Method and system for characterizing keratinous fibers, in particular human eyelashes

1 . A method for characterizing eyelashes or eyebrows which is implemented by computer and comprises the following steps, which are directed toward:

receiving data corresponding to a set of pixels in a close-up image of a body area comprising a plurality of eyelashes, the entirety of a row of eyelashes or of an eyebrow, to be characterized;

applying at least one computer vision step so as to obtain, on the basis of the image data received, a numerical evaluation of at least one characteristic numerical parameter from among, notably, the number of fibers, an average length of the fibers, an average thickness of the fibers, and a curvature of the fibers,

said method being characterized in that it comprises the following additional steps:

identifying, in the image of the body area received, a facial feature having an average size which is fixed among a population,

determining a number of pixels which is associated with the size of the facial feature,

converting distances determined in number of pixels into metric distance,

determining a length of the fibers by determining a number of pixels from root to tip of each fiber, and

determining an average length on the basis of a number of fibers which is counted;

wherein in that the computer vision step comprises at least one step of identification by segmenting the fibers of the eyelashes or of the eyebrows;

wherein the segmentation step is followed by at least one step of computing the parameter on the basis of the pixels classified by segmentation.

2 . The method as claimed in claim 1 , characterized in that the computer vision step implements a regression method by applying a residual neural network, in particular of ResNet type.

3 . The method as claimed in claim 1 , characterized in that the computer vision step comprises a step of identification by segmenting roots or tips of the eyelashes or of the eyebrows.

4 . The method as claimed in claim 1 , characterized in that it comprises a step of generating an image highlighting the elements identified by segmentation in the image received.

5 . The method as claimed in claim 1 , characterized in that the number of fibers is determined by a pixel jumping method.

6 . The method as claimed in claim 1 , characterized in that it comprises an additional step of classifying the eyelashes or the eyebrows among at least two typologies established on the basis of at least one characteristic parameter of said eyelashes, on the basis of at least two parameters.

7 . The method as claimed in claim 1 , characterized in that it comprises a step of querying a database on the basis of at least one characteristic parameter determined, and notably on the basis of the typology determined, and of a desired result of treatment of the eyelashes or of the eyebrows which is chosen by a user, so as to determine at least one recommended cosmetic product (RP).

8 . The method as claimed in claim 7 , characterized in that it comprises a step of showing the one or more recommended products (RP) on a screen.

9 . A computer system for implementing a method as claimed in claim 1 , comprising:

at least one for importing or for acquiring a close-up image,

computing for implementing all or some of the processing steps and notably the computer vision steps and the numerical evaluation steps, and

at least one display, configured to allow the numerical evaluation results obtained to be displayed.

10 . The computer system as claimed in claim 9 , characterized in that all or some of the computing which are able to carry out the computer vision or numerical evaluation steps belong to a remote server which is distinct from a device incorporating the image importation or acquisition.