group of the new individual sample not included in the parent group related to the plurality of shape feature types, wherein the analysis processing means is The feature set of the new individual sample and the set of the feature quantity group of the parent group perform the above factor analysis, and use the factor load group re-acquired by the above factor analysis to separately calculate the addition of the new individual sample to the parent group. The factor score of each ageing common factor of each individual sample of the new parent group is updated from the above parent group by the factor load group obtained above and the factor scores of the respective samples of the above new parent group. The original information.

The age-analyzing device according to any one of <17> to <30> wherein the plurality of shape feature types associated with age include at least one of a length under the nose and a thickness of the lips, and an eye size At least two of the inclination of the eye, the inclination of the eyebrows, and the bulging of the lower jaw.

<32> The age-analyzing device according to any one of <17> to <31, wherein the age is an appearance age, the number of the age-related common factors is five, and includes a plurality of thicknesses representing the length of the nose and the thickness of the lips. The feature quantity, a plurality of feature quantities indicating the relative sizes of the eyes with respect to the width of the cheek, a plurality of feature amounts indicating the inclination of the eyes in the up and down direction and the front and rear directions, and a plurality of feature amounts indicating the state of the eyebrows of the cheek width And a plurality of feature quantities indicating the magnitude of the corner width below the cheek width and the positional relationship between the tip, the ear, and the lower jaw angle.

The aging analysis device according to any one of <17> to <32> wherein the plurality of shape feature types associated with age are based on feature amounts related to the plurality of shape feature species The group is set as a complex correlation coefficient for each shape feature obtained by complex regression analysis which describes the variable group and the appearance age is set as the target variable, and is selected from a plurality of shape feature types on the surface of the human head.

<34> An auxiliary method for aging repair comprising the aging analysis method according to any one of <1> to <16>.

<35> A program that causes at least one computer to perform an ageing analysis method such as any one of <1> to <16>.

Priority is claimed on the basis of Japanese Patent Application No. 2014-033374, filed on Jan.

S11, S13, S15‧‧‧ steps

Claims (21)

An ageing analysis method comprising the steps of: obtaining a feature quantity group of a subject of a plurality of shape feature types associated with age; and performing factor analysis using a set of feature quantity groups of a parent group of plural individuals Determining a plurality of age-related common factors and the characteristic quantity group of the test subject, and determining a factor expression pattern indicating the performance status of the plurality of age-related common factors of the test subject; and based on the testee The above factors represent patterns and obtain the ageing information of the above-mentioned subjects. The aging analysis method according to claim 1, wherein the determining of the factor expression pattern comprises using the above-mentioned feature quantity group of the subject to calculate a factor score of the test subject for each of the plurality of age-related common factors And determining the factor expression pattern of the subject according to the above-mentioned factor score of each of the above-mentioned age-related common factors of the subject. The ageing analysis method of claim 2, wherein the determining of the factor expression pattern comprises: by using a threshold value indicating a specific position in the factor score distribution of the parent group, and the individual included in the test subject or the parent group The factor score of the sample is compared with the plurality of common age common factors, and the performance status of the age-related common factor of the testee or the individual sample is determined. For the testee or the individual sample, the decision is determined to be determined as The above-mentioned factor expression pattern of at least one age-integrating factor expressed by the above performance state. The ageing analysis method of claim 2, wherein the factor score of the test subject is calculated by applying the feature amount group of the test subject to each complex regression formula, wherein each of the complex regression formulas is The set of the above-mentioned feature quantity groups of the parent group is applied to the complex regression analysis of each of the above-mentioned common age common factors, which is characterized by the plurality of shape feature types, and is used to describe the variable group and set the factor score as the target variable. And separately obtained. The ageing analysis method according to claim 4, wherein the parent group does not include the subject, and the ageing analysis method further includes: a set of the feature group related to the parent group; and the feature amount of the subject a group, applying a complex regression analysis of each of the above-mentioned common ageing common factors, using each complex regression formula obtained by complex regression analysis of each of the above-mentioned ageing common factors, and replacing each of the original obtained from the parent group Complex regression formula. The ageing analysis method according to claim 2, wherein the parent group includes the subject, and the factor score of the subject is calculated by using the set of the feature groups for the parent group and the plurality of shape features The correlation coefficient group between the two and the factor load group obtained by the above factor analysis respectively calculate the factor scores of the respective ageing common factors of the respective samples of the parent group. The ageing analysis method according to claim 1, wherein for each of the parent groups, a factor expression pattern is determined according to a factor score of each of the above-mentioned respective ageing common factors of the individual sample, and the factor expression pattern is determined Included: for each of the plurality of factor representation patterns determined by the parent group, Obtaining representative values of the feature quantity groups of the individual samples having the same factor expression pattern, respectively calculating a distance between the feature quantity group of the test subject and the representative value of each of the factor expression patterns, and determining the above-mentioned The above factors of the tester showed a pattern. The ageing analysis method according to any one of claims 1 to 7, wherein a coefficient of correlation between a factor score and an age of an individual sample included in the parent group related to each of the above-mentioned ageing common factors is greater than that of the parent group 1% of the number of samples is a significant level of the limit. The ageing analysis method of claim 1, wherein the obtaining of the ageing information comprises: classifying all factor expression patterns corresponding to all combinations of the performance states of the above-mentioned ageing common factors according to the commonality of the ageing information. Among the plurality of groups obtained, a group corresponding to the factor expression pattern of the above-mentioned subject is selected. The ageing analysis method according to claim 9, wherein for each of the parent groups, a factor expression pattern is determined according to the factor scores of the respective ageing common factors of the individual samples, and all the factor representation patterns are based on having The same factor represents at least one of the age, factor score, and factor performance of each sample of the pattern, and is classified into the above plurality of groups. The ageing analysis method according to claim 9, wherein the plurality of groups are formed according to an age layer, and a factor expression pattern that does not exhibit an age-related common factor or exhibits only one age-age common factor is classified into the plurality of The group with the youngest age in the group, showing the factor expression pattern of all age-related common factors and the factor expression pattern showing the number of age-related common factors one less than the total age-related common factor is classified into The oldest group in the plural group. The ageing analysis method according to claim 9, wherein the obtaining of the ageing information further comprises obtaining ageing information given to the group selected according to the factor expression pattern of the subject. The ageing analysis method according to any one of claims 1 to 7 and 9 to 12, wherein the plurality of shape feature types associated with age include at least one of a length under the nose and a thickness of the lips, an eye size, At least two of the inclination of the eye, the inclination of the eyebrows, and the bulging of the lower jaw. The ageing analysis method according to any one of claims 1 to 7 and 9 to 12, wherein the age is an appearance age, and the number of the age-related common factors is five, and the plurality of shapes associated with the appearance of the appearance age The characteristic quantity group of the characteristic species includes a plurality of feature quantities indicating the length of the nose and the thickness of the lip, a plurality of feature amounts indicating the relative sizes of the eyes with respect to the width of the cheek, and the inclination of the eyes indicating the up and down direction and the front and rear directions. a plurality of feature quantities, a plurality of feature quantities indicating a state of the eyebrow sag relative to the width of the cheek, and a plurality of feature amounts indicating a width of the ridge angle with respect to the width of the cheek and a positional relationship between the apex, the ear and the chin angle . The ageing analysis method according to any one of claims 1 to 7 and 9 to 12, wherein the plurality of shape feature types associated with age are based on a feature quantity group related to the plurality of shape feature species The complex correlation coefficient of each shape feature obtained by the complex regression analysis in which the variable group is set and the age is set as the target variable is selected from a plurality of shape feature types on the surface of the human head. An auxiliary method for aging repair, comprising the method of aging analysis according to any one of claims 1 to 7 and 9 to 12. An ageing analysis device having: An obtaining unit that obtains a feature quantity group of a subject having a plurality of shape feature types associated with age; and a determining unit that selects a factor analysis of a set of feature quantity groups of the parent group of the plurality of individuals a plurality of age-aged common factors and the characteristic quantity group of the test subject, and determining a factor expression pattern indicating the performance status of the plurality of age-related common factors of the test subject; and an output mechanism based on the above-mentioned test The above factors represent the pattern and output the ageing information of the above-mentioned subjects. The age-increasing analysis device according to claim 17, wherein the output means outputs the name given to each of the plurality of age-related common factors and the age-increasing information indicating a factor expression pattern of the subject. The ageing analysis device according to claim 17, wherein the output means outputs the ageing information given to the group selected based on the factor expression pattern of the subject. The ageing analysis device according to any one of claims 17 to 19, further comprising: an analysis processing unit that uses the above-mentioned age-generating common to the feature quantity group of the plurality of individual samples of the plurality of shape feature types Factor analysis of the number of factors, wherein the obtaining means further obtains a feature quantity group of the new individual sample not included in the parent group related to the plurality of shape feature types, and the analysis processing means is a feature quantity of the new individual sample The set of the feature quantity groups of the group and the parent group performs the above factor analysis, and the factor load group re-acquired by the above factor analysis is used to separately calculate the new parent group obtained by adding the new individual sample to the parent group. a factor score for each ageing common factor for each individual sample, The original information obtained from the parent group is updated based on the above-obtained factor load group and the factor scores for the respective samples of the above-mentioned new parent group. A recording medium recording a program for causing at least one computer to execute the ageing analysis method according to any one of claims 1 to 7 and 9 to 12.