FR2917830A1 - Film brightness evaluating method for e.g. lipstick, involves calculating brightness index perceived from index representative of brightness intensity and another index representative of brightness heterogeneity of film - Google Patents

Abstract

The method involves measuring an index representative of an intensity of brightness, which is not diffused from a part of a film of a cosmetic product, and determining another index representative of heterogeneity of the brightness of the film. A brightness index perceived from the two indices is calculated, where one of the indices is calculated from a number, size, form, spatial distribution and distribution of intensity of brightness zones on the film. The film of product is applied to a keratinous material and lips. Independent claims are also included for the following: (1) a system for evaluating brightness perceived by a human observer of a film of a cosmetic product (2) a method for evaluation of wet brightness of a film of a cosmetic product.

Description

The present invention relates to methods and systems for

  to evaluate the perceived brilliance, that is to say the brilliance that is representative of the way the human perceives it. One of the properties that the cosmetics industry is trying to improve in the 5 consumer product offerings is brilliance. This is particularly the case of lipsticks, nail polish or hair products. There is a need to measure the perceived brilliance of these products on the lips, nails or hair. No definition or existing method can satisfactorily characterize the brilliance of the cosmetic products so that it accounts for the brightness perceived by the consumer. Existing gloss measurement methods generally involve measuring, when illuminating a surface, the amount of light reflected by that surface in different directions. This measurement can be done using optical instruments of the goniometric type, for example gloss meter or gonioreflectometer, which measure quantities such as specular reflection, diffusion indicator, etc. These methods are rather reserved for flat surfaces because they require geometric referencing difficult to achieve in vivo. Measurement of the brightness can also be done by imaging, for example by polarimetric imaging, for example by measuring the polarization rate of the reflected light. The imaging methods have the advantage of being able to choose in the image the measurement zone for which the light source, the surface to be evaluated and the camera are in a known measurement configuration, for example specular reflection, and to quantify the intensity of the reflected light by image analysis. However, the Applicant has found that the brightness measured from these parameters does not always make it possible to account for the differences in brightness perceived between the products. The invention aims to provide a method for evaluating the brightness perceived by the human observer.

  According to one of its aspects, the subject of the invention is a method for evaluating the brightness perceived by a human observer of a film of a cosmetic product, comprising the steps of, the film being illuminated by at least one source of light: - measuring a first index representative of the intensity of non-diffuse brightness of at least a portion of the film, - determining a second index representative of a heterogeneity of brightness of the film, - calculating a brightness index perceived at from the first and second indices at least.

  The Applicant has discovered that the perception of brightness strongly depends on the ability of the product film to reflect more or less clearly the image of the light source or sources present, whether they are point or semi-extended. This property can be quantified by characterizing the heterogeneity, that is the level of fragmentation, of the observable brightness zones.

  The invention can make it possible to quantify the brightness of a cosmetic product according to the perception and can also make it possible to decompose the visual effects contributing to the perception of brightness so as to offer several ways of formulation to improve it. The second index may be calculated from at least the spatial distribution of gloss areas on the film and / or from at least the brightness area intensity distribution of the film. The second index can be calculated from the number, size, shape, spatial distribution and intensity distribution of glossy areas of the film. The second index can be calculated by image analysis at least from a segmentation of gloss areas of the film. The second index can be evaluated from at least one image representative of the light non-diffusely reflected by the film. The representative image of light non-diffusely reflected by the film can be obtained by polarimetric imaging.

  The method may comprise the step of acquiring an image (P) in parallel polarization, the representative image of the non-diffuse reflected light being obtained at least from this image (P).

  The method may comprise the step of acquiring a cross-polarized image (C), the representative image of the diffuse and / or backscattered light being obtained at least from this image (C). The representative image of the non-diffusely reflected light of the film can be obtained by subtracting two superimposed images acquired one with a parallel polarization and the other with a cross polarization. A grayscale histogram can be generated from an image representative of light non-diffusely reflected by the film. The first index can be calculated from at least one gray level of the lightest pixels of the image representative of the light non-diffusely reflected by the film. The brightest pixels can be the last 0.5% of the pixels in the histogram, for example. The second index can be evaluated by comparison for example with reference images of an atlas.

  The product film can be applied to keratin materials, for example the lips. The product film can be illuminated by at least one polarized artificial light source, better by two polarized artificial light sources arranged substantially symmetrically with respect to a median plane for the film.

  The invention also relates to a system for evaluating the brightness perceived by a human observer of a film of a cosmetic product, comprising: first means for measuring a first index representative of the non-diffuse brightness of at least a portion of the film, second means for generating an image for determining a second index representative of a brightness heterogeneity of the film, third means for calculating a perceived brightness index from the first and second indices at least. The first means may include an acquisition bank comprising at least one polarized light source and a camera for acquiring at least one image in cross polarization or parallel. The second means may comprise at least one computer system for adding or subtracting two images acquired respectively in parallel polarization and in cross polarization and analyzing an image representative of the non-diffuse brightness. The third means may include a computer system that may be the same as above or another computer system.

  The subject of the invention is also a method for evaluating the wet gloss of a film of cosmetic product, comprising the steps of: generating an image representative of the light non-diffusely reflected by the surface of the film; determine from an evaluation of the heterogeneity of gloss of the film on this image a wet gloss index. The evaluation of the heterogeneity of brightness can be performed by comparison with reference images of an atlas. The evaluation of gloss heterogeneity can also be evaluated by automatic image analysis.

  In an exemplary implementation of the invention, the method for measuring the perceived brightness comprises the steps of: - acquiring by polarimetric imaging two images of keratin materials, in particular lipsticks, made up with a cosmetic product, one of (C) in cross-polarized configuration and the other (P) in parallel polarized configuration, -difference of the two images to obtain a resultant image representative of light non-diffusely reflected by the film surface, resulting image (PC) a region of interest to be analyzed, in particular a rectangle encompassing the lips, -generate a histogram of gray levels on the region of interest, -calculate a first index representative of the average gray level of the pixels the more clearly, especially the last 0.5% of the histogram, -evaluate a second index representative of the heterogeneity of brightness by visually scoring the image (PC) using an atlas or by image analysis, in particular by segmenting the region of interest of the areas of brightness, - determining a perceived brightness index from the first and second indices, in particular by summing at least the first and second indices.

  The subject of the invention is also a method for promoting the sale of a cosmetic product, comprising the step of reporting information representative of the perceived brightness, measured by the process as defined above, or even to be done state of information representative of the wet gloss.

  The information in question may be a numerical value or an identifier representative of a class of perceived or wet brightness values. This information may appear on a product packaging or on any material or digital media associated with the product, such as a user's manual or advertisement.

  The invention will be better understood on reading the following detailed description, non-limiting examples of implementation thereof, as well as on examining the appended drawing, in which: FIG. 1 schematically shows in perspective an acquisition bank for acquiring images of the cosmetic product film; - FIG. 2 illustrates the generation of an image representative of the non-diffusely reflected light by subtraction of an image by parallel polarization and a cross-polarization image; FIGS. 3A to 3D are examples of delimitation of a region of interest; FIG. 4 illustrates the transformation of the representative image of the reflected light by processing of 5A, 5B, 6A, 6B, 7A and 7B schematically illustrate different appearances of gloss depending in particular on the surface state and the nature of the composition; FIGS. 8A to 8G show show different images of an exemplary atlas for evaluating the heterogeneity of the brightness zones, FIGS. 9A to 9C represent different images that can be used to calculate the wet gloss, FIGS. 10A to 10C and 11A to 11C. 9A to 9C are analogous images for other cosmetics, and FIGS. 12 to 15 illustrate different examples of cosmetic film glosses.

  FIG. 1 shows a bench 1 for acquiring images of a film of a cosmetic product, for example a film of a product to be applied to the skin, the lips or the keratinous fibers, for example the hair. This bench 1 is part of a perceived brightness evaluation system 5 comprising a computer system, such as a microcomputer for example, to perform image processing. The acquisition bench 1 comprises in the example in question a polarimetric vision system intended for the in vivo measurement of the brightness of a part of the face, in this case the lips. This system is arranged to successively acquire a parallel polarization image P carrying the diffuse reflection and non-diffuse reflection information and a cross-polarized image C with the diffuse reflection information. A representative image of the non-diffuse brightness PC can be extracted by difference between these two images, as illustrated in FIG. 2. In order to acquire the images P and C, the subject is illuminated with polarized light using two fluorescent projectors 3 of the same intensity, equipped with linear polarizing filters, located on either side of a set 4 comprising a camera, for example monochrome, and a liquid crystal analyzer whose polarization axis switches to the video rate in two perpendicular directions. The positioning of the model can be done using a chinrest 6. The system can be calibrated with one or more brightness standards, the calibration can be a function for example of the nature of the product whose gloss is to be measured. , the brightness intensity may be different depending on the products, being for example stronger for a product to be applied to the lips than for a foundation. Once the resulting P-C image is generated, a region of interest can be defined, as illustrated in one of FIGS. 3A-3C. This region of interest is, for example, a contour determined automatically by image analysis, for example of a polygonal contour including the lower lip, as illustrated in FIG. 3A, of rectangular contour including the upper and lower lips, as illustrated in FIG. Figure 3B, or may be manually drawn to include, for example, the lower lip and the upper lip as shown in Figure 3C.

  Once the region of interest is defined, a histogram of the gray levels on it can be generated in order to calculate the average gray level of the lightest pixels, for example the last 0.5% of the pixels of the histogram . This average value can define a first index representative of the non-diffuse brightness of the film.

  According to one aspect of the invention, a second index representative of the brightness heterogeneity of the film can be determined. This second index can be determined on a region of interest which is for example of rectangular shape framing the lower lip, as shown in Figure 3D, within the larger rectangle defining the region used to calculate the first index. The non-diffuse brightness image PC may be subjected to a treatment intended to facilitate the evaluation of the heterogeneity of the brightness zones, as illustrated in FIG. 4. This treatment may in particular aim to eliminate the noise and to perform a thresholding and then a classification of the gloss spots that appear on the image according to their clarity, size, shape, etc. The image processing may be useful especially in the case where the index representative of the heterogeneity of brightness is determined by image analysis, automatically. In the case where the heterogeneity of the brightness is evaluated by comparison with reference images of an atlas, such image processing may not be performed. Depending on the nature of the product applied, the gloss as it appears on the P-C image may be more or less heterogeneous, as shown in Figures 5B, 6B and 7B. FIG. 5B corresponds to the non-diffuse brightness of a film of product F which has a smooth surface S, favorable to a glassy reflection with a weak diffusion of light. In this example, the product is the lip gloss sold under the reference B. LOLITA by the company AGNES B, which makes it possible to obtain, as illustrated in FIG. 5A, a smooth surface favorable to a continuous glassy reflection with spots of relatively large extent. big. FIG. 6A illustrates the case where the product film, for example of the JUICY WEAR brand marketed by LANCAME, does not perfectly smooth the surface on which it is applied, so that part of the light is diffused, this which gives rise to a discontinuous vitreous reflection with fragmented spots on the image and a moderate contrast. FIG. 7A illustrates the case where the product film, for example of the GLAMSHINE brand marketed by L'OREAL PARIS, comprises reflecting particles capable of creating highlight points which reflect light in multiple directions. Such a product can lead to a glassy reflection with intense point reflections on the particles, resulting in the presence of spots of both relatively large extent and small extent, and a relatively low contrast. In an exemplary implementation of the invention, the second index, representative of the heterogeneity of the brightness, also called wet gloss index, can be evaluated by comparing the PC image representative of the non-diffuse brightness with atlas reference images such as those illustrated in Figures 8A-8G.

  This comparison can be performed by an experienced observer or alternatively by an image comparison software. Once the second index has been determined, a perceived brightness can be determined, such as, for example, the sum of the first index and the second index. The second index can still be determined by image analysis, especially on an image processed to remove the noise and perform a thresholding and classification of gloss spots, as mentioned above. In an exemplary implementation of the invention, this second index is determined by a formula which is for example a function of the size and contrast of the gloss zones.

  For example, the image processing may include the detection of high-area spots, for example with an area greater than 100 pixels, and the corresponding total area Ab may be measured. The treatment may also include the detection of small spots, for example less than 20 pixels and the corresponding total area As can be measured.

  The wet gloss index may for example be defined by the formula 40 log [(1 + Ab) / (1 + As)].

  Figure 9B corresponds to the P-C image when the product film is B. LOLITA lip gloss. Figure 9B shows the image of Figure 9A after processing to reveal the extended spots. In the example considered, the total area Ab is equal to 1843. FIG. 9C represents the spots of small extent, the corresponding total area As being equal to 8, which leads to a wet gloss index of 93. In the example of Figure 10A, the product film is JUICY WEAR reference. FIG. 10B corresponds to the extended spots, of sum Ab equal to 1330 in the example under consideration. FIG. 10C corresponds to the spots of small extent, whose total area As is equal to 40, which leads to a wet gloss index equal to 61.

  Figure 11A corresponds to the application of GLAMSHINE lipstick. Figure 11B shows the largest spots with a total area of 3221, and Figure 11C are small patches with a total area of 592, resulting in a wet gloss of 24. L The invention may allow a classification of cosmetic products according to, for example, the value of perceived brightness. FIGS. 12 to 15 show four images corresponding to four different classes of wet gloss. The first class corresponds for example to a wet gloss index of between 100 and 150, which may correspond, for example, to 2 to 4 extended spots of marked outline. The second class corresponds, for example, to a wet gloss index which ranges from 80 to 100, which may correspond, for example, to 4 to 10 spots of medium extent, of relatively marked outline, the spots slightly diffusing the light, the product being able to have small, bright particles where appropriate. The third class may correspond to a wet gloss index value ranging from 40 to 80, which may correspond to a multitude of small spots whose outlines are fuzzy, imparting an impression of diffuse brightness. Finally, the fourth class may correspond to a wet gloss index ranging from 0 to 40, which may correspond to a lower lip covered with small spots and giving a general impression of blur. The invention is not limited to the examples which have just been described.

  The representative image of the non-diffuse brightness can be generated other than by difference of the P-C images, for example by the operation (P-C) / (P + C) carried out pixel by pixel. The representative image of non-diffuse brightness can be obtained by colorimetric or spectral polarimetric imaging. In these cases, the chromatic or spectral characteristics of the image and in particular the brightness areas may be taken into account in calculating the different indices. In the case of high gloss, that is to say when the brightness areas are sufficiently contrasted with the diffuse background, it is possible to calculate the wet gloss index on non-polarized images generated under visible, ultraviolet light or infrared. The phrase having one must be understood as being synonymous with having at least one.

Claims (25)

  1. A method for evaluating the brightness perceived by a human observer of a film of a cosmetic product, comprising the steps of, the film being illuminated by at least one light source (3): measuring a first representative index of the intensity of the non-diffuse brightness of at least a portion of the film, - determining a second index representative of a brightness heterogeneity of the film, - calculating a brightness index perceived from the first and second indices at least.   2. Method according to claim 1, the second index being calculated from at least the spatial distribution of gloss zones on the film.   3. Method according to claim 1 or 2, the second index being calculated from at least the intensity distribution of brightness areas of the film.   4. Method according to any one of claims 1 to 3, the second index being calculated from the number, size, shape, spatial distribution and intensity distribution of brightness areas of the film.   5. Method according to any one of the preceding claims, the second index being calculated by image analysis at least from a segmentation of gloss zones of the film.   6. Method according to any one of the preceding claims, the second index being evaluated from at least one image representative of the light non-diffusely reflected by the film.   7. Method according to the preceding claim, the representative image of the light non-diffusely reflected by the film being obtained by polarimetric imaging.   8. Method according to claim 7, comprising the step of acquiring an image (P) in parallel polarization, the representative image of the non-diffuse reflected light being obtained at least from this image (P). . 30   The method according to claim 7 or 8, comprising the step of acquiring a cross-polarized image (C), the representative image of the non-diffuse reflected light being obtained at least from this image ( VS).   10. The method according to claims 8 and 9, the image (PC) representative of the light reflected by the non-diffuse film being obtained by subtracting two superposed images acquired one with a parallel polarization and the other with a cross polarization.   The method of any one of claims 1 to 10, wherein a greyscale histogram is generated from an image representative of light non-diffusely reflected by the film.   The method of claim 11, wherein the first index is calculated from at least one gray level of the brightest pixels of the image representative of the light non-diffusely reflected by the film.   13. The method of claim 12, the brightest pixels corresponding to the last 0.5% of the pixels of the histogram.   14. A method according to any one of the preceding claims, the second index being evaluated by comparison with reference images of an atlas. 15   15. Method according to any one of the preceding claims, the product film being applied to keratin materials.   16. The method as claimed in the preceding claim, the product film being applied to the lips.   17. A method according to any one of the preceding claims, wherein the product film is illuminated by at least one polarized artificial light source.   18. A method according to any one of the preceding claims, the film being illuminated by two polarized artificial light sources arranged substantially symmetrically with respect to a median plane for the film.   19. A system for evaluating the perceived brightness by a human observer of a film of a cosmetic product, comprising: first means for measuring a first index representative of the non-diffuse brightness of at least part of a film; second means for generating an image for determining a second indicia representative of a brightness heterogeneity of the film; - third means for calculating a perceived brightness index from the first and second indices at least.   20. The system of claim 19, the first means comprising an acquisition bank (1) comprising at least one polarized light source (3) and a camera for acquiring at least one image in cross polarization or parallel.   21. The system of claim 19 or 20, the second means comprising at least one computer system for adding or subtracting two images acquired respectively in parallel polarization and cross polarization and analyze an image representative of the non-diffuse brightness.   22. A method for evaluating the wet gloss of a cosmetic product film, comprising the steps of: - generating an image representative of the light non-diffusely reflected by the surface of the film, - determining from a evaluation of the gloss heterogeneity of the film on this image a wet gloss index.   23. The method of claim 22, the evaluation of the brightness heterogeneity being performed by comparison with reference images of an atlas.   24. The method of claim 22, the evaluation of brightness heterogeneity being evaluated by image analysis.   The method of claim 22, the representative image of the non-diffuse reflected light being generated by polarimetric imaging.