Method for determining a target cosmetic composition for hair coloring, including hair, electronic device for determination and associated computer program product

Abstract

Method for determining a target cosmetic composition for hair coloring, in particular head hair, electronic determination device and associated computer program product. The present invention relates to a method for determining a target cosmetic composition (11*) for hair coloring (12), in particular head hair. The method comprises an operating phase including the acquisition of at least one target colorimetric parameter, the obtaining of a set of cosmetic compositions to be tested, and the obtaining of at least one colorimetric parameter associated with each cosmetic composition to be tested. The operating phase also includes filtering the cosmetic compositions to be tested according to the at least one colorimetric parameter associated with them, to obtain filtered cosmetic compositions, and the formation of several groups of filtered cosmetic compositions.The exploitation phase also includes the selection of one of the groups and the determination of the target cosmetic composition based on the filtered cosmetic compositions of the selected group. Figure for the summary: Figure 1.

Description

Title of the invention: Method for determining a target cosmetic composition for coloring hair, in particular hair, associated electronic determination device and computer program product

[0001] The present invention relates to a method for determining a target cosmetic composition for coloring hair, in particular hair, as a function of at least one target colorimetric parameter. The present invention also relates to an associated computer program product and electronic determination device.

[0002] The present invention relates to the field of cosmetic products, preferably cosmetic compositions for coloring hair, in particular hair.

[0003] More generally, a “cosmetic product” is a product as defined in

[0004] Regulation CEN-1223 / 2009 of the European Parliament and of the Council dated November 30

[0005] 2009, relating to cosmetic products.

[0006] One goal of the cosmetics industry is to improve the experience of its consumers. In particular, there is a strong trend towards offering products that are increasingly tailored to the specific needs and characteristics of the user. This trend is generally referred to as "personalization."

[0007] The personalization of cosmetic products and services can concern any part of the human body but is of particular interest for exposed body parts, such as the face (makeup or care products), and the hair or a beard where appropriate (coloring products for example).

[0008] In the field of cosmetic compositions for coloring hair(s), particularly hair, new cosmetic compositions are regularly developed to adapt to user demand.

[0009] These new cosmetic compositions aim, in a first case, to achieve a new color or a new visual effect of the hairs on which they are applied. In a second case, these new cosmetic compositions aim to achieve an already known color but include components which were not previously used for this purpose. This second case is particularly interesting when the supply of certain components becomes difficult, is accompanied by high costs, or presents a risk for the environment.

[0010] However, obtaining new cosmetic compositions results from a process

[0011]

[0012]

[0013]

[0014]

[0015]

[0016]

[0017] long and complex. First, chemical experts develop prototypes of cosmetic compositions. These prototypes are the result of the experience of chemical experts and are governed by certain chemical principles. These prototypes are then tested individually in the laboratory to verify that the visual effect obtained meets the expectations of the chemical experts. In some cases, and unpredictably, the visual effect associated with a newly developed prototype does not meet expectations. It is therefore not always possible to capitalize on the work of the chemical experts who developed these prototypes. In addition, it is sometimes desired to obtain several cosmetic compositions leading to the same color or the same visual effect. This makes it possible to anticipate future problems due to the unavailability of certain components, while ensuring that the demand of users wishing to obtain the said color or visual effect can continue to be met. There is therefore a need for tools to simplify the development of a new cosmetic composition. To this end, the present invention relates to a method for determining a target cosmetic composition for coloring hair, in particular hair, as a function of at least one target colorimetric parameter, the method being implemented by an electronic determination device and comprising an operating phase comprising the following steps: - the acquisition of at least one target colorimetric parameter, - obtaining a set of cosmetic compositions to test, each com cosmetic position to be tested including components, - obtaining at least one colorimetric parameter associated with each cosmetic composition to be tested, - filtering the cosmetic compositions to be tested according to at least one colorimetric parameter associated with them, to obtain filtered cosmetic compositions, the at least one colorimetric parameter associated with each filtered cosmetic composition meeting a criterion relating to the at least one target colorimetric parameter, - the formation of several groups of filtered cosmetic compositions so that the filtered cosmetic compositions of the same group comprise the same components, - the selection of a group of filtered cosmetic compositions, called the selected group, and - the determination of the target cosmetic composition based on the com- filtered cosmetic positions of the selected group.

[0018] Thanks to the steps of filtering and forming groups of cosmetic compositions, it is possible to group together the cosmetic compositions of which at least one colorimetric parameter is close to the at least one target colorimetric parameter and of which the components are identical. Thus, it is possible to quickly determine one or more cosmetic compositions of which at least one colorimetric parameter corresponds to the target colorimetric parameter(s).

[0019] According to particular embodiments, the determination method according to the invention comprises one or more of the following characteristics, taken in isolation or in all technically possible combinations: - each filtered cosmetic composition comprises, for each of its components, a quantity of said component in the filtered cosmetic composition,

[0020] the step of determining the operating phase comprising: • for each component of the filtered cosmetic compositions of the selected group, the calculation of an average value of the quantities of said component among said filtered cosmetic compositions, • the determination of an average cosmetic composition comprising, for each of said components, a quantity equal to the respective calculated average value of the component, and • the formation of the target cosmetic composition based on the average cosmetic composition; • during the determination step of the exploitation phase, the formation of the target cosmetic composition comprises the calculation of at least one colorimetric parameter associated with the average cosmetic composition and the optimization of the quantity of each component of the average cosmetic composition so that the at least one colorimetric parameter associated with the average cosmetic composition approaches the at least one target colorimetric parameter,

[0021] the target cosmetic composition being the average cosmetic composition resulting from the optimization of its quantities of components; - during the selection step of the exploitation phase, M groups of filtered cosmetic compositions are selected, M being greater than or equal to two, during the determination step of the exploitation phase, for each selected group, a corresponding target cosmetic composition being determined; - during the selection step of the exploitation phase, the M groups are selected so that, for each selected group, the components included in the filtered cosmetic compositions of said selected group are only included in said selected group; - the method comprises, prior to the operating phase, a training phase comprising the following steps: • the receipt of a set of constraints on the viability of cosmetic composition(s), • the acquisition of a training data set, each training data being specific to a cosmetic composition respecting the set of viability constraints received,

[0022] each training data comprising: • a set of quantities representing the quantity of each component of the cosmetic composition, and • a value of at least one colorimetric parameter associated with the cosmetic composition, and • training an artificial intelligence model on the basis of the training data set, to obtain a trained model,

[0023] during the step of obtaining at least one colorimetric parameter associated with each cosmetic composition to be tested, of the exploitation phase, each colorimetric parameter being obtained by applying the trained model to a corresponding cosmetic composition to be tested; - the stage of obtaining a set of cosmetic compositions to be tested, the exploitation phase includes: • the receipt of a number N corresponding to the number of cosmetic compositions to be tested, the number N being greater than or equal to two, • the generation of N test data,

[0024] each test data being relative to a cosmetic composition whose components are chosen from a predefined list of components and such that the cosmetic composition associated with each test data is included in a reduced space of viable cosmetic compositions, the reduced space comprising only the cosmetic compositions respecting a set of viability constraint(s), and

[0025] the N test data being representative of said reduced space of viable cosmetic compositions,

[0026] the set of cosmetic compositions to be tested being formed from the N generated test data; - during the exploitation phase, the generation of N test data of the obtaining step comprises the calculation of magnitudes of the N test data so that, for each test data, the distance between said test data and the other test data is maximum,

[0027] each quantity of each test data representing the quantity of a component in the corresponding cosmetic composition to be tested - the at least one colorimetric parameter comprises: • a triplet of values ​​characterizing a hair color after application of the cosmetic composition, or • a color fading value of the cosmetic composition after washing the hair, or • a selectivity value characterizing a color difference between a root and a tip of the hairs; and - the exploitation phase further comprises a step of producing at least one sample of the target cosmetic composition, with a view to applying it to hairs to verify the value of the at least one target colorimetric parameter.

[0028] The present invention also relates to a computer program product on which is stored a computer program comprising program instructions, the computer program being loaded onto a data processing unit and causing the implementation of such a method when the computer program is implemented on the data processing unit.

[0029] The present invention also relates to an electronic device for determining a target cosmetic composition for coloring hair, in particular head hair, as a function of at least one target colorimetric parameter, the electronic determination device comprising a processing unit capable of implementing such a method.

[0030] The present invention also relates to a readable information medium on which is stored a computer program product comprising program instructions, the computer program being loaded onto a data processing unit and causing the implementation of such a determination method when the computer program is implemented on the data processing unit.

[0031] Other characteristics and advantages of the invention will appear on reading the following description of embodiments of the invention, given by way of example only and with reference to the drawings which are: - [Fig.l] [Fig.l] is a schematic view of a concept of the invention; - [Fig.2] [Fig.2] and a schematic view of an electronic device for determining mination according to the invention; - [Fig.3] [Fig.3] is a flowchart of a determination method according to the invention implemented by the electronic determination device of [Fig.2]; - [Fig.4] [Fig.4] is a two-dimensional schematic representation of a reduced space for viable cosmetic compositions; and - [Fig.5] [Fig.5] is a two-dimensional schematic representation of steps of the determination process illustrated in [Fig.3].

[0032] In [Fig.l] is shown an electronic device 10 for determining a target cosmetic composition 11* for coloring hairs 12, in particular head hair, as a function of at least one target colorimetric parameter 13*.

[0033] As visible in [Fig.l], the target cosmetic composition 11* is a cosmetic composition for coloring hair 12. Preferably, the target cosmetic composition 11* is suitable for being applied to the hair 12 of a user, to color said hair 12.

[0034] The target cosmetic composition 11* comprises a plurality of components 14 capable of interacting with each other, and with the hair 12 of the user, to color it.

[0035] Preferably, during this interaction, the components 14 of the target cosmetic composition 11* begin by bleaching the hair 12. Then, the components 14 of the cosmetic composition 11 interact with the bleached hair to fix pigments of the chosen color and thus form colored hair 12*.

[0036] The colored hair 12* has a color corresponding to the at least one target colorimetric parameter 13*.

[0037] The components 14 are for example present in the target cosmetic composition 11* in the form of powder, gel, emulsion or oil.

[0038] The electronic determination device 10 is configured to determine, from the at least one target colorimetric parameter 13*, a set of quantities representing the quantity of each component 14 forming a target cosmetic composition 11*. The target cosmetic composition 11* is intended to produce a hair coloring corresponding to the at least one target colorimetric parameter 13*, when it is applied to the hair 12 of the user.

[0039] With reference to [Fig.2], the determination device 10 comprises a processing unit 15.

[0040] The determination device 10 optionally further comprises a display screen 16 and / or a unit 17 for producing cosmetic compositions 11.

[0041] The processing unit 15 comprises, for example, a calculator interacting with a computer program product. The processing unit 15 is, for example, a computer.

[0042] The calculator comprises, for example, a processor comprising a data processing unit, memories and an information medium reader, as well as possibly a human-machine interface.

[0043] The computer program product comprises an information medium.

[0044] The information medium is a medium readable by the computer, usually by the data processing unit. The readable information medium is a medium suitable for storing electronic instructions and capable of being coupled to a bus of a computer system.

[0045] For example, the information medium is a USB key, a floppy disk or flexible disk (from the English name "Floppy disk"), an optical disk, a CD-ROM, a magneto-optical disk, a ROM memory, a RAM memory, an EPROM memory, an EEPROM memory, a magnetic card or an optical card.

[0046] The computer program comprising program instructions is stored on the information medium.

[0047] The computer program is loadable onto the processing unit 15 and is adapted to cause the implementation of a method for determining a target cosmetic composition 11*, when the computer program is implemented on the processing unit of the computer. Such a determination method will be described in the remainder of the description.

[0048] The operation of the electronic device 10, implementing a method for determining a target cosmetic composition 11*, will now be described with reference to the flowchart of [Fig. 3], as well as to the examples of Figures 4 and 5.

[0049] The determination method optionally comprises a training phase 100.

[0050] The training phase 100 comprises a step 110 of receiving a set of cosmetic composition viability constraints 11.

[0051] Preferably, each cosmetic composition(s) 11 comprises at least one component 14 of a first type and at least one component 14 of a second type.

[0052] Each component 14 of the first type is for example a base, and each component 14 of the second type is for example a coupler.

[0053] Preferably, each component 14 is chosen from a predefined list of components.

[0054] Preferably, the set of viability constraints of cosmetic composition(s) 11 comprises one or more of the following constraints: - a ratio between a quantity of component 14 of the first type and a quantity of component 14 of the second type is between a first threshold and a second threshold, - a total quantity of the cosmetic composition 11 is less than a third threshold, - a quantity of each component 14 in the cosmetic composition 11 is less than a fourth threshold, - a number of components 14 in the cosmetic composition 11 is less Advantageously, the set of constraint(s) comprises each of the aforementioned constraints.

[0055] These constraints are optionally derived from chemical principles making it possible to ensure the viability of the cosmetic composition 11. The expression “viability of the cosmetic composition 11” is understood here to mean that a composition respecting the set of constraints: is effective, and / or does not present any risks for the user 13, and / or complies with production standards, and is preferably environmentally friendly.

[0056] The optional training phase 100 further comprises a step 120 of acquiring a training data set 18.

[0057] Each training data item 18 of the training data set 18 is specific to a cosmetic composition 11 respecting the set of viability constraints received during the reception step 110.

[0058] Each training data 18 comprises a set of quantities representing the quantity of each component 14 in the cosmetic composition 11, and a value of at least one colorimetric parameter 13 associated with the cosmetic composition 11.

[0059] Each quantity is for example a quantity of the corresponding component 14 in the associated cosmetic composition 11.

[0060] Each quantity of component 14 is optionally a mass of component, a volume of component, a mass percentage of component 14 in cosmetic composition 11, or a volume percentage of component 14 in cosmetic composition 11.

[0061] The or each colorimetric parameter 13 characterizes the cosmetic composition 11 for coloring hair 12.

[0062] Preferably, each colorimetric parameter 13 is representative of a visual effect of the hair 12 to which the cosmetic composition 11 is applied.

[0063] Advantageously, the at least one colorimetric parameter 13 comprises: - a triplet of values ​​characterizing a hair color 12 after application of the cosmetic composition 11, and / or - a color fading value of the cosmetic composition 11 after washing the hair 12, and / or - a selectivity value characterizing a difference in color between a root and a tip of the hair 12.

[0064] Preferably, each training data 18 comprises each of the aforementioned values.

[0065] The triplet of values ​​is for example the triplet of values ​​of the CIE L*a*b* color space.

[0066] The CIE L*a*b* color space, often abbreviated CIELAB, is a color space- metric for surface colors defined by the International Commission on Illumination (CIE) in 1976. Based on the evaluations of the CIE XYZ system, it abandons its linearity in order to more faithfully represent the differences between colors perceived by human vision. In this model, three quantities characterize the colors, the lightness L*, which derives from the luminance (Y) of the XYZ evaluation, and two parameters a* and b*, which express the deviation of the color from that of a gray surface of the same lightness, such as chrominance. The definition of a gray, uncolored, achromatic surface involves explicitly indicating the composition of the light that illuminates the colored surface. This illuminant is often daylight, which corresponds to the standardized standard D65.

[0067] Also, in this case, the triplet of values ​​includes a value of clarity L*, and two values ​​a*, b* of deviation of the color from that of a gray surface of the same clarity.

[0068] Any other color space is possible.

[0069] Alternatively, the triplet of values ​​is an RGB triplet. The triplet then includes an R value for the color red, a G value for green and a B value for blue.

[0070] More generally, any representations allowing spectra to be associated would also be usable in this context.

[0071] The color of a hair is not identical at every point of the hair. In particular, the tips of the hair 12 are generally lighter than the roots. Thus, the selectivity value is a parameter of interest.

[0072] Preferably, during the acquisition step 120 of the training phase 100, the training data set 18 is acquired, at least in part, from a sensor capable of measuring the at least one colorimetric parameter 13.

[0073] The sensor is for example a spectro-colorimeter.

[0074] According to one embodiment, the sensor is capable of capturing images of hair areas of the individual and of extracting colorimetric measurements from the captured images.

[0075] Preferably, each training data 18 comes from a viable cosmetic composition 11, i.e. respecting the set of viability constraints.

[0076] Preferably, each training data corresponds to a commercialized cosmetic composition. Thus, before the commercialization of said cosmetic compositions, their colorimetric parameters 13 were measured during testing on hair 12.

[0077] As an optional addition, the cosmetic compositions associated with the training data are preselected by chemical experts from among all commercially available cosmetic compositions.

[0078] The optional training phase 100 further comprises a step 130 training an artificial intelligence model from the training dataset 18, to obtain a trained model.

[0079] Preferably, the artificial intelligence model comprises at least one of the following models: a support vector machine, a decision tree forest (from the English Random Forest), a gradient boosting mechanism (from the English Gradient BoostingY or kriging, also called Gaussian Process Regressor).

[0080] In each of the preceding cases, the artificial intelligence model comprises adjustable parameters. During the training step 130, each adjustable parameter is adjusted so that, when a respective set of quantities of a training data item 18 is given to it as input, said trained model provides as output one or more colorimetric parameters 13 substantially identical to the colorimetric parameter(s) of said training data item 18.

[0081] It is clear that the training phase 100 described above aims to obtain the trained model.

[0082] According to a variant not shown, the determination method does not include the training phase 100.

[0083] For example, the method instead comprises a step of receiving the trained model. In another example, the method instead comprises nothing.

[0084] The method further comprises an operating phase 200.

[0085] The operating phase 200 comprises a step 210 of acquiring the at least one target colorimetric parameter 13*. The at least one target colorimetric parameter 13* is for example provided by an operator of the determination device 10.

[0086] The operating phase 200 further comprises a step 220 of obtaining a set of cosmetic compositions 11 to be tested.

[0087] The set of cosmetic compositions 11 to be tested is preferably formed from a plurality of test data 19.

[0088] The test data 19 comprise, for each cosmetic composition 11 to be tested, a set of quantities representing the quantity of each component 14 in the cosmetic composition 11 to be tested.

[0089] Preferably, the components 14 of each cosmetic composition 11 to be tested are chosen from the predefined list of components 14 previously described.

[0090] Each cosmetic composition 11 to be tested complies with the set of viability constraints received during the training phase 100.

[0091] If the method does not include the training phase 100, then the step 220 of obtaining the set of cosmetic compositions 11 to be tested further comprises receiving said set of viability constraints.

[0092] Step 220 of obtaining the set of cosmetic compositions 11 to be tested advantageously comprises receiving a number N corresponding to a number of cosmetic compositions 11 to be tested. Optionally, the number N is greater than or equal to two. Preferably, the number N is greater than one hundred, for example equal to one hundred and fifty.

[0093] The obtaining step 220 further advantageously comprises the generation of N test data 19. Each test data 19 relates to a cosmetic composition 11 whose components 14 are chosen from the predefined list of components 14 and such that the cosmetic composition 11 associated with each test data 19 is included in a reduced space 20 of viable cosmetic compositions. The reduced space 20 of viable cosmetic compositions comprises only the cosmetic compositions 11 respecting the set of viability constraints. The test data 19 are representative of the reduced space 20 of viable cosmetic compositions.

[0094] The expression “test data 19 representative of the reduced space 20” is understood here to mean that the test data 19 are judiciously chosen in the reduced space 20 to cover substantially the entirety of the reduced space 20.

[0095] [Fig.4] illustrates, via a two-dimensional schematic diagram, a set 22 representing all possible cosmetic compositions 11 and the reduced space 20. It is clear that this set 22 and this space 20 are, in reality, not representable in two dimensions but rather in a number of dimensions equal to the number of components 14 in the predefined list of components.

[0096] In [Fig.4], the set 22 of all possible cosmetic compositions 11 corresponds to the plane of [Fig.4]. As visible in [Fig.4], the reduced space 20 is included in said set 22 because the reduced space 20 only includes the cosmetic compositions 11 of the set 20 respecting the viability constraints.

[0097] In [Fig.4] still, we can see the test data 19 in the reduced space 20. As an example, ten test data 19 are represented in [Fig.4].

[0098] Preferably, the generation of N test data 19 comprises the calculation of magnitudes of the N test data 19 so that, for each test data 19, the distance between said test data 19 and the other test data 19 is maximum.

[0099] Each quantity of each test data 19 represents the quantity of a component 14 in the corresponding cosmetic composition 11 to be tested.

[0100] The generation of the N test data 19 comprises the calculation of quantities of N sets of quantities for which a distance between each pair of sets of quantities is maximum. Each quantity of a respective set of quantities represents the quantity of a component 14 in a cosmetic composition 11 associated with said set of quantities, said cosmetic composition 11 respecting the set of constraints. By way of example, a test data 19 comprises a first size equal to 4 mol of a first component 14A and 6 mol of a second component 14B.

[0101] Advantageously, each test data 19 is a vector comprising a coefficient for each component 14 of the predefined list of components.

[0102] Each coefficient represents the quantity of said component 14 in the respective cosmetic composition 11 of the associated vector.

[0103] A distance between two test data 19, i.e. between two vectors, is for example defined by the algebraic norm according to the following equation:

[0104] [Math.l] distance(X, Y) - || X- F || =

[0105] where X and Y are two sets of P quantities,

[0106] x,'> >=Ç->p and y, are the respective magnitudes of said sets X and F,

[0107] is the square root function, and

[0108] is the sum operator.

[0109] The calculation of the N vectors is for example carried out by applying a design of experiments technique, also called DoE technique (from the English Design of Experimentation). Advantageously, the DoE technique implements an SFD algorithm (from the English Space Filling Design). This algorithm makes it possible, even with a reduced number of vectors, to ensure that the chosen vectors are representative of the reduced space 20. Indeed, when the number of vectors is small, the assumptions of the law of large numbers are not sufficiently respected so that a random distribution of vectors can be representative of the reduced space 20.

[0110] In [Fig.4], it is visible that the test data 19 are spaced from each other so as to maximize the distances between the test data 19.

[0111] Alternatively, the obtaining step 220 comprises only the acquisition of one or more test data 19 chosen by an operator of the determination device 10. In this case, the operator chooses one or more cosmetic compositions by selecting, for each cosmetic composition 11, components 14 from the predefined list of components, and the quantity of each of said components 14. When making his choice, the operator ensures that each chosen cosmetic composition 11 complies with the set of constraints.

[0112] The set of cosmetic compositions to be tested is formed by the N test data 19.

[0113] The operating phase 200 further comprises a step 230 of obtaining at least one colorimetric parameter 13 associated with each cosmetic composition 11 to be tested.

[0114] Preferably, the obtaining step 230 comprises the application, to the test data 19, forming the set of cosmetic compositions 11 obtained, of the model trained to determine the at least one colorimetric parameter 13 of the or each cosmetic composition 11 to be tested.

[0115] Preferably, during the obtaining step 230, each test data 19 is successively provided to the trained model. The trained model then determines, for each test data 19, the associated colorimetric parameter(s) 13.

[0116] Advantageously, during the obtaining step 230, the cosmetic composition 11 and the colorimetric parameter(s) 13 associated with each test data item are stored in the memory(s) of the processing unit 15.

[0117] Alternatively, the obtaining step 230 comprises receiving, from the operator of said device 10, at least one colorimetric parameter 13 associated with each cosmetic composition 11 of the set of cosmetic compositions 11 to be tested.

[0118] In this case, said at least one colorimetric parameter 13 is for example measured by the operator of the determination device 10 for example by means of a sensor, as explained previously for the test data 19.

[0119] This is particularly advantageous if the method does not include the training phase 100, nor the reception of the trained model.

[0120] This variant combines particularly advantageously with the variant of step 220 of obtaining the set of cosmetic compositions 11 to be tested, according to which the test data 19 associated with the cosmetic compositions 11 to be tested are received from the operator.

[0121] The operating phase 200 further comprises a filtering step 235, during which the cosmetic compositions 11 to be tested are filtered according to the at least one colorimetric parameter 13 associated with them. The filtering step 235 makes it possible to obtain filtered cosmetic compositions 30, for which the at least one colorimetric parameter 13 meets a criterion relating to the at least one target colorimetric parameter 13*.

[0122] Preferably, the criterion is that a relative difference between the at least one colorimetric parameter 13 associated with each cosmetic composition 11 to be tested and the at least one target colorimetric parameter 13* is less than a sixth threshold. The sixth threshold is for example equal to 10%. The filtered cosmetic compositions 30 are only the cosmetic compositions 11 to be tested for which said relative difference is less than the sixth threshold. In other words, the filtered cosmetic compositions 30 are the cosmetic compositions 11 to be tested for which each colorimetric parameter 13 deviates, from the at least one target colorimetric parameter 13*, by at most the sixth threshold.

[0123] The operating phase 200 further comprises a step 240 of forming several groups 35 of filtered cosmetic compositions 30. For each group 35 (from the English cluster), the filtered cosmetic compositions 30 comprise the same components 14. The training step 240 comprises for example the application of an unsupervised learning algorithm, such as a K-means algorithm.

[0124] The filtered cosmetic compositions 30 of each group 35 are substantially similar. In other words, two filtered cosmetic compositions 30 comprising the same components 14 but in substantially different proportions are not included in the same group 35. For example, a filtered cosmetic composition 30 comprising 10% of a component 14A and 90% of a component 14B is not included in the same group 35 as a filtered cosmetic composition 30 comprising 80% of component 14A and 20% of component 14B.

[0125] In [Fig.5], several groups 35A, 35B, 35C, 35D are represented by ellipses. The filtered cosmetic compositions 30 of each group 35A, 35B, 35C, 35D are represented by crosses.

[0126] As visible in [Fig.5], a first group 35A comprises the filtered cosmetic compositions 30 formed from components 14A and 14B. A second group 35B comprises the filtered cosmetic compositions 30 formed from components 14A, 14B and 14C. A third group 35C comprises the filtered cosmetic compositions 30 formed from components 14C and 14D, in proportions substantially also at 80 / 20. A fourth group 35D comprises the filtered cosmetic compositions 30 formed from components 14C and 14D, in proportions substantially equal to 40 / 60.

[0127] The operating phase 200 further comprises a step 250 of selecting at least one group 35 of filtered cosmetic compositions 30, called selected group 35*. The selected group 35* is for example selected randomly.

[0128] According to a variant, during the selection step 250, M groups 35 are selected, M being greater than or equal to two, advantageously equal to four.

[0129] For this, the M selected groups 35* are preferentially selected so that, for each selected group 35*, the components 14 included in the filtered cosmetic compositions 30 of said selected group 35* are only included in said selected group 35*. In other words, the components 14 included in the filtered cosmetic compositions 30 of said selected group 35* are exclusive to said group 35*. Formulated still differently, the components 14 included in the filtered cosmetic compositions 30 of said selected group 35* are not included in the filtered cosmetic compositions 30 of the other selected groups 35*.

[0130] In other words, two selected groups 35* do not include filtered cosmetic compositions 30 having the same component 14.

[0131] In the example illustrated in [Fig.5], the first 35A and second 35B groups cannot therefore together form part of the M selected groups 35*. Similarly, the third 35B and fourth 35D groups cannot therefore together form part of the M selected groups 35*.

[0132] Thus, in the example of [Fig.5], the selected groups 35* are for example: - the first 35A and the third 35C groups, - the first 35A and the fourth 35D groups, - the second 35B and the third 35C groups, or - the second 35B and the fourth 35D groups.

[0133] Preferably, the selected groups 35* are the groups 35 further comprising the filtered cosmetic compositions 30 for which each associated colorimetric parameter 13 is closest to the at least one target colorimetric parameter 13*.

[0134] The exploitation phase 200 further comprises a step 260 of determining the target cosmetic composition 11* as a function of the filtered cosmetic compositions 30 of the selected group 35*.

[0135] Advantageously, when several groups 35 are selected, during the determination step 260, a target cosmetic composition 11* is determined for each group 35 selected.

[0136] The determination step 260 preferably comprises, for each component 14 of the filtered cosmetic compositions 30 of the selected group 35*, the calculation of an average value of the quantities of said component 14 among said filtered cosmetic compositions 30. In the example of [Fig.5], if the selected group 35* is the first group 35A, said calculation is the calculation of an average value of the quantity of component 14A in each filtered cosmetic composition 30 of the first group 35A and the calculation of an average value of the quantity of component 14B in each filtered cosmetic composition 30 of the first group 35A.

[0137] The determination step 260 further preferably comprises the determination of an average cosmetic composition 40 comprising, for each of said components 14, a quantity equal to the respective calculated average value of the component 14. In [Fig.5], the average cosmetic composition 40A, 40B, 40C, 40D of each group 35A, 35B, 35C, 35D is represented by a square.

[0138] It is then understood that the average cosmetic composition 40 is a barycenter of the filtered cosmetic compositions 30 of the selected group 35*.

[0139] Advantageously, the determination step 260 further comprises the formation of the target cosmetic composition 11* as a function of the average cosmetic composition 40.

[0140] For example, the target cosmetic composition 11* is the cosmetic composition average 40.

[0141] Alternatively, the determination of the target cosmetic composition 11* as a function of the average cosmetic composition 40 comprises the calculation of at least one colorimetric parameter 13 associated with the average cosmetic composition 40, and the optimization of the quantity of each component 14 of the average cosmetic composition 40 so that the at least one colorimetric parameter 13 associated with the average cosmetic composition 40 approaches the at least one target colorimetric parameter 13*.

[0142] In other words, an optimization algorithm is applied to a difference between the at least one colorimetric parameter 13 associated with the average cosmetic composition 40 and the at least one target colorimetric parameter 13*, to reduce this difference by varying the quantity of each component 14 of the average cosmetic composition 40.

[0143] According to this variant, the target cosmetic composition 11* is the average cosmetic composition 40 resulting from the optimization of the quantity of its components 14.

[0144] The calculation of the at least one colorimetric parameter 13 associated with the average cosmetic composition 40 is for example implemented by the trained model.

[0145] As an optional addition, when M groups 35 are selected, said determining step 260 is repeated for each selected group 35*. Thus, a target composition 11* is determined for each selected group 35*.

[0146] As an optional addition, the operating phase 200 further comprises a step 270 of sending the or each target cosmetic composition 11* to the display screen 16. The display screen 16 then displays the or each target cosmetic composition 11* for the operator of the determination device 10.

[0147] According to another optional addition, the operating phase 200 further comprises a step 280 of producing at least one sample of the or each target cosmetic composition 11*. The production of the at least one sample is preferably carried out by the unit 17 for producing cosmetic compositions 11.

[0148] Alternatively, the production step 280 comprises receiving, from the operator, a selection of one or more target cosmetic compositions 11* to be produced, from among the determined target cosmetic compositions 11. The production unit 17 then produces only the sample(s) of the target cosmetic compositions 11* selected by the operator.

[0149] The sample(s) of target cosmetic compositions 11* produced are intended to be applied to hair 12, preferably a lock of hair 12, to verify the colorimetric parameter(s).

[0150] Thus, following the application of each sample on hair 12, the parameters colorimetrics are measured using the same sensors used to form the training data 18.

[0151] Alternatively, the operating phase 200 does not include the sending step 270 and / or the production step 280.

[0152] As a further variant, each cosmetic composition 11 is suitable for being applied to any type of hair 12, and not only to hair, for example to eyelashes, eyebrows or a beard of the user.

[0153] The exploitation phase 200 is preferably repeated a plurality of times, forming iterations. At each iteration, the at least one target colorimetric parameter 13* acquired during the acquisition step 210 is distinct from the at least one target colorimetric parameter 13* acquired for the previous iteration(s).

[0154] Preferably, during each repetition, i.e. during each iteration successive to the first iteration, the steps of obtaining 220 the set of cosmetic compositions, and of obtaining 230 at least one colorimetric parameter 13 associated with each cosmetic composition 11 to be tested, and of formation 240, are not implemented.

[0155] All the variants and optional additions listed above can be combined with each other.

[0156] With the determination method according to the invention, it is possible to automatically obtain the target cosmetic composition 11* from at least one target colorimetric parameter 13*.

Claims

Claims

1. Method for determining a target cosmetic composition (11*) for coloring hair (12), in particular hair, as a function of at least one target colorimetric parameter (13*), the method being implemented by an electronic determination device (10) and comprising an operating phase (200) comprising the following steps: - acquiring (210) the at least one target colorimetric parameter (13*), - obtaining (220) a set of cosmetic compositions (11) to be tested, each cosmetic composition (11) to be tested comprising components (14), - obtaining (230) at least one colorimetric parameter (13) associated with each cosmetic composition (11) to be tested, - filtering (235) the cosmetic compositions (11) to be tested as a function of the at least one colorimetric parameter (13) associated with them, to obtain filtered cosmetic compositions (30),the at least one colorimetric parameter (13) associated with each filtered cosmetic composition (30) meeting a criterion relating to the at least one target colorimetric parameter (13*), - the formation (240) of several groups (35) of filtered cosmetic compositions (30) so that the filtered cosmetic compositions (30) of the same group (35) comprise the same components (14), - the selection (250) of a group (35) of filtered cosmetic compositions (30), called selected group (35*), and - the determination (260) of the target cosmetic composition (11*) as a function of the filtered cosmetic compositions (30) of the selected group (35*).,

2. Method according to claim 1, in which, each filtered cosmetic composition (30) comprises, for each of its components (14), an amount of said component (14) in the filtered cosmetic composition (30), the step of determining (260) the exploitation phase (200) comprising: - for each component (14) of the filtered cosmetic compositions (30) of the selected group (35*), calculating an average value of the quantities of said component (14) among said filtered cosmetic compositions (30), - determining an average cosmetic composition (40) comprising, for each of said components (14), an amount equal to the respective calculated average value of the component (14), and - forming the target cosmetic composition (11*) as a function of the average cosmetic composition (40).

3. The method of claim 1 or 2, wherein in the determining step (260) of the operating phase (200), the formation of the target cosmetic composition comprises calculating at least one colorimetric parameter (13) associated with the average cosmetic composition (40) and optimizing the quantity of each component (14) of the average cosmetic composition (40) such that the at least one colorimetric parameter (13) associated with the average cosmetic composition (40) approaches the at least one target colorimetric parameter (13*), the target cosmetic composition (11*) being the average cosmetic composition (40) resulting from the optimization of its quantities of components (14).

4. Method according to any one of claims 1 to 3, wherein during the selection step (250) of the exploitation phase (200), M groups (35) of filtered cosmetic compositions (30) are selected, M being greater than or equal to two, during the determination step (260) of the exploitation phase (200), for each selected group (35*), a corresponding target cosmetic composition (11*) being determined.

5. Method according to claim 4, wherein during the selection step (250) of the exploitation phase (200), the M groups (35*) are selected so that, for each selected group (35*), the components (14) included in the cosmetic compositions (14) filtered (30) of said selected group (35*) are only included in said selected group (35*).

6. Method according to any one of claims 1 to 5, in which the method comprises, prior to the operating phase (200), a training phase (100) comprising the following steps: - receiving (110) a set of viability constraints of cosmetic composition(s), - the acquisition (120) of a training data set (18), each training data (18) being specific to a cosmetic composition (11) respecting the set of viability constraints received, each training data (11) comprising: • a set of quantities representing the quantity of each component (14) of the cosmetic composition (11), and • a value of at least one colorimetric parameter associated with the cosmetic composition (11), and • training (130) an artificial intelligence model based on the training data set (18), to obtain a trained model, during the step (230) of obtaining at least one colorimetric parameter (13) associated with each cosmetic composition to be tested, of the exploitation phase (200), each colorimetric parameter (13) being obtained by applying the trained model to a corresponding cosmetic composition (11) to be tested.

7. Method according to any one of claims 1 to 6, in which the step (220) of obtaining a set of cosmetic compositions to be tested, of the exploitation phase (200) comprises: - the receipt of a number N corresponding to the number of cosmetic compositions (11) to be tested, the number N being greater than or equal to two, - the generation of N test data (19), each test data (19) being relative to a cosmetic composition (11) whose components (14) are chosen from a predefined list of components (14) and such that the cosmetic composition (11) associated with each test data (19) is included in a space reduced space of viable cosmetic compositions, the reduced space comprising only the cosmetic compositions (11) respecting a set of viability constraints, and the N test data (19) being representative of said reduced space of viable cosmetic compositions, the set of cosmetic compositions (11) to be tested being formed from the N test data (19) generated.

8. Method according to claim 7, wherein, during the exploitation phase (200), the generation of N test data (19) of the obtaining step (220) comprises the calculation of magnitudes of the N test data (19) so that, for each test data (19), the distance between said test data (19) and the other test data (19) is maximum, each magnitude of each test data (19) representing the quantity of a component (14) in the corresponding cosmetic composition (11) to be tested.

9. Method according to any one of claims 1 to 8, in which the at least one colorimetric parameter (13) comprises: - a triplet of values (L*, a*, b*; R, G, B) characterizing a color of hairs (12) after application of the cosmetic composition (11), or - a value of erasure of the color of the cosmetic composition (11) after washing the hairs (12), or - a selectivity value characterizing a difference in color between a root and a tip of the hairs (12).

10. Method according to any one of claims 1 to 9, in which the exploitation phase (200) further comprises a step of producing (280) at least one sample of the target cosmetic composition (11*), with a view to applying it to hairs (12) to verify the value of the at least one target colorimetric parameter (13*).

11. A computer program product on which is stored a computer program comprising program instructions, the computer program being loaded onto a data processing unit and causing the implementation of a method according to any one of claims 1 to 10 when the computer program is implemented on the data processing unit.

12. Electronic device (10) for determining a target cosmetic composition (11*) for coloring hair (12), in particular head hair, as a function of at least one target colorimetric parameter (13*), the electronic determination device (10) comprising a processing unit capable of implementing a method according to any one of claims 1 to 10.

Images