Computer program for hair condition analysis device for analyzing hair condition, hair condition analysis device, and hair condition analysis method

Abstract

To provide a technology for outputting information related to hair condition.SOLUTION: A hair condition analysis device for analyzing hair condition includes: an acquisition part which acquires at least one spectroscopic spectrum of human hair measured using light of at least one wavelength in a wavelength range of 700-2000 nm; a determination part which determines an index indicating a damage degree of the human hair by applying at least one of the obtained spectroscopic spectrum to a learning model machine-learned from a plurality of sample hair spectroscopic spectra measured using the wavelength range; and an output part which outputs related information related to the determined index.SELECTED DRAWING: Figure 9

Description

Technical Field

[0001] This specification relates to a technique for analyzing the state of hair.

Background Art

[0002] Patent Document 1 discloses a technique for selecting an appropriate hair dye for hair.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] This specification provides a novel technique for outputting information related to the state of hair.

Means for Solving the Problems

[0005] A first aspect disclosed by this specification is a computer program for a hair state analysis device that analyzes the state of hair. The computer program causes a computer of the hair state analysis device to function as an acquisition unit that acquires at least one spectral spectrum of human hair measured using light of at least one wavelength in a wavelength range of 700 nm or more and 2000 nm or less, a determination unit that applies the at least one spectral spectrum acquired to a learning model obtained by machine learning of each spectral spectrum of a plurality of sample hairs measured using light in the wavelength range to determine an index indicating the degree of damage of the human hair, and an output unit that outputs related information related to the determined index. According to this configuration, the hair state analysis device can appropriately analyze the state of the hair by applying the spectral spectrum of human hair to the learning model and outputting related information related to the index indicating the degree of damage of the hair.

[0006] In a second aspect disclosed by this specification, in the above first aspect, the acquisition unit may acquire a plurality of spectral spectra of the human hair measured using light of a plurality of wavelengths within the wavelength range. According to this configuration, since the hair condition analysis device acquires spectral spectra using light of a plurality of wavelengths, it can analyze the condition of the hair with higher accuracy.

[0007] In a third aspect disclosed by this specification, in the above second aspect, each of the plurality of wavelengths may be set with a predetermined wavelength interval. According to this configuration, since spectral spectra are acquired using wavelengths within the wavelength range of 700 nm or more and 1500 nm or less in a well-balanced manner, the condition of the hair can be analyzed with higher accuracy.

[0008] In a fourth aspect disclosed by this specification, in any one of the above first to third aspects, the index may include at least one of the degree of bleaching of the human hair, the degree of heat treatment performed on the human hair, and the degree of permanent wave treatment performed on the human hair. According to this configuration, the hair condition analysis device can appropriately determine an index indicating the degree of damage.

[0009] In a fifth aspect disclosed by this specification, in any one of the above first to fourth aspects, the related information may be information indicating at least one of the degree of bleaching of the human hair, the degree of heat treatment performed on the human hair, and the degree of permanent wave treatment performed on the human hair. According to this configuration, the hair condition analysis device can output useful information related to an index indicating the degree of damage.

[0010] In the sixth aspect disclosed by this specification, in any one of the first to fifth aspects, the index may include at least one of the type of dye applied to the human hair and the amount of dye applied to the human hair. According to this configuration, the hair condition analysis device can appropriately determine an index indicating the degree of damage.

[0011] In the seventh aspect disclosed by this specification, in any one of the first to sixth aspects, the related information may be information indicating at least one of the type of dye applied to the human hair and the amount of dye applied to the human hair. According to this configuration, the hair condition analysis device can output useful information related to an index indicating the degree of damage.

[0012] In the eighth aspect disclosed by this specification, in the sixth or seventh aspect, the learning model may perform machine learning by associating color information related to the colors of the plurality of sample hairs measured using light in the wavelength range of the visible light region with the respective spectral spectra of the plurality of sample hairs, the acquisition unit may further acquire color information related to the color of the human hair measured using light in the wavelength range of the visible light region, and the determination unit may apply the at least one spectral spectrum and the acquired color information that have been acquired to the learning model to determine the index. According to this configuration, the hair condition analysis device can analyze the type and amount of dye applied to the hair with higher accuracy.

[0013] A computer-readable storage medium storing the above computer program is also novel and useful. A hair condition analysis device for analyzing the condition of hair, and a hair condition analysis method for analyzing the condition of hair are also novel and useful. Further, a system including a hair condition analysis device and a measuring device for measuring spectral spectra is also novel and useful.

Brief Description of the Drawings

[0014]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Best Mode for Carrying Out the Invention

[0015] (First Embodiment) (Configuration of Communication System 2) As shown in FIG. 1, the communication system 2 includes a terminal device 10 and a measuring device 100. The terminal device 10 and the measuring device 100 are each installed at a place (such as a beauty salon) where human hair is handled. The terminal device 10 and the measuring device 100 can communicate with each other using any one of communication means such as wireless LAN, wired LAN, Bluetooth (registered trademark), and NFC (Near Field Communication, etc.).

[0016] (Configuration of Terminal Device 10) The terminal device 10 is a portable user terminal such as a tablet PC or a smartphone. In a modified example, the terminal device 10 may be a stationary user terminal such as a desktop PC. The terminal device 10 includes a communication interface 12, an operation unit 14, a display unit 16, and a control unit 30. Each of the units 12 to 30 is connected to a bus line.

[0017] The communication interface 12 is an interface for executing communication according to wireless LAN, wired LAN, Bluetooth, NFC, etc. The operation unit 14 includes buttons for receiving instructions from the user in response to being operated by the user. The display unit 16 is a display for displaying various information. The display unit 16 functions as a so-called touch panel. That is, the display unit 16 also functions as an operation unit operated by the user.

[0018] The control unit 30 includes a CPU 32 and a memory 34. The CPU 32 executes various processes according to programs 36 and 38 stored in the memory 34. The memory 34 is composed of a volatile memory, a non-volatile memory, etc. The OS program 36 is a program for realizing the basic operations of the terminal device 10. The analysis program 38 is a program for executing processes for analyzing the state of hair. The analysis program 38 may be installed in the terminal device 10 from a server (not shown) on the Internet provided by the vendor of the measuring device 100 described later, or may be installed in the terminal device 10 from a medium shipped together with the measuring device 100.

[0019] In addition to the programs 36 and 38, the memory 34 stores three learning models 40, 42, and 44. Each of the learning models 40 to 44 is a model obtained by machine learning of each spectral spectrum measured for a plurality of sample hairs, and is a model used when analyzing the state of human hair. Each of the learning models 40 to 44 is installed in the terminal device 10 together with the analysis program 38. Although the memory 34 is illustrated as storing a learning model 46, this will be described in the second embodiment described later.

[0020] (Configuration of the measuring device 100) The measuring device 100 is a portable device for measuring the spectral spectrum of human hair. The measuring device 100 is manufactured and sold, for example, by a manufacturer that manufactures and sells hair dyes, hair bleaches, permanent wave agents, etc. The measuring device 100 includes a communication interface 112, an operation unit 114, a display unit 116, a light source 118, a spectroscopic unit 120, a detection unit 122, and a control unit 130. Each of the units 112 to 130 is connected to a bus line.

[0021] The communication interface 112 is an interface for performing communication according to wireless LAN, wired LAN, Bluetooth, NFC, etc. The operation unit 114 includes buttons for receiving instructions from the user in response to being operated by the user. The display unit 116 is a display for displaying various information. The light source 118 outputs light for measuring the spectral spectrum of human hair. The light source 118 can output light in a wavelength range of 700 nm or more and 2000 nm or less. The spectroscopic unit 120 splits the light output from the light source 118 into a plurality of wavelengths. In this embodiment, the spectroscopic unit 120 splits the light in the wavelength range of 700 nm or more and 2000 nm or less at 5 nm intervals. As the spectroscopic unit 120, for example, a diffraction grating, a prism, etc. can be used. The detection unit 122 receives the reflected light from the hair for each wavelength and detects the intensity of the reflected light for each wavelength.

[0022] The control unit 130 includes a CPU 132 and a memory 134. The CPU 132 executes various processes according to the programs 136 and 138 stored in the memory 134. The memory 134 is composed of a volatile memory, a non-volatile memory, etc. The OS program 136 is a program for realizing the basic operations of the measuring device 100. The measurement program 138 is a program for measuring the spectral spectrum of human hair. By executing the measurement program 138, the CPU 132 calculates the reflectance of the reflected light from the hair based on the intensity of the light detected by the detection unit 122, and obtains a spectral spectrum showing the relationship between the reflectance and the wavelength.

[0023] Note that the measuring device 100 is illustrated as including a light source 119 and a detection unit 123, and a measurement program 139 is stored in the memory 134, but these will be described in the second embodiment below.

[0024] Here, the spectral spectrum obtained from hair will be described. FIG. 2 shows an example of the spectral spectra of three types of hair with different degrees of decolorization. "Sh" indicates the hair treated once with a Vigent treatment shampoo (manufactured by Hoyu Co., Ltd.), "LT" indicates the hair treated once with a treatment agent obtained by mixing Promaster LT (a first agent of a hair bleaching agent manufactured by Hoyu Co., Ltd.) and PRO-OXIDE 6% (a second agent of a hair bleaching agent manufactured by Hoyu Co., Ltd.) at a ratio of 1:2, and "PW2" indicates the hair treated twice with a Beauty Base Up Bleach (a hair bleaching agent manufactured by Hoyu Co., Ltd.). As shown in FIG. 2, it can be seen that in the wavelength range of 700 nm or more and 1500 nm or less, the spectral spectrum varies greatly depending on the degree of decolorization of the hair. That is, by measuring the spectral spectrum of the hair of the subject, the degree of decolorization of the hair can be analyzed.

[0025] However, when the hair is in a dyed state, since the dye absorbs light, the obtained spectral spectrum changes. FIGS. 3 to 6 show examples of the respective spectral spectra of the hair dyed after treatment with Sh, LT, and PW2. FIG. 3 shows the spectral spectrum of the hair dyed with an oxidative dye (toluene-2,5-diamine; pTD) after treatment with LT. FIG. 4 shows the respective spectral spectra of the hair dyed with each of two acid dyes (Red 102, Violet 401) after treatment with LT. FIG. 5 shows the respective spectral spectra of the hair dyed with an oxidative dye (toluene-2,5-diamine + m-aminophenol; pTD + mAP) after treatment with Sh, LT, and PW2, respectively. FIG. 6 shows the respective spectral spectra of the hair dyed with an acid dye (Orange 205) after treatment with Sh, LT, and PW2, respectively. As shown in FIGS. 3 to 6, it can be seen that in any case, the obtained spectral spectrum changes due to the absorption of light by the dye.

[0026] In addition, when heat treatment (e.g., styling using a hair iron) is performed on the hair or when a permanent wave treatment is performed on the hair, the hair is damaged, and thus the acquired spectral spectrum changes. FIG. 7 also shows an example of each spectral spectrum of hair (HT1) that has been heat-treated at 180° C. for 10 minutes using a straight hair iron (A’s styleEX) in addition to the treatments by Sh, LT, and PW2. FIG. 8 also shows an example of each spectral spectrum of hair (PM3) on which a permanent wave treatment has been performed in addition to the treatments by Sh, LT, and PW2. As shown in FIGS. 7 and 8, it can be seen that the acquired spectral spectrum changes due to the execution of the heat treatment and the execution of the permanent wave treatment.

[0027] As described above, it has been found that the spectral spectrum obtained changes due to hair dyeing, the execution of heat treatment, or the execution of permanent wave treatment, but in any case, the change is not so large. From the above, the inventors have found that by using each learning model 40, 42, 44 obtained by machine learning the spectral spectra of a plurality of sample hairs on which these treatments have been performed, the state of the hair can be accurately analyzed.

[0028] The first learning model 40 is a model for determining the degree of hair bleaching for human hair. The first learning model 40 is created by machine learning each spectral spectrum measured at 5 nm intervals using light in the wavelength range of 700 nm to 2000 nm for a plurality of sample hairs dyed after being treated by any one of Sh, LT, and PW2. By using the first learning model 40, the degree of hair bleaching can be analyzed with high accuracy.

[0029] The second learning model 42 is a model for determining the degree of heat treatment performed on human hair. The second learning model 42 is created by machine learning the respective spectral spectra measured at 5 nm intervals using light in the wavelength range of 700 nm to 2000 nm for a plurality of sample hairs subjected to heat treatment A or heat treatment B, and a plurality of sample hairs not subjected to heat treatment (untreated). In heat treatment A, the sample hair is treated at 180° C. for 30 minutes using a constant temperature dryer (DVS), and in heat treatment B, the sample hair is treated at 200° C. for 40 minutes using a constant temperature dryer (DVS). By using the second learning model 42, the degree of heat treatment performed on the hair can be analyzed with high accuracy.

[0030] The third learning model 44 is a model for determining the degree of permanent wave treatment (hereinafter referred to as perm treatment) performed on human hair. The third learning model 44 is created by machine learning the respective spectral spectra measured at 5 nm intervals using light in the wavelength range of 700 nm to 2000 nm for a plurality of sample hairs subjected to perm treatment A or perm treatment B, and a plurality of sample hairs not subjected to perm treatment (untreated). In perm treatment A, the sample hair is treated once with Biwave 30 (a perm treatment agent manufactured by Hohyu Co., Ltd.), and in perm treatment B, the sample hair is treated three times with Biwave 30 (a perm treatment agent manufactured by Hohyu Co., Ltd.). By using the third learning model 44, the degree of permanent wave treatment performed on the hair can be analyzed with high accuracy.

[0031] In this embodiment, each of the learning models 40, 42, 44 is created by machine learning the spectral spectra measured for 600 or more sample hairs. In this embodiment, the terminal device 10 executes the hair condition analysis process described below using these learning models 40, 42, 44.

[0032] (Hair Condition Analysis Process; FIG. 9) Referring to FIG. 9, the content of the hair condition analysis process executed by the CPU 32 of the terminal device 10 according to the program 38 will be described.

[0033] In S10, the CPU 32 of the terminal device 10 monitors for receiving the spectral spectrum obtained by measuring the hair of the subject from the measuring device 100. The measuring device 100 measures the spectral spectrum of the hair of a subject who is a customer at, for example, a beauty salon, and transmits the spectral spectrum to the terminal device 10 via the communication interface 112. In this case, the CPU 32 receives the spectral spectrum (YES in S10) and proceeds to S12.

[0034] In S12, the CPU 32 determines whether or not it has received a user operation instructing the execution of the analysis process of the received spectral spectrum. When the CPU 32 receives the user operation (YES in S12), it proceeds to S14.

[0035] In S14, the CPU 32 applies the received spectral spectrum to the first learning model 40 to determine the degree of hair discoloration. In this embodiment, the CPU 32 determines whether the hair has been processed by any of Sh, LT, and PW2. FIG. 10 shows the results of analyzing the hair of a plurality of humans using the first learning model 40 in actual use. As is clear from FIG. 10, the coincidence rate between the degree of discoloration (predicted class) determined using the first learning model 40 and the actual degree of discoloration (true class) is 97.3%. By using the first learning model 40, the degree of hair discoloration can be analyzed with high accuracy. Note that the hair analyzed in FIG. 10 includes hair in a dyed state. The same applies to FIGS. 11 and 12 described later.

[0036] In S16, the CPU 32 applies the received spectral spectrum to the second learning model 42 to determine the degree of heat treatment performed on the hair. In this embodiment, the CPU 32 determines whether the hair has undergone heat treatment A, heat treatment B, or no heat treatment. FIG. 11 shows the results of analyzing a plurality of spectral spectra obtained by actually measuring human hair using the second learning model 42. More specifically, FIG. 11 shows the results obtained by using the second learning model 42 after determining the degree of decolorization using the first learning model 40. As is clear from FIG. 11, the coincidence rate between the degree of heat treatment (predicted class) determined using the second learning model 42 and the degree of heat treatment actually performed (true class) is 80% or more in any of the Sh system, LT system, and PW system. By using the second learning model 42, the degree of heat treatment performed on the hair can be analyzed with high accuracy.

[0037] In S18, the CPU 32 applies the received spectral spectrum to the third learning model 44 to determine the degree of perm treatment performed on the hair. In this embodiment, the CPU 32 determines whether the hair has undergone perm treatment A, perm treatment B, or no permanent wave treatment. FIG. 12 shows the results of analyzing a plurality of spectral spectra obtained by actually measuring human hair using the third learning model 44. More specifically, FIG. 12 shows the results obtained by using the third learning model 44 after determining the degree of decolorization using the first learning model 40. As is clear from FIG. 12, the coincidence rate between the degree of perm treatment (predicted class) determined using the third learning model 44 and the degree of perm treatment actually performed (true class) is 80% or more in any of the Sh system, LT system, and PW system. By using the third learning model 44, the degree of perm treatment performed on the hair can be analyzed with high accuracy.

[0038] In S20, the CPU 32 causes the display unit 16 to display the information determined in S14 to S18. Although it is an example of the display content on the display unit 16, for example, as shown in FIG. 9, the CPU 32 causes the display unit 16 to display information indicating that the degree of decolorization "LT", the degree of heat treatment "heat treatment B", and the degree of perm treatment "no treatment". By looking at the information displayed on the display unit 16, the user can grasp the state of the hair to be measured.

[0039] Even if the hair looks the same in color, the actual degree of damage may vary depending on the degree of treatment applied to the hair, such as decolorization and decoloring. For this reason, for example, even if a beautician predicts the finished color of the hair by a treatment using a hair cosmetic based on the apparent color of the customer's hair, there is a possibility that the expected finished color may not be achieved because the degree of hair damage is different. However, as described above, in this embodiment, the terminal device 10 can appropriately analyze the degree of hair damage (that is, the degree of decolorization, the degree of heat treatment, the degree of perm treatment). Therefore, the beautician can select a hair cosmetic based on the degree of hair damage indicated by the analysis result, so that the finished color of the hair can be made closer to the assumed color. Also, when new treatment should be performed on damaged hair, there is a concern about the accumulation of damage. In this embodiment, since the beautician can accurately grasp the degree of hair damage, an appropriate treatment can be selected according to the degree of damage.

[0040] (Corresponding relationship) The analysis program 38 and the terminal device 10 are examples of "program" and "hair state analysis device", respectively. Each of Sh, LT, PW2, heat treatment A, heat treatment B, perm treatment A, perm treatment B 3 times, and no treatment is an example of "index" and "related information". The 5 nm interval is an example of "predetermined wavelength interval".

[0041] The process of S10 in FIG. 9 is an example of a process executed by the "acquisition unit". The processes of S14, S16, and S18 are examples of processes executed by the "determination unit". The process of S20 is an example of a process executed by the "output unit".

[0042] (Second Embodiment) As shown in FIG. 1, in the second embodiment, the measuring device 100 further includes a light source 119 and a detection unit 123. Further, the memory 134 of the control unit 130 stores a measurement program 139.

[0043] The light source 119 can output light in a wavelength range different from that of the light source 118, specifically, in the wavelength range of the visible light region (for example, 380 nm or more and 780 nm or less). The detection unit 123 detects the reflected light from the hair of the light output from the light source 119. The measurement program 139 is a program for measuring the coordinate values in the Lab color space of human hair. The control unit 130 converts the light detected by the detection unit 123 into the Lab color space by executing the measurement program 139, and acquires the coordinate values in the Lab color space of human hair.

[0044] As shown in FIG. 1, in the second embodiment, the memory 34 of the terminal device 10 stores a learning model 46. The learning model 46 is installed in the terminal device 10 together with the analysis program 38.

[0045] The learning model 46 is a model for determining the type and amount of dye applied to human hair. The learning model 46 is created by machine learning the spectral spectra measured at 5 nm intervals using light in the wavelength range of 700 nm to 2000 nm for a plurality of sample hairs treated with a hair dye containing various dyes and a plurality of sample hairs not treated with a hair dye after treatment with any of Sh, LT, PW2. FIGS. 13 to 15 together show an example of each spectral spectrum of hair treated with a hair dye containing a predetermined amount of various dyes after treatment with any of Sh, LT, PW2. FIG. 13 shows the spectral spectrum of hair treated with a hair dye containing a predetermined amount of p-phenylenediamine (pPD) (FIG. 13(a)) or a predetermined amount of pTD (FIG. 13(b)) after treatment with Sh. FIG. 14 shows the spectral spectrum of hair treated with a hair dye containing a predetermined amount of pPD (FIG. 14(a)) or a predetermined amount of pTD (FIG. 14(b)) after treatment with LT. FIG. 15 shows the spectral spectrum of hair treated with a hair dye containing a predetermined amount of pPD (FIG. 15(a)) or a predetermined amount of pTD (FIG. 15(b)) after treatment with PW2.

[0046] In addition, the coordinate values in the Lab color space measured using light in the wavelength range of the visible light region for each sample hair are associated with each spectral spectrum used for machine learning. FIG. 16 shows an example of the learning data used for creating the learning model 46. For example, sample #1 is a sample hair treated with a hair dye containing 2.00% pTD as a dye after being treated with Sh. The coordinate values L = 17.06, a = 1.02, b = 0.16 in the Lab color space are associated with the spectral data of the sample hair. The learning model 46 is created by machine learning these learning data.

[0047] As described in the first embodiment, when the hair is in a dyed state, the acquired spectral spectrum changes because the dye absorbs light. However, the present inventors have found that by using the learning model 46 described above, the type and amount of the dye applied to the hair can be analyzed with high accuracy. In this embodiment, the learning model 46 is created by machine learning the learning data obtained by measuring 600 or more sample hairs. The terminal device 10 executes the hair state analysis process described below using the learning model 46.

[0048] (Hair state analysis process; Fig. 9) In S10 of Fig. 9, the CPU 32 of the terminal device 10 monitors receiving the spectral spectrum of the subject's hair measured by the measuring device 100 and the coordinate values in the Lab color space of the hair. The measuring device 100 measures the spectral spectrum of the hair of a subject who is a customer at a beauty salon or the like and the coordinate values in the Lab color space, and transmits the spectral spectrum and the coordinate values to the terminal device 10 via the communication interface 112. In this case, the CPU 32 receives the spectral spectrum and the coordinate values (YES in S10) and proceeds to S12. S12 is the same as in the first embodiment.

[0049] In S114, the CPU 32 applies the received spectral spectrum and coordinate values to the learning model 46 to determine the type and amount of the dye applied to the hair. In this embodiment, the CPU 32 determines whether the hair has been treated with any of Sh, LT, and PW2, and with which type of hair dye containing what amount of dye. Fig. 17 shows the results of analyzing the hairs of a plurality of humans using the learning model 46 actually. As is clear from Fig. 17, the coincidence rate between the type and amount of the dye (predicted class) determined using the learning model 46 and the type and amount of the dye actually applied (true class) is 92.1%. By using the learning model 46, the type and amount of the dye applied to the hair can be analyzed with high accuracy.

[0050] In S20, the CPU 32 causes the display unit 16 to display the information determined in S114. Although it is an example of the display content on the display unit 16, for example, as shown in FIG. 9, the CPU 32 causes the display unit 16 to display information indicating that the degree of decolorization "Sh", the type of dye "pPD", and the amount of dye "1.00%". By viewing the information displayed on the display unit 16, the user can grasp the state of the hair to be measured.

[0051] (Corresponding relationship) pTD and pPD are examples of "type of dye", "index", and "related information". 2.00%, 1.00%, etc. are examples of "amount of dye", "index", and "related information". The coordinate values "L, a, b" in the Lab color space are an example of "color information". The process of S114 in FIG. 9 is an example of the process executed by the "determination unit".

[0052] The specific examples of the present invention have been described in detail above, but these are merely examples and do not limit the scope of the claims. The technology described in the claims includes various modifications and changes of the specific examples illustrated above. The modifications of the above embodiments are listed below.

[0053] (Modification 1) The process of FIG. 9 above does not have to be realized by the terminal device 10. For example, it may be realized by a server installed by a manufacturer that manufactures and sells hair coloring agents, hair bleaching agents, permanent wave agents, etc. In this case, instead of S20 in FIG. 9, the server transmits each information determined in S14 to S18 and S114 to the measuring device 100. Then, each information is displayed on the display unit 116 of the measuring device 100. In this modification, the server is an example of the "hair state analysis device". In another modification, the process of FIG. 9 may be executed by the measuring device 100. In this case, the measuring device 100 executes S14 to S18 and S114 using the measured spectral spectrum without executing S10, and in S20, causes each information to be displayed on the display unit 116. In this modification, the measuring device 100 is an example of the "hair state analysis device".

[0054] (Modification Example 2) In the above-described first embodiment, in the hair condition analysis process, three indicators were determined: the degree of bleaching, the temperature of the heat treatment, and the number of times of using a permanent wave agent. However, in the hair condition analysis process, any one or two of these indicators may be determined. For example, in FIG. 9, the CPU 32 may execute only one of the processes S14, S16, and S18, or may execute the processes of S14 and S16, S14 and S18, or S16 and S18. That is, the processes of S14 to S18 may be executed in any combination.

[0055] (Modification Example 3) In each of the above-described embodiments, as the degree of hair damage, the degree of bleaching, the degree of heat treatment, the degree of perming, the type of dye, and the amount of dye were exemplified and described. However, the types of the degree of hair damage are not limited to these. The degree of hair damage may include various chemical damages and physical damages. Chemical damage may be, for example, damage caused by an oxidation reaction or a reduction reaction on the hair. Damage caused by an oxidation reaction may be, for example, bleaching, hair color, etc., and damage caused by a reduction reaction may be, for example, a perm. Physical damage may be, for example, frictional damage caused by combing or UV damage caused by ultraviolet rays. Further, as an index of the degree of hair damage, a gloss value indicating the degree of gloss of the hair may be determined. According to the technology disclosed in this specification, by applying the spectral spectrum of the hair to the learning model, the degrees of the various damages described above can be analyzed.

[0056] (Modification Example 4) In S20 of FIG. 9, instead of causing the display unit 16 to display "LT", "Heat Treatment A", "No Treatment", etc. shown in FIG. 9 as information indicating the degree of hair damage, for example, the terminal device 10 may cause the display unit 16 to display information obtained by quantifying the degree of damage based on the indexes determined in S14 to S18. For example, in each of S14, S16, and S18, the terminal device 10 may evaluate the degree of damage to the determined index with three levels of numerical values (for example, 1 to 3), and in S20, may display the evaluation values corresponding to the respective indexes, or may display the value obtained by summing up the respective evaluation values. Specifically, the index "Sh" with a low degree of damage may be set as the evaluation value 1, the index "LT" with a medium degree of damage may be set as the evaluation value 2, and the index "PW2" with a high degree of damage may be set as the evaluation value 3. The same applies to the degree of heat treatment and the degree of permanent wave treatment for the hair. Note that these evaluation values may be further subdivided into more levels instead of three levels. In this modification example, each of the evaluation values corresponding to the respective indexes, or the value obtained by summing up the respective evaluation values, is an example of "related information". Also, for example, the terminal device 10 may cause the display unit 16 to display the recommended brand name of hair cosmetics, the recommended hairstyle, etc. according to the degree of damage based on the indexes determined in S14 to S18. In this modification example, the recommended brand name of hair cosmetics and the recommended hairstyle are examples of "related information". Generally speaking, the "index" and the "related information" do not necessarily have to match. Also, instead of causing the display unit 16 to display the type of hair treatment agent (that is, Sh, LT, PW2) as information indicating the degree of bleaching, for example, the terminal device 10 may cause the display unit 16 to display the lightness of the hair (so-called under level). The same applies to the second embodiment.

[0057] (Modification Example 5) In the above-described second embodiment, pPD and pTD were described as examples of dyes, but the types of dyes to be analyzed are not limited to these. In the technology disclosed in this specification, for example, other oxidative dyes such as m-aminophenol, 2,4-diaminophenoxyethanol hydrochloride, 5-amino-o-cresol, resorcinol, 4-nitro-o-phenylenediamine, α-naphthol, N,N'-bis(2-hydroxyethyl)-2-nitro-p-phenylenediamine, acid dyes (direct dyes) such as Red 102, Violet 401, Orange 205, Red 227, Yellow 203, Blue 1, Black 401, Brown 16, HC Blue 2, Red 51, Blue 75, Violet 2, or combinations thereof can accurately analyze the type and amount of the dye.

[0058] (Modification Example 6) In each of the above embodiments, the spectral spectrum of hair measured at 5 nm intervals using light in the wavelength range of 700 nm or more and 2000 nm or less was analyzed. However, in the measurement of the spectral spectrum of hair, light in a partial wavelength band within the above wavelength range (for example, 800 nm or more and 1500 nm or less, or for example, 900 nm or more and 1200 nm or less, or for example, 1000 nm or more and 1100 nm or less) may be used, or only light of a single wavelength may be used. Also, the wavelength interval when using light of a plurality of wavelengths is not limited to 5 nm. The wavelength interval may be a value less than 5 nm or a value greater than 5 nm. Here, increasing the wavelength interval can increase the processing speed of the terminal device 10 and the measuring device 100, but if the wavelength interval becomes too large, the accuracy of the analysis may decrease. Therefore, the wavelength interval is preferably less than 200 nm, and more preferably less than 100 nm. In another modification example, the wavelength interval may not be uniform. In the measurement of the spectral spectrum of hair, when using only light of a single wavelength, for example, as the light source 118, one that outputs light of the single wavelength may be used, or an optical filter that allows only a specific wavelength to pass through may be used as the spectroscopic unit 120.

[0059] (Modification Example 7) In the above-described Second Embodiment, the coordinate values in the Lab color space are used as the hair color information. However, instead of this, or in addition to this, the coordinate values in other color spaces (for example, RGB color space, HSV color space, XYZ color space, etc.) may be used.

[0060] (Modification Example 8) In each of the above-described embodiments, one learning model is used to analyze one type of damage degree. However, a plurality of learning models may be used to analyze one type of damage degree, or a single learning model may be used to analyze a plurality of types of damage degrees.

[0061] (Modification Example 9) In each of the above-described embodiments, each learning model is created by machine learning the normal spectral spectra of a plurality of sample hairs. However, instead of this, or in addition to this, for example, each learning model may be created by machine learning a differential spectrum obtained by differentiating the spectral spectrum, a logarithmic spectrum obtained by logarithmically transforming the spectral spectrum, or the like. According to this configuration, the damage degree of the hair can be analyzed more accurately.

[0062] (Modification Example 10) In each of the above-described embodiments, each process in FIG. 9 is realized by the CPU 32 executing the analysis program 38. Instead of this, any one of the processes in FIG. 9 may be realized by hardware such as a logic circuit.

[0063] Moreover, the technical elements described in this specification or the drawings exhibit technical utility alone or in various combinations, and are not limited to the combinations described in the claims at the time of filing. In addition, the technology exemplified in this specification or the drawings achieves a plurality of purposes simultaneously, and has technical utility by achieving one of those purposes itself.

Description of Reference Numerals

[0064] 2: Communication system, 6: Internet, 10: Terminal device, 12: Communication interface, 14: Operation unit, 16: Display unit, 30: Control unit, 32: CPU, 34: Memory, 36: OS program, 38: Analysis program, 40, 42, 44, 46: Learning model, 100: Measuring device, 112: Communication interface, 114: Operation unit, 116: Display unit, 118, 119: Light source, 120: Spectral analysis unit, 122, 123: Detection unit, 130: Control unit, 132: CPU, 134: Memory, 136: OS program, 138, 139: Measurement program

Claims

1. A computer program for a hair condition analysis device that analyzes the condition of hair, causing a computer of the hair condition analysis device to function as each of the following parts, namely, an acquisition unit that acquires at least one spectral spectrum of human hair measured using light of at least one wavelength in a wavelength range of 700 nm or more and 2000 nm or less; a determination unit that applies the at least one spectral spectrum acquired to a learning model obtained by machine learning the spectral spectra of a plurality of sample hairs measured using light in the wavelength range, and determines an index indicating the degree of damage of the human hair; an output unit that outputs related information related to the determined index; A computer program.

2. The computer program according to claim 1, wherein the acquisition unit acquires a plurality of spectral spectra of the human hair measured using light of a plurality of wavelengths in the wavelength range.

3. The computer program according to claim 2, wherein each of the plurality of wavelengths is set with a predetermined wavelength interval.

4. The index includes the degree of bleaching of the human hair, the degree of heat treatment performed on the human hair, the degree of permanent wave treatment performed on the human hair, at least one of them. The computer program according to claim 1.

5. The related information is the degree of bleaching of the human hair, the degree of heat treatment performed on the human hair, the degree of permanent wave treatment performed on the human hair, Information indicating at least one of them. The computer program according to claim 1.

6. The index includes the type of dye applied to the human hair, the amount of dye applied to the human hair, at least one of them. The computer program according to claim 1.

7. The related information is the type of dye applied to the human hair, the amount of dye applied to the human hair, Information indicating at least one of them. The computer program according to claim 1.

8. The learning model has performed machine learning by associating color information related to the colors of the plurality of sample hairs measured using light in the wavelength range of the visible light region with the respective spectral spectra of the plurality of sample hairs. The acquisition unit further acquires color information related to the color of the human hair measured using light in the wavelength range of the visible light region. The determination unit determines the index by applying the at least one spectral spectrum and the acquired color information to the learning model. The computer program according to claim 6 or 7.

9. A hair condition analysis device for analyzing the condition of hair, an acquisition unit that acquires at least one spectral spectrum of human hair measured using light of at least one wavelength in the wavelength range of 700 nm or more and 2000 nm or less; a determination unit that determines an index indicating the degree of damage to the human hair by applying the at least one spectral spectrum to a learning model obtained by machine learning of each spectral spectrum of a plurality of sample hairs measured using light in the wavelength range; an output unit that outputs related information related to the determined index; A hair condition analysis device comprising the above.

10. A hair condition analysis method for analyzing the condition of hair, an acquisition step of acquiring at least one spectral spectrum of human hair measured using light of at least one wavelength in the wavelength range of 700 nm or more and 2000 nm or less; a determination step of determining an index indicating the degree of damage to the human hair by applying the at least one spectral spectrum to a learning model obtained by machine learning of each spectral spectrum of a plurality of sample hairs measured using light in the wavelength range; an output step of outputting related information related to the determined index; A hair condition analysis method comprising the above.

Images