A method for quantifying the impression of volume conferred by hair

Abstract

Quantification of an impression of volume conferred by hair Method for quantifying an impression of volume conferred by hair (4), comprising the steps of: a) acquiring a front image and / or a profile image of the hair of a person (1) to be evaluated, b) segmenting the hair on each image, c) defining within the hair segmented on the front image at least one region of interest comprising hair roots (7) and / or at least one region of interest comprising hair lengths (8);and / or define within the segmented hair on the profile image a region of interest comprising a mass of hair (6), d) deliver a value of at least one characteristic parameter, extracted from one of the regions of interest or corresponding to the sum of intermediate parameters extracted from several regions of interest, the characteristic parameter being associated with a visual estimate v of the volume conferred by the hair in the regions of interest. Figure for the abstract: Fig. 1;

Description

Title of the invention: Method for quantifying the impression of volume conferred by hair technical field

[0001] The present invention relates to methods for quantifying the impression of volume conferred by a person's hair and / or highlighting the effects of at least one cosmetic treatment and / or stimulus on this impression of volume. Prior art

[0002] The perception of a person's hair volume, which can also be described as the "impression of volume" conferred by said hair, is a concept difficult to characterize objectively. The impression of volume depends on several criteria affecting various areas of the hair, which are sometimes perceived differently by observers.

[0003] Current image processing methods known from the prior art make it possible to approximate the exact volume, in the mathematical sense of the term, of hair that an individual possesses, but this exact volume cannot be confused with the impression of volume.

[0004] Many cosmetic treatments seek to increase or decrease the impression of volume given by hair, but their effects are difficult to demonstrate in the absence of objective measurement of the impression of volume.

[0005] There is therefore a need for processes that allow for the objective quantification of the impression of volume conferred by an individual's hair and for highlighting the effects of at least one cosmetic treatment and / or stimulus on this impression of volume. Summary of the invention

[0006] The invention aims to meet this need and thus relates, according to one of its aspects, to a method for quantifying the impression of volume conferred by hair, comprising the steps of:

[0007] a) acquire, in particular using a digital acquisition device such as a camera, a front and / or profile and / or back image of the hair of a person to be assessed,

[0008] b) segment, during image processing performed on a computer or other computing device, the hair on each image,

[0009] c) define, in particular automatically by image processing:

[0010] - within the segmented hair on the frontal image at least one region of interest including hair roots and / or at least one area of ​​interest including hair lengths; -and / or within the segmented hair on the profile image, a region of interest comprising a mass of hair;

[0011] - and / or within the segmented hair on the back view image at least one region of interest including a portion of the hair

[0012] d) deliver, in particular automatically, a value of at least one characteristic parameter, extracted from one of the regions of interest or corresponding to the sum or average, possibly weighted, of intermediate parameters extracted from several regions of interest, the characteristic parameter being associated with a visual estimate of the volume conferred by the hair in the region(s) of interest.

[0013] The quantification method according to the invention makes it possible to objectively characterize the impression of volume conferred by the hair by means of at least one parameter. This parameter is described as "characteristic" because of its association with a visual estimation of the volume.

[0014] In one embodiment, the method can deliver a value for a single characteristic parameter. Alternatively, it can deliver a value for several characteristic parameters.

[0015] By using a front and / or profile and / or back image and possibly several regions of interest, the quantification process takes into account the multiplicity of criteria allowing the volume of a head of hair to be assessed.

[0016] The computer or other computing device is, for example, a personal computer, a smart phone, a server, a specialized circuit, and may include a graphics card.

[0017] The above value can be delivered by display on a screen and / or printed and / or sent as a message to a remote terminal. Picture

[0018] The method may include a step a0) preceding step a) consisting of placing the person in front of a plain background, preferably gray, and / or draping their torso with a piece of plain fabric, preferably gray, so that their hair falls over the fabric. This step facilitates the segmentation of the hair in the images.

[0019] The lighting conditions and / or acquisition parameters are preferably predefined to correspond to a standard image acquisition protocol. This facilitates the comparison of the values ​​of a characteristic parameter obtained for different hair types.

[0020] The front image can be acquired after styling the hair so that the hairline on the top of the forehead is visible and the hair, when its length allows, is brought back to the front of the person's torso.

[0021] The profile image can be acquired after bringing all of the person's hair back onto the back of their chest.

[0022] The image from behind can be acquired after bringing all of the person's hair back over the back of their torso.

[0023] The person's hair can be arranged so as to cover their ears. This arrangement of the hair over the ears helps to avoid distorting the volume of the hair. Arranging the hair behind the ears tends to flatten it artificially.

[0024] These positionings of the person's hair also contribute to the standardization of the image acquisition protocol.

[0025] Segmentation is a classic task in the field of image processing.

[0026] Hair segmentation can be implemented by algorithms inspired by conventional image processing methods and / or by algorithms based on artificial intelligence. Regions of interest

[0027] By "region of interest" is meant a part or sub-part of the hair segmented in an image. In some embodiments, the region of interest may encompass the entire segmented hair in an image.

[0028] At least one of the regions of interest can be delimited by at least one anatomical landmark of the person, in particular an anatomical landmark of the person's face.

[0029] By "delimited by at least one anatomical landmark", it is understood that a region of interest can be delimited by at least one horizontal line and / or at least one vertical line whose position is fixed by an anatomical point of the person, such as his eyes, his nose, his mouth, his ears or his shoulders among others, this list not being exhaustive.

[0030] By "region of interest including hair roots" is meant a region of interest encompassing a part of the person's head so as to include at least a part of the person's hair roots.

[0031] At least one, and preferably each, region of interest comprising hair roots may encompass the entire segmented hair extending between a substantially vertical line passing through the middle of the person's face and a horizontal line passing through an outer corner of one of the person's eyes.

[0032] Preferably, two regions of interest including hair roots are defined on either side of the person's face in the front image.

[0033] By "region of interest including lengths of hair" is meant a region encompassing a part of the hair not touching the person's skull and potentially extending to the tips of the hair.

[0034] At least one, and preferably each, region of interest comprising hair lengths may encompass the entire segmented hair extending from a horizontal line passing through an outer corner of one of the person's eyes to the hair tips.

[0035] Preferably, two regions of interest comprising hair lengths are defined on either side of the person's face in the front image.

[0036] In one embodiment, two regions of interest comprising hair roots and two regions of interest comprising hair lengths are defined on either side of the person's face in the front image such that each region of interest comprising hair roots is bordered at one end by a region of interest comprising hair roots and at the other end by a region of interest comprising hair lengths.

[0037] By "region of interest comprising a mass of hair" is meant a region defined on the profile image encompassing at least the free part of the hair detached from the skull of the person.

[0038] Preferably, the region of interest comprising a mass of hair encompasses the entire segmented hair on the profile image.

[0039] By "region of interest including a part of the hair" is meant a region defined on the image from the back encompassing a part of the hair segmented on that image.

[0040] At least one, and preferably each, region of interest comprising a part of the hair may encompass the entire segmented hair extending on one side of a substantially vertical line passing through the middle of the face.

[0041] Preferably, two regions of interest comprising a part of the hair are defined on either side of the substantially vertical line passing through the middle of the face.

[0042] The regions of interest can be defined automatically by implementing algorithms inspired by conventional image processing methods and / or by algorithms based on artificial intelligence, preferably using neural networks, in particular convolutional neural networks (CNNs).

[0043] Automatically defined regions of interest can be specified manually. Alternatively, the regions of interest can be defined entirely manually. Visual estimation

[0044] By "visual estimation," we mean that at least one observer, and preferably a panel of observers, has given their perception of the person's hair volume. The visual estimation can be modeled by an ordinal categorical variable. This variable can have, for example, five levels ranging from a very low impression of volume to a high impression of volume.

[0045] Visual estimation can be performed in each region of interest.

[0046] When the chosen characteristic parameter corresponds to the sum of intermediate parameters extracted from several regions of interest, the visual estimation can be carried out globally on the regions of interest concerned.

[0047] Visual estimation can be done by direct notation, using an atlas, which may have, for example, five levels, or by indirect notation such as pairwise comparisons.

[0048] The atlas may be specific to the region of interest concerned.

[0049] The visual assessment can be carried out by at least one observer other than the person whose hair is being assessed. Alternatively, the visual assessment can be carried out by the person whose hair is being assessed. In another embodiment, the visual assessment can correspond to the majority vote of several observers (i.e., the most frequently given score is retained), or to the average of their votes. Parameters

[0050] The expression "extracted parameter" of a region of interest is characteristic of the field of image processing. An extracted parameter of a region of interest is a parameter defined within the region of interest. It may be, for example, a distance between two specific points in the region of interest, but also a measurement such as height, width, area, or perimeter, calculated within the region of interest; this list is not exhaustive.

[0051] Certain parameters, such as the width of the hair at the shoulders, can only be defined for long hair. Therefore, hair length can restrict the number of parameters that can be defined within a region of interest.

[0052] By characteristic parameter "associated with a visual estimate", it is understood that a mathematical relationship exists between the characteristic parameter and the visual estimate.

[0053] The association between characteristic parameters and visual estimation may depend on hair length. Some parameters may be associated with visual estimation in short hair, but not have the same association in medium-length or long hair.

[0054] The characteristic parameter can be predictive of the visual estimation of the volume conferred by the hair in the region(s) of interest.

[0055] The characteristic parameter can be correlated with the visual estimate with a correlation coefficient of at least greater than 0.45, better greater than 0.5, better greater than 0.6, even better greater than 0.7, even better greater than 0.8, even better greater than 0.9.

[0056] The correlation coefficient can be calculated using Spearman's method. Alternatively, the correlation coefficient can be calculated using another method, for example using Pearson's method.

[0057] In an embodiment where at least one characteristic parameter is extracted from a region of interest comprising hair roots, the characteristic parameter can be chosen from an area E, El of the region of interest and a maximum height of the root separation L4, L4.1.

[0058] By "maximum height of root detachment" is meant a parameter L4, L4.1. consisting of the height between the highest point along a vertical axis Z of the region of interest including roots and a horizontal line extending from the highest point of the person's face.

[0059] In an embodiment where two regions of interest comprising hair roots are defined on either side of the person's face, at least one characteristic parameter may correspond to the sum or average, possibly weighted, of two identical intermediate parameters each extracted from one of the two regions, the intermediate parameter being chosen for example from an area E, E.1 of the region of interest and a height of the root separation L4, L4.1.

[0060] In an embodiment where at least one characteristic parameter is extracted from a region of interest comprising hair lengths, the characteristic parameter can be chosen from an area C, Cl of the region of interest comprising hair lengths, a hair width at the chin e3, e3.1 and a hair width at the mouth e2, e2.1.

[0061] The "hair width at the chin" corresponds to the length, measured in the region of interest comprising hair lengths, of a horizontal line extending from the lowest point along a vertical axis Z of the person's chin.

[0062] The "hair width at the mouth" corresponds to the length, measured in the region of interest comprising hair lengths, of a horizontal line extending from a corner of the person's lips.

[0063] In an embodiment where two regions of interest comprising hair lengths are defined on either side of the person's face, at least one characteristic parameter may correspond to the sum or average, possibly weighted, of two identical intermediate parameters, each extracted from one of the two regions, the intermediate parameter being chosen, for example among an area C, Cl of the region of interest comprising hair lengths, a hair width at the chin e3, e3.1 and a hair width at the mouth e2, e2.1.

[0064] In an embodiment where at least one characteristic parameter is extracted from a region of interest comprising a mass of hair, the characteristic parameter can be chosen from an area Ai of the region of interest comprising a mass of hair, a restricted area A2 of the mass of hair and a hair width d2 at the level of the opening of the ear canal.

[0065] In an embodiment where at least one characteristic parameter is extracted from a region of interest comprising a part of the hair, the characteristic parameter can be chosen from: an area A, Al of the region of interest, a maximum length L, Ll of the hair in the region of interest, a width of the hair d, dl at different positions. Stimulus and treatment

[0066] The stimulus according to the invention can be chosen from a chemical stimulus, a thermal stimulus, a mechanical stimulus, an electromagnetic stimulus, in particular light or microwave or a combination thereof.

[0067] The treatment is non-therapeutic. The treatment is a cosmetic treatment, in particular via the application of a cosmetic product promoting hair regrowth or beautification, in particular hair thickening.

[0068] The term “cosmetic product” means a product as defined in Directive 93 / 35 / EEC of 14 June 1993, amending Directive 76 / 768 / EEC.

[0069] The cosmetic product may be selected from rinse-off and leave-on products, styling products and hair lotions, shampoos, conditioners, hairsprays, anti-hair loss products, mousses, sprays, oils, among others, this list not being exhaustive.

[0070] The cosmetic treatment may include mechanical shaping, in particular by means of styling tools and / or devices associated or not with products, for example a hair dryer, a straightener, a styling brush, this list not being exhaustive.

[0071] By "reference stimulus and / or treatment" is meant a stimulus and / or treatment whose effects on an impression of volume conferred by hair, for example an effect leading to an impression of greater volume, are known from the state of the art and / or from previous experiments.

[0072] Methods for evaluating the effect of a stimulus and / or a treatment

[0073] According to another aspect of the invention, the invention also relates to a method for evaluating the effect of a stimulus and / or a treatment on the impression of volume conferred by hair, comprising the steps of:

[0074] a) quantify the impression of volume conferred by the hair using the quantification method as defined above,

[0075] b) repeat step a) on the hair after exposure to the action of the stimulus and / or treatment,

[0076] c) compare the values ​​of the characteristic parameters delivered at the two stages and generate at least one piece of information relating to the action of the stimulus and / or the processing on the impression of volume.

[0077] In a variant of the method for evaluating the effect of a stimulus and / or treatment previously described, step c) can be replaced by step c') consisting of delivering, in particular automatically, a value of at least one characteristic parameter, extracted from one of the regions of interest or corresponding to the sum or average, possibly weighted, of intermediate parameters extracted from several regions of interest, the characteristic parameter having a variation before and after exposure to said stimulus and / or treatment associated with the variation in visual estimation of the volume conferred by the hair before and after exposure, in particular correlated with a correlation coefficient, preferably calculated according to Spearman's method, at least greater than 0.7, better greater than 0.8, even better greater than 0.9.

[0078] The invention also relates, according to another aspect, to a method for evaluating the effect of a stimulus and / or a treatment on an impression of volume conferred by a head of hair, in which at least two heads of hair are exposed differently to the stimulus and / or the treatment, and in which the value of at least one characteristic parameter delivered by means of the quantification method as previously defined is compared for the different heads of hair.

[0079] The duration of exposure to the stimulus may vary between hairs. The treatment may be applied at different doses between hairs.

[0080] One of the hairs may not be exposed to the stimulus and / or treatment, unlike the other or other hairs. Processing method

[0081] The invention also relates to a cosmetic hair treatment process for a person, comprising the steps of: a. To quantify the impression of volume conferred by the hair using the quantification method as defined above, b. Select a hair treatment, including a treatment involving the application of a cosmetic product and / or styling mechanical by means of styling tools, depending on the result of the quantification in a) and a target volume printing to be achieved, c. carry out said treatment.

[0082] Preferably, in the case where the treatment involves the application of a cosmetic product, the selection of the cosmetic product is made from a database which lists for each of the products the effect of the product on the impression of volume.

[0083] Preferably, the selection of the cosmetic product is also carried out by the user entering a choice of the desired volume printing objective.

[0084] The selection process may include the steps of determining which product best achieves the desired impression of volume, given the individual's current hair volume. For example, if the individual wants hair that gives the impression of high volume, but their hair currently gives very little, a product with a strong volumizing effect could be selected. Conversely, if the individual wants to maintain their current volume and their hair already gives a high impression of volume, a product with a neutral effect on the impression of volume could be selected. Similarly, for an individual wishing to reduce their perceived volume, an appropriate product or styling method could be selected.

[0085] Finally, for a person wishing for an impression of medium volume, products with a low volumizing effect or, on the contrary, reducing the volume somewhat, can be chosen according to the initial impression of volume. System

[0086] The invention also relates, according to another aspect, to a system for implementing a method according to the invention, comprising: - an image acquisition unit, for example a camera, - a computing means, in particular a unit capable of storing the images and / or sending them to an embedded computer or a remote computing server via a wired or wireless connection, this computing means being configured to segment the hair on the image(s), define at least one region of interest and deliver a value of at least one characteristic parameter extracted from one of the regions of interest or corresponding to the sum or average, possibly weighted, of intermediate parameters extracted from several regions of interest.

[0087] The image acquisition unit can be separated from the rest of the system and connected by a wired or wireless connection.

[0088] The computing means may include a computer integrated into the system, as mentioned above, such as a microcomputer, or be remote such as a computing server connected to the system by a wired or wireless connection.

[0089] The computing means can be a computer, a smart object, for example a smart mirror, a smartphone,

[0090] The calculation means can be integrated into any image capture device for automatic hair volume measurement, for example into the following existing devices: “Canfield scientific Vectra”, “Cosderma photographic bench”, “Headscan and Lrobot from Orion Technolab”.

[0091] The user interface may be included in the system or separate, such as a display screen.

[0092] The system can be miniaturized for use at a point of advice or sale, a hair salon or a user's home.

[0093] The system can be used to recommend products, devices and / or services or for diagnosing hair volume. It can also be used to monitor hair volume, particularly in relation to treatments and / or stimuli applied to the hair and the impact of the exposome on the hair.

[0094] Such monitoring can be used to measure the progress of mastery of treatment techniques, for example of styling or shaping, by professionals or users. Brief description of the drawings

[0095] The invention will be better understood upon reading the following description, non-limiting examples of its implementation, and upon examination of the accompanying drawing, in which:

[0096] [Fig-1] represents in a schematic view different stages of an example of a quantification method according to the invention,

[0097] [Fig.2] represents a profile view of a user's hair and a system for acquiring and processing the profile view,

[0098] [Fig.3] is a front view image of the hair of the person in [Fig.2],

[0099] [Fig.4] is the same front view image as [Fig.3] on which are represented different parameters,

[0100] [Fig.5] is a correlation matrix between parameters relating to regions comprising hair roots defined in Figures 3 and 4, and a visual estimation of the impression of volume,

[0101] [Fig.6] is a set of three correlation matrices between parameters relating to regions comprising hair lengths defined in Figures 3 and 4 and a visual estimation of the impression of volume, each matrix representing a situation where the person's hair is of a different length,

[0102] [Fig.7] is the same profile image as [Fig.2] on which different parameters are represented,

[0103] [Fig.8] is a set of three correlation matrices between parameters relating to a region comprising a mass of hair defined in Figures 2 and 7 and a visual estimate of the impression of volume, each matrix representing a situation where the person's hair is of a different length,

[0104] [Fig.9] is a back view image of the hair of the person in [Fig.2],

[0105] [Fig. 10] is a correlation matrix between variations before and after exposure to a stimulus and / or processing of parameters relating to regions including hair roots defined in Figures 3 and 4, and a pre- and post-exposure variation of a visual estimate of the impression of volume,

[0106] [Fig. 11] is a correlation matrix between pre- and post-exposure variations to a stimulus and / or processing of parameters relating to regions comprising hair lengths defined in Figures 3 and 4, and a pre- and post-exposure variation of a visual estimate of volume impression,

[0107] [Fig. 12] is a correlation matrix between pre- and post-exposure variations to a stimulus and / or processing of parameters relating to a region comprising a mass of hair defined in Figures 2 and 7, and a pre- and post-exposure variation of a visual estimate of the impression of volume. Detailed description

[0108] With reference to figures 1 to 11, we will describe different stages of a method for quantifying the impression of volume conferred by hair according to the invention.

[0109] Step a0 of [Fig.1] aims to create standard acquisition conditions and may consist of placing, as illustrated in [Fig.2], a person 1 in front of a plain background 2 and draping their torso with a piece of plain fabric 3 so as to let their hair 4 fall over the piece of fabric.

[0110] In step a), all of the person's hair 4 is brought back onto the back of their torso, the hair arranged so as to cover at least part of the ears, as seen in [Fig. 2] a). A profile image is acquired, using any digital image acquisition device 9, for example a color camera with a CMOS sensor. A rear image, as seen in [Fig. 9], is then acquired.

[0111] Person 1's hair 4 is then styled so that the hairline on the upper forehead is visible and the hair is brought back onto the front of the person's bust as illustrated in [Fig.3]. A front image of the hair 4 is acquired using the digital image acquisition device 9.

[0112] In a step b), the previously acquired images are processed by a computer 10 of type PC for example; or any other suitable computing means, as shown in [Fig.2] b), and the hair 4 is segmented by means of a segmentation algorithm, for example an algorithm using convolutional neural networks (CNNs).

[0113] In step c), a region of interest comprising a mass of hair 6 is defined on the profile image, for example automatically by image processing, or alternatively defined by an operator using a tool for drawing a contour on the image. This region encompasses all the segmented hair on the profile image, as shown in [Fig. 2]. On the back image, as seen in [Fig. 9], two regions of interest 11 and 12 comprising a portion of the hair are defined on either side of the substantially vertical line passing through the middle of the face 5.

[0114] In step c), two regions of interest comprising hair roots 7 are defined on the frontal image, shown in [Fig. 3], on either side of the person's face. These two regions 7 encompass the entire segmented hair extending between a substantially vertical line 5 passing through the middle of the face and a horizontal line passing through an outer corner of one of the person's eyes.

[0115] Two regions of interest comprising hair lengths 8 are also defined on the frontal image on either side of the person's face. These two regions 8 encompass the entire segmented hair extending from a horizontal line passing through an outer corner of one of the person's eyes to the tips of the hair.

[0116] A set of parameters can be extracted by calculation using a computer 10 or any other suitable computing means, from regions 7 and 8 as shown in [Fig. 4], the list below not being exhaustive, the extraction relating to all or part of the listed parameters:

[0117] A: the area of ​​the upper part of the hair 4 on the right starting at the outer corner of the eye and stopping at the middle parting 5,

[0118] A1: the area of ​​the upper part of the left hair starting at the outer corner of the eye and stopping at the middle parting 5,

[0119] At: the sum of A and Al,

[0120] E: the area of ​​the region of interest comprising hair roots 7 located to the right of the person's face,

[0121] El: the area of ​​the region of interest comprising hair roots 7 located to the left of the person's face,

[0122] And: the sum of E and El,

[0123] C: the area of ​​the region of interest comprising hair lengths 8 located to the right of the person's face,

[0124] Cl: To make the area of ​​interest including hair roots 8 located to the left of the person's face,

[0125] Ct: the sum of C and Cl,

[0126] L5: positioned midway between the outer corner of the eye and the middle parting, L5 illustrates the length between the base of the hair and the crown of the hair on the upper right forehead,

[0127] L5.1: positioned midway between the outer corner of the eye and the middle line, L5.1 illustrates the length between the base of the hair and the crown of the hair on the upper left forehead,

[0128] L5m: the average between L5 and L5.1,

[0129] L4: the maximum height of the root detachment on the right,

[0130] L4.1: the maximum height of the root detachment on the left,

[0131] L4m: the average between L4 and L4.1,

[0132] el: the width to the right of the hair at eye level,

[0133] el. 1: the width on the left side of the hair at eye level,

[0134] elt: the sum of el and el.l,

[0135] e2: the width to the right of the hair at the level of the mouth,

[0136] e2.1: the width to the left of the hair at the level of the mouth,

[0137] e2t: the sum of e2 and e2.1,

[0138] e3: the width to the right of the hair at the level of the chin,

[0139] e3.1: the width to the left of the hair at chin level,

[0140] e3t: the sum of e3 and e3.1,

[0141] e4: the width to the right of the hair at shoulder level,

[0142] e4.1: the width on the left side of the hair at shoulder level,

[0143] e4t: the sum of e4 and e4.1,

[0144] e5: the width to the right of the hair at chest level,

[0145] e5.1: the width on the left side of the hair at chest level,

[0146] e5t: the sum of e5 and e5.1,

[0147] e6: the width to the right of the hair immediately above the hair tips,

[0148] e6.1: the width to the right of the hair immediately above the tips of the hair,

[0149] e6t: the sum of e6 and e6.1,

[0150] Standard parameters such as facial parameters, particularly those relating to its shape, are also annotated on the frontal image to serve as elements of comparison in the correlation matrices:

[0151] L2: the length of the hair between a horizontal line passing through an outer corner of the person's eye and the tips of the hair,

[0152] L1:1a total length of hair from the top of the head to the tips

[0153] b: half the length of the face or major radius of the face ellipse,

[0154] a: half the width of the face or minor radius of the face ellipse,

[0155] B: the area of ​​the face.

[0156] In the correlation matrices of Figures 5, 6, and 8, v is the visual estimate of the volume conferred by the hair in the region(s) of interest. v can be modeled by a multilevel ordinal categorical variable, for example, with 5 levels, ranging from a very low impression of volume to a high impression of volume. The visual estimation could be carried out by a panel of observers, for example, lay observers, consumers, or experts in the field, by comparing sets of images pairwise, or using several atlases, for example, three, in self-assessment or image evaluation, each specific to one of the regions comprising hair roots, hair lengths, or hair mass. The correlation coefficients for all the matrices were calculated using Spearman's rank correlation coefficient.

[0157] The correlation matrix in [Fig. 5] focuses on the parameters extracted from the regions of interest 7, which include hair roots, or on the sum or mean of these parameters, which then serve as intermediate parameters. The parameters with a correlation coefficient greater than 0.45 with the visual estimate are the areas E and E1 of the regions of interest and the maximum heights of the root detachment L4 and L4.1, as well as the respective sums of these parameters E1 and L4.1.

[0158] The correlation matrices of [Fig.6] are interested in the parameters extracted from the regions of interest 8 comprising hair lengths or in the sum of these parameters, which then act as intermediate parameters.

[0159] The matrix in [Fig. 6] a) relates to short hair, so not all parameters are defined. The matrix in [Fig. 6] b) relates to medium-length hair, with more parameters than the matrix in [Fig. 6] a). The matrix in [Fig. 6] c) relates to long hair, so an even greater number of parameters are defined.

[0160] Depending on the length of the person's hair, the strength of the correlation between similarly defined parameters and visual estimation varies. The parameters with the highest correlation coefficients are the C and C areas of the regions of interest comprising hair lengths and hair widths at the level of the chin e3, e3.1 and the hair widths at the mouth e2, e2.1, as well as the respective sums of these parameters: Ct, e3t and e2t.

[0161] On the profile image, another set of parameters is extracted from the region of interest comprising a mass of hair 6, as shown in [Fig.7], the list below not being exhaustive, the extraction relating to all or part of the listed parameters:

[0162] A1: the area of ​​the region of interest comprising a mass of hair,

[0163] A2: the restricted area of ​​the hair mass,

[0164] dl: the width of the hair at the middle of the forehead,

[0165] d2: the width of the hair at the level of the opening of the ear canal

[0166] d3: the width of the hair at the level of the bottom of the mouth,

[0167] d4: the width of the hair at shoulder level,

[0168] d5: the width of the hair at chest level,

[0169] d6: the width of the hair immediately before the hair tips.

[0170] Similar to [Fig.6], the matrices of [Fig.8] concern lengths of different hair types. The parameters with the highest correlation coefficients with the visual estimate are the area Ai of the region of interest including a mass of hair, the restricted area A2 of the mass of hair and the width of the hair d2 at the level of the opening of the ear canal.

[0171] On the image from the back, another set of parameters is extracted from each of the two regions of interest 11 and 12 comprising a part of the hair, as shown in [Fig.9], the list below not being exhaustive, the extraction relating to all or part of the listed parameters:

[0172] A, Al: the area of ​​the region of interest,

[0173] L, L1: the length of the region of interest,

[0174] dl, dl.l to d6, d6.1: the width of the hair in each region of interest at different positions.

[0175] The correlation matrices in Figures 10 to 12 illustrate another aspect of the invention, the method for evaluating the effect of a stimulus and / or a treatment on an impression of volume conferred by the hair.

[0176] The correlation matrices of figures 10 to 12, concerning respectively the regions comprising hair roots 7, the regions of interest comprising hair lengths 8 and the region of interest comprising a mass of hair 6.

[0177] During the evaluation process, the impression of volume conferred by a person's hair is quantified using the previously described method before and after exposure to a stimulus and / or treatment. The association between the pre- and post-exposure variations in the parameter values ​​provided by the quantification method and the pre- and post-exposure variations in the visual estimation is then examined.

[0178] For reasons of simplicity, the variations of a parameter are designated by the name of the parameter in the legends of these matrices.

[0179] The invention is not limited to the examples described above.

[0180] Other regional boundaries may be established.

[0181] Other parameters or sums, possibly weighted, of parameters may be defined.

[0182] The selection thresholds for characteristic parameters may vary.

Claims

Demands

1. A method for quantifying the impression of volume conferred by hair (4), comprising the steps of: a. to acquire, in particular by means of a digital acquisition device (9) such as a camera, a front and / or profile and / or back image of the hair of a person (1) to be assessed, b. to segment, during image processing carried out on a computer or other computing means (10), the hair on each image, c. define, in particular automatically through image processing: - within the segmented hair on the front image at least one region of interest including hair roots and / or at least one region of interest including hair lengths; - and / or within the segmented hair on the profile image, a region of interest comprising a mass of hair; - and / or within the segmented hair on the back view image, at least one region of interest including a portion of the hair d. deliver, in particular automatically, a value of at least one characteristic parameter, extracted from one of the regions of interest or corresponding to the sum or average, possibly weighted, of intermediate parameters extracted from several regions of interest, the characteristic parameter being associated with a visual estimate v of the volume conferred by the hair in the region(s) of interest.

2. A method of quantification according to the preceding claim, the characteristic parameter being correlated with the visual estimate v with a correlation coefficient of at least greater than 0.45, better greater than 0.5, better greater than 0.6, even better greater than 0.7, even better greater than 0.8, even better greater than 0.

9.

3. A method of quantification according to the preceding claim, the correlation coefficient being calculated using Spearman's method.

4. A method of quantification according to any one of the preceding claims, at least one of the regions of interest being delimited by at least one anatomical landmark of the person, in particular an anatomical landmark of the person's face.

5. A method of quantification according to the preceding claim, each region of interest comprising hair roots (7) encompassing the entire segmented hair extending between a substantially vertical line passing through the middle of the person's face (1) and a horizontal line passing through an outer corner of one of the person's eyes.

6. A method of quantification according to the preceding claim, wherein at least one characteristic parameter is extracted from a region of interest comprising hair roots (7), the characteristic parameter being chosen from an area E, E.1 of the region of interest and a maximum height of the root detachment L4, L4.

1.

7. A quantification method according to one of the two immediately preceding claims, wherein two regions of interest comprising hair roots (7) are defined on either side of the person's face, at least one characteristic parameter corresponding to the sum or average of two identical intermediate parameters extracted each from one of the two regions, the intermediate parameter being chosen for example from an area E, E.1 of the region of interest and a height of the root detachment L4, L4.

1.

8. A method of quantification according to any one of claims 4 to 7, each region of interest comprising hair lengths (8) encompassing the entire segmented hair (4) extending from a horizontal line passing through an outer corner of one of the person's eyes to the tips of the hair.

9. A method of quantification according to the preceding claim, wherein at least one characteristic parameter is extracted from a region of interest comprising hair lengths (8), the characteristic parameter being chosen from an area C, Cl of the region of interest comprising hair lengths, a hair width at the chin e3, e3.1 and a hair width at the mouth e2, e2.

1.

10. A quantification method according to one of the two immediately preceding claims, wherein two regions of interest comprising hair lengths (8) are defined on either side of the person's face, at least one characteristic parameter corresponding to the sum or average of two identical intermediate parameters extracted each from one of the two regions, the intermediate parameter being chosen for example from an area C, Cl of the region of interest comprising hair lengths, a hair width at the chin e3, e3.1 and a hair width at the mouth e2, e2.

1.

11. A quantification method according to any one of the preceding claims, the region of interest comprising a mass of hair (6) encompassing the entire segmented hair on the profile image.

12. A method of quantification according to the preceding claim, at least one characteristic parameter being extracted from a region of interest comprising a mass of hair (6), the characteristic parameter being chosen from an area Ai of the region of interest comprising a mass of hair, a restricted area A2 of the mass of hair and a hair width d2 at the level of the opening of the ear canal.

13. A method for evaluating the effect of a stimulus and / or a treatment on the impression of volume conferred by hair (4), comprising the steps of: a. quantifying the impression of volume conferred by the hair by means of the quantification method according to any one of the preceding claims, b. repeating step a) on the hair after exposure to the action of the stimulus and / or the treatment, c. comparing the values ​​of the characteristic parameters delivered at the two steps and generating at least one piece of information relating to the action of the stimulus and / or the treatment on the impression of volume.

14. A method for evaluating the effect of a stimulus and / or treatment on the impression of volume conferred by a head of hair (4), wherein at least two heads of hair are exposed differently to the stimulus and / or treatment, and wherein the value of the stimulus and / or treatment is compared less a characteristic parameter delivered by means of the quantification method according to any one of claims 1 to 13, for the different hair types.

15. A method according to one of the two immediately preceding claims, the treatment being a cosmetic treatment, in particular via the application of a product promoting hair regrowth or beautification, in particular hair thickening.

16. A cosmetic hair treatment method for a person, comprising the steps of: a. quantifying the impression of volume conferred by the hair by means of the quantification method according to any one of claims 1 to 13, b. selecting a hair treatment, in particular a treatment comprising the application of a cosmetic product and / or mechanical shaping by means of styling tools, according to the result of the quantification in a) and a target of impression of volume to be achieved, c. carrying out said treatment.

17. A system for implementing a method according to any one of the preceding claims, comprising: - an image acquisition unit, for example a camera, - a computing means, in particular a unit capable of storing the images and / or sending them to an integrated computer or to a remote computing server via a wired or wireless connection, this computing means being configured to segment the hair on the image(s), define at least one region of interest and deliver a value of at least one characteristic parameter extracted from one of the regions of interest or corresponding to the sum or average, possibly weighted, of intermediate parameters extracted from several regions of interest.

Images