METHOD FOR ACQUIRING A PHOTOGRAPHIC PORTRAIT OF AN INDIVIDUAL AND INSTALLATION IMPLEMENTING THIS METHOD

Abstract

This method for acquiring photographic portraits of an individual consists of: – acquiring a biometric image of the individual's face using at least one digital camera (22); – concurrently with this acquisition, illuminating the individual's face with at least two distinct light sources, not centered on the face, and positioned laterally relative to the field of view (23) of the camera, said light sources being composed of light-emitting diodes (38) controlled by a PWM system. At least two acquisitions of the face are carried out, respectively: – under standard lighting, to allow the biometric acquisition of the face by the camera (22); – under differentiated lighting, that is to say, under which the at least two light sources (36, 37) generate a lighting effect on the face, also acquired by the camera.These two acquisitions are performed not simultaneously and randomly over a short period. The central control and processing unit is capable of processing these acquisitions and, after processing the acquisition under different lighting conditions, of determining the presence or absence of an artifact. Figure 1 for the abbreviation.

Description

Title of the invention: METHOD FOR ACQUIRING A PHOTOGRAPHIC PORTRAIT OF AN INDIVIDUAL AND INSTALLATION IMPLEMENTING THIS METHOD Scope of the invention

[0001] The present invention relates to the technical field relating to the acquisition of photographs of individuals, in particular within booth-type devices or apparatus.

[0002] The invention relates more particularly to the field of securing the acquisition of such photographs, in addition to optimizing the quality of the latter, in particular because of their intended use for identity or access documents. Prior state of the art

[0003] Facial photography equipment is widely used. Typically, it has been developed under the name "Photomaton," particularly in France, and consists of a booth enclosed by side walls and with access to the interior of the booth, traditionally concealed by a curtain or similar device. A device for acquiring the photograph is integrated into one of the walls; this device typically consists of a camera or video camera, while the individual's face is illuminated simultaneously with the acquisition of the photograph.

[0004] One of the challenges faced by designers of such installations lies in optimizing the lighting of the individual's face. More specifically, the aim is to avoid, as far as possible, "flattening" the image and to promote the distinction of the different raised areas of the face. Overexposure of certain parts of the face is also to be avoided.

[0005] Various solutions have been developed to overcome these drawbacks. In particular, document EP 2 860 582 describes an installation of the type in question, employing two complementary light-emitting surfaces positioned laterally to the field of view of the camera or photographic apparatus, and located between the wall incorporating said acquisition element and the object focal plane of said acquisition apparatus. This installation allows for homogeneous illumination of an individual's face, even when the individual is posing approximately relative to a reference position, resulting in high-quality photographic portraits, regardless of the morphology and positioning of the individual's face relative to the reference position of said installation.

[0006] Undoubtedly, the installation described in this document makes it possible to significa- actively improve the quality of the photographs thus obtained.

[0007] However, with the more specific objective of producing such photographs for use in identity or security documents, government and / or security services wish to eliminate any risk of identity theft, commonly referred to by the Anglo-Saxon term "anti-spoofing". Indeed, such identity theft, also known as a "biometric attack", can result, for example, from placing a raised mask on the face of the individual in question, but also from presenting, in the focal plane of the acquisition device, a portrait on a flat or curved medium representing an electronic print or display, which could thus either generate spoofed photographs or allow access to a secure location, thereby deceiving the databases in which authorized users are listed.This identity theft can also result from the digital association of at least two distinct faces, which has the effect of allowing the usurpation of each of the identities used in this association. This digital association is also known by the English term "morphing".

[0008] The invention therefore aims to overcome this difficulty and to combat these risks of usurpation. Summary of the invention

[0009] According to a first aspect, the process of acquiring photographic portraits of an individual consists of: - to acquire a biometric image of the individual's face using at least one digital camera, - concomitantly with this acquisition, to illuminate said face of the individual by at least two distinct light sources, not centered on said face, and positioned laterally in relation to the field of vision of said camera said light sources being made up of light-emitting diodes (LEDs) managed by a PWM system, (Anglo-Saxon acronym for "pulse width modulation", and also known by the French expression MLI for "Modulation de Largeur d'Impulse"); the PWM system and the operation of said at least one camera being managed by a central control and processing unit.

[0010] According to the invention, at least two acquisitions of said face are carried out, respectively: - according to standard lighting, that is to say according to which said at least two light sources are activated simultaneously so as to illuminate said face uniformly, in order to allow the biometric acquisition of said face by said at least one camera, that is to say of the view subsequently used for example for identity papers or access control media; - according to differentiated lighting, that is to say according to which the said at least two light sources generate a lighting effect at the level of the said face, also acquired by the said at least one camera; these at least two acquisitions being carried out not simultaneously and randomly over a short period of time; the central command and processing unit being capable of processing said acquisitions, and of determining after processing the acquisition according to differentiated lighting the presence or absence of an artifact.

[0011] For the purposes of this invention, the term "artifact" means the means of identity theft referred to in the preamble to this application, and generally any data that does not correspond to the real face of the individual whose photographic portrait one wishes to acquire.

[0012] Thus, the invention makes it possible to combat any attempt at identity theft, in a simple and rapid manner. Indeed, typically the total acquisition time lasts less than one second, and the two acquisitions are undetectable by the individual in question or by an advanced electronic system implemented by said individual, and therefore unpredictable, especially since said acquisitions occur randomly during this acquisition phase and in any order, that is to say that the acquisition under standard lighting can occur before or after the acquisition under differentiated lighting.

[0013] According to the invention, differentiated lighting corresponds to: - to a lateralization by weighting of the lighting of the face; - to a moiré effect achieved through synchronization, and in particular through the coordination and adaptation of the camera and PWM system parameters; - to partial illumination by said at least two light sources in the infrared, or even in a spectrum of wavelengths other than the visible; in this configuration, a second digital camera is used, equipped with a sensor whose sensitivity is adapted to the wavelength or range of wavelengths considered; - or to a variation in the light intensity generated by said at least two light sources.

[0014] According to a first embodiment of the invention, the acquisition of the individual's face under differentiated lighting is implemented by weighting said lighting, and in particular by lateralizing said lighting. This lateralization is made possible by the two aforementioned light sources, the intensity of which can be modulated during said acquisition using the PWM system. In other words, the two parts of the individual's face located substantially on either side of a median plane passing through the center between the two eyes and the bridge of the nose are illuminated with an intensity different. The processing of the acquired image makes it possible to distinguish anything that is not three-dimensional, such as a portrait printed on a flat or curved sheet of paper, or a photo appearing on an electronic display, such as a tablet or smartphone. Indeed, due to the implementation of lateral lighting, a difference in brightness will appear in the acquired image if it is actually a three-dimensional object, because of the shadows resulting from the relief of the face.

[0015] According to a second embodiment of the invention, the acquisition of the individual's face under differentiated lighting results from the formation of a moiré effect (better known by the English term "aliasing"), detectable by the acquisition camera, which again makes it possible to differentiate between tablet and smartphone screens. More precisely, the parameters of said camera, and in particular its scanning speed, are synchronized, and more specifically coordinated and adapted with the PWM system acting on the light sources, and therefore on the illumination frequency of the LEDs that constitute them.

[0016] This moiré effect is clearly visible in the presence of a real person in the image acquired by the camera, and across the entire surface exposed to the light. However, this moiré effect is less pronounced on the back of the booth if it receives sufficient light, assuming the back of the booth is actually equipped with a light source (see below). The background of the artifact (appearing on a printout or on an electronic display such as a tablet or smartphone) and the background of the booth have different patterns, thus allowing them to be identified.

[0017] According to a third embodiment of the invention, the acquisition of the individual's face under differentiated lighting employs partial illumination, for example, infrared. To this end, at least one of the LEDs of at least one of the two light sources emits in the infrared range. This partial infrared illumination makes it possible to detect the presence of a real face, and more specifically, the skin of said face. Indeed, the reflectivity of a surface, and for example, of skin, varies with wavelength, thus making it possible to distinguish the skin.

[0018] According to a fourth embodiment of the invention, the acquisition of the individual's face under differentiated lighting relies on a variation in the light intensity generated by said at least two light sources. Typically, if the light intensity of said sources is doubled, then this doubling appears on the subject's face. Conversely, this doubling does not appear on an electronic display held by said subject due to the light intensity generated by said display. The processing unit makes it possible to detect the presence or absence of this proportionality, and consequently to identify an artifact.

[0019] According to yet another aspect of the invention, the installation of the type in question may include a background, facing the wall of the booth integrating the acquisition camera, said background also being equipped with a lighting source controlled by the central unit. In this configuration, the acquisition under differentiated lighting of the individual's face implements the formation of moiré (aliasing) on ​​said background, by managing the background illumination frequency and the camera scanning frequency. Indeed, in such a configuration, the background lighting source is positioned behind the subject, so that the flickering effect appears only on the background. Thus, if the subject is holding up an electronic display, the captured background does not exhibit aliasing. In this way, it is possible to distinguish the individual actually present within the installation from screens such as tablets or smartphones.

[0020] Whatever the differentiated lighting method implemented, possibly combined together, the image thus acquired by the camera is processed by the central management and processing unit, which, by artificial intelligence, will be able to discriminate an individual actually present in front of said camera from a 2D image or rendering, or even from the application of a mask on the face of the individual concerned.

[0021] In other words, the invention generally consists of modulating the lighting of the face of the individual whose portrait one wishes to acquire, in order to combat the risks of impersonation, in particular inherent in the wearing of a mask, the exposure of a smartphone or tablet screen or a printed sheet, or even of "morphing".

[0022] The invention also relates to a photographic portrait acquisition system for an individual implementing the method described above. This system consists of a booth delimiting an interior space, within which the individual may position themselves, and comprising: - a first wall delimiting part of the interior space of the cabin, at least part of which is transparent; - at least one digital camera arranged behind a transparent or semi-transparent part of said wall, and comprising a field of vision directed towards the interior of the cabin and whose object focal plane is located in the interior space of the cabin; - at least two light sources illuminating the interior of the cabin, arranged so that they illuminate the individual's face laterally, and located outside the field of vision of the acquisition camera.

[0023] According to the invention, these light sources consist of PWM-controlled light-emitting diodes, and said at least one camera is of the digital type, the management of the illumination of the light-emitting diodes and the acquisition of the camera being managed by a central unit also ensuring the processing of the images acquired by said camera.

[0024] According to a variant of the invention, the aforementioned installation further comprises a background, opposite said wall integrating the acquisition camera, said background being provided with a lighting source, also managed by the PWM system.

[0025] According to yet another embodiment of the invention, the entire process of acquiring the individual's image is supervised, typically by a supervisory authority. In other words, when secure enrollment is required, as may be necessary in a number of applications, an operator external to the installation must be able to monitor the various stages of said acquisition. To this end, the installation may include an additional camera designed to transmit the contents of the booth remotely and in real time. Alternatively, at least one of the walls delimiting the internal space of the booth may be equipped with variable opacity properties. Brief description of the figures

[0026] The manner in which the invention can be implemented and the resulting advantages will be more apparent from the following embodiment examples, given by way of illustration and not limitation, in support of the attached figures.

[0027] [Fig.1] Fig.1 is a schematic cross-sectional representation of an example of an embodiment of a photographic portrait acquisition installation according to the invention.

[0028] [Fig.2] Fig.2 is a schematic representation of a front view of one of the walls of the installation of the [Fig.l].

[0029] [Fig.3] [Fig.3] is a schematic view analogous to [Fig.2], of a variant of lighting design.

[0030] [Fig.4] Fig.4 is a schematic top view illustrating a first mode differentiated lighting, in this case by lateralization of the lighting.

[0031] [Fig. 5] Figure 5 schematically illustrates the image acquired of a face following the Lateralization of lighting.

[0032] [Fig.6] Fig.6 illustrates the image of Fig.5 under normal lighting.

[0033] [Fig.7] Fig.7 is a schematic view of the image acquired of a face under aliasing according to a second differentiated lighting mode.

[0034] [Fig.8] Fig.8 is a schematic view of the acquisition of a smartphone screen acquired under aliasing, to be compared with the view in [Fig.7].

[0035] [Fig.9] [Fig.9] is a schematic top view, analogous to [Fig.4], illustrating another method of differentiated lighting, in this case by partial infrared lighting.

[0036] [Fig. 10] The [Fig. 10] schematically illustrates the image acquired of a face under aliasing from the background of the booth according to another differentiated lighting mode.

[0037] [Fig.11] The [Fig.11] is an analogous view of the [Fig.10], illustrating a variation of the parameters causing aliasing.

[0038] [Fig. 12] The [Fig. 12] is a schematic view of the acquisition of a smartphone screen under the same conditions as those of the [Fig. 10].

[0039] [Fig. 13] The [Fig. 13] is a schematic view of the acquisition of a printed sheet under the same conditions as those of the [Fig. 10].

[0040] [Fig. 14] The [Fig. 14] is a schematic view illustrating the image acquired of a face under aliasing of the background of the booth and side lighting, i.e. mixing two of the differentiated lighting modes illustrated previously.

[0041] [Fig. 15] The [Fig. 15] is a schematic view illustrating the image acquired of a face implementing the preceding mode, with camera saturation by background lighting.

[0042] [Fig. 16] The [Fig. 16] is a schematic view analogous to the [Fig. 15], illustrating the image acquired from a sheet printed under the same conditions.

[0043] [Fig. 17] [Fig. 17] is a view analogous to [Fig. 16], illustrating the image acquired from a smartphone, again under the same conditions as those of [Fig. 15]. Detailed description of the invention

[0044] Figure 1 illustrates an example of the embodiment of a photographic portrait acquisition installation 10 according to the invention. It comprises a booth 12 delimited by four walls 14, 16, 18 and 20, in addition to the ceiling 24.

[0045] A first wall 14 has at least one transparent part, in order to allow a digital camera 22 positioned behind said wall to have its field of vision 23 and its object focal plane 26 inside the cabin.

[0046] A seat 25 may be present in the cabin to allow a user 30 to sit facing and near the object focal plane 26 of the digital camera 22. More specifically, the seat is positioned to include the object focal plane of said camera. An adjustment device allows the user to adjust the position of the seat or the digital camera to place their face 31 in a reference position 32. The reference position is centered in the plane defined by the intersection of the field of vision 23 and the object focal plane 26 of the digital camera 22.

[0047] The installation 10 includes means enabling the user to position his face in the reference position, whether he is standing, or where applicable, sitting on the seat 25. These means may be visual indications located on the wall 14 in front of the digital camera 22, such as for example a marker to position his eyes at the correct height in the field of vision of said camera.

[0048] The installation includes a light emission source 35 positioned between the object focal plane 26 of the digital camera 22 and the first wall 14 of the cabin. According to the present example, it is fixed to the ceiling 24 of the cabin by appropriate means, or embedded within the latter.

[0049] As illustrated in [Fig. 2], the light-emitting source 35 comprises a first and a second luminaire, that is to say, it actually defines two light sources. More precisely, a first luminaire 36 is arranged to illuminate the second wall 16 delimiting the cabin and adjacent to said first wall 14, and a second luminaire 37 is positioned to illuminate the third wall 18 of said cabin, facing the second wall 16. The second and third walls are therefore positioned opposite each other and intersecting the first wall 14, either orthogonally or in such a way as to define an obtuse angle with it. In this latter configuration, the cross-section of the cabin is no longer rectangular, but trapezoidal.

[0050] According to the embodiment illustrated in Figures 1 to 3, the first and second luminaires each consist of an array of LEDs 38 held and oriented opposite the side walls 16, 18 of the cabin, for example, by means of a first wedge 39 and a second wedge 40, respectively. However, any other arrangement could be considered. In other words, the longitudinal arrangements of the LEDs are substantially parallel to the second wall 16 and / or the third wall 18. The LEDs emit warm white light, with a color rendering index greater than 85, a temperature of approximately 4000 Kelvin, an intensity of at least 550 lumens (per luminaire), and a beam angle between 30° and 40°.These various LEDs are powered by appropriate current sources, and their power supplies are periodically switched by a PWM system, for which a pulse frequency of several hertz to several kilohertz and a variable duty cycle of 0 to 100% are defined according to the desired effect. This PWM system is itself managed via a USB interface by a central control and processing unit (not shown) via an interface electronic board.

[0051] More particularly, according to a preferred embodiment of the invention, the angles of the wedges 39 and 40 are chosen so that the luminaires 36, 37 illuminate areas of the second 16 and third 18 walls located at the same height in the cabin as the surface delimited by the intersection between the field of vision 23 and the object focal plane 26 of the digital camera 22. The inclination generated by the wedges may not be identical for each of the luminaires 36, 37, the optimization of these angles typically depending on the distance between the user (reference position 32) and the walls 16 and 18, as well as the distance between the user and the height of the light source 35, and the center-to-center distance between the two luminaires 36 and 37.

[0052] Thus, the light emitted by the light emission source 35 is reflected onto the areas 41, 42 of the walls 16, 18, thereby forming a first diffusing emitting surface light is emitted from the second wall 16, and a second light-emitting diffusing surface is emitted from the second wall 18, positioned in front of and to the sides of the user's face 31. The light thus generated is also reflected from the first wall 14, these different walls being coated with a suitable light-reflecting material.

[0053] Advantageously, the light-emitting surfaces 41, 42 and 14 illuminate all parts of the user's face 30 from numerous incident directions. As a result, the face is illuminated homogeneously.

[0054] Due to the presence of these two luminaires, their location, and their mode of operation, and in particular the reflection of the light they emit on the diffusing surfaces 41, 42, and 14, in addition to their technical characteristics (luminous intensity), the installation of the invention can do without any face-mounted light source, provided that the inner face of the wall 14 is coated with a reflective layer. Thus, a portion of the light emitted by the light sources 36 and 37 is reflected off this wall 14 and serves as face (or frontal) lighting for the user's face.

[0055] However, such a face light source can be implemented. It has been represented by reference 45 in Figures 1 to 3. This face light source can consist of an LED 46, emitting a warm white light characterized by a color rendering index greater than 85, a temperature of about 4000 Kelvins, a native intensity greater than 550 Lumens and a diffusion angle of 120°.

[0056] In fact, due to the presence of the two lateral light sources 36, 37, it is possible either to reduce the intensity of the light emitted by the front light source 45 when it is present, or to compensate for its absence by adjusting the intensity of the illumination of said lateral light sources 36, 37. The glare to the user by the front light source 45 is then less significant, or even absent, thus offering greater comfort for the user of the installation.

[0057] According to the invention, a fourth wall 20 closes off the cabin and forms the back, opposite the first wall 14. This fourth wall 20 may be at least partially transparent, to allow another light-emitting device 50 arranged behind said wall to illuminate the interior of the cabin. This light-emitting device 50 may comprise a matrix array of LEDs 51 of different colors, for example blue, red, and green, enabling the back of the cabin to be illuminated in several colors. The LEDs may be similar to those described above. This LED array 51 is also controlled by the PWM system mentioned previously.

[0058] One face of the fourth wall 20 may also be covered by an anti-reflective (or matte) and dark, for example black, and semi-transparent layer. This face of the The fourth wall can then absorb some of the light coming from inside the cabin, to allow optimal saturation and color homogeneity of the cabin background by the fourth emission device.

[0059] Furthermore, at least one of the two side walls 16, 18 of the cabin must necessarily allow the user to enter the volume it defines. To this end, either one of the two walls 16, 18 slides, or it is replaced by a curtain or any other blackout support. In the latter case, and in order to take into account the difference in nature between such a curtain and the side wall opposite it, and with the aim of having suitable lighting, the light sources 36, 37 are adjusted by differentiating their intensity, in particular by means of the PWM system.

[0060] Furthermore, in the context of implementing such a booth for secure enrollment, involving supervision of the facial acquisition process by a third party, for example by an operator specifically designated for this purpose, the side walls 16, 18, as well as the back wall 20, can be equipped with variable opacity properties. Typically, all or part of said walls can be made of a transparent material, for example glass, and coated with a liquid crystal film. Such a film has electro-optical properties such that it switches from an opacity state to a transparent state under the application of an electric field. In this case, the electrical pulse controlling the transition between the opacity state and the transparent state is synchronized with the at least two facial acquisitions by the central control and processing unit, in order to ensure the desired illumination.The actual acquisition phases are carried out under the occulting state.

[0061] It is understood that such a film could be replaced by any system ensuring the same function, and for example of the type of Venetian or Californian blinds, the orientation of which is controlled by an electric motor, also on the order of the control and processing unit.

[0062] According to a variant of the previous embodiment, and still with a view to having secure enrollment, a surveillance camera is positioned inside the cabin, for example on the ceiling, filming the interior of said cabin in real time, allowing an operator or a supervisory authority to control operations remotely.

[0063] The process of acquiring biometric photographic portraits of an individual, using the installation described above, will now be described.

[0064] According to an essential feature of the invention, once the individual concerned is in place inside the booth, at least two acquisitions of said individual's face are carried out, respectively: - an acquisition carried out under standard lighting, that is to say, under which the two light sources 36, 37 are activated simultaneously so as to illuminate said face in a uniform manner, in order to allow the biometric acquisition of said face by the digital camera 22, i.e. of the view subsequently used for example for identity papers or access control media; - and an acquisition carried out under differentiated lighting, that is to say under which the two light sources 36, 37 generate a lighting effect at the level of said face, also acquired by the digital camera 22.

[0065] These two acquisitions are not carried out simultaneously, but on the other hand are carried out randomly for a short period, typically less than 1 second, and preferably less than 300 milliseconds.

[0066] These at least two acquisitions are processed by the control and processing unit integrated within the cabin, which is able to determine, after processing the acquisition carried out under differentiated lighting, the presence or absence of an artifact.

[0067] The different modes of differentiated lighting will now be described.

[0068] According to a first embodiment of the invention, illustrated in relation to Figures 4 to 6, this differentiated lighting is constituted by the weighting of said lighting, and more specifically by a lateralization of said lighting. This lateralization is made possible by the two light sources 36, 37, the intensity of which can be modulated during said acquisition using the PWM system. In other words, the two parts of the individual's face located substantially on either side of a median plane passing through the center between the two eyes and the bridge of the nose are illuminated with different intensities, as illustrated in [Fig. 4] by arrows of unequal lengths emanating in this case from the light source 37.

[0069] This produces the image illustrated in [Fig. 5], which clearly shows a difference in illumination between the left and right sides of the face. In this respect, it is possible to vary, again using the PWM system, the difference in intensity between the two light sources 36, 37.

[0070] The processing of the image thus acquired makes it possible to discriminate against anything that is not three-dimensional, and for example, a portrait printed on a flat or curved sheet, or a photo appearing on an electronic display, such as a tablet or smartphone. Indeed, due to the implementation of lateral lighting, a difference in brightness will appear in the acquired image ([Fig. 5]) if a three-dimensional object is indeed present, because of the shadows resulting from the relief of the face.

[0071] Typically, using a suitable algorithm, the central processing unit is able to detect a variance in the face, which is an affine function of the modulation rate of the lighting from one light source relative to another. Thus, a lack of linearity is a sign of the presentation in front of the camera 22 of a flat surface, such as printed sheet or tablet or smartphone screen.

[0072] According to a second embodiment of the invention, illustrated in relation to figures 7 and 8, this differentiated lighting is constituted by the formation of a moiré effect, detectable by the digital camera 22.

[0073] In this case, the parameters of the digital camera 22, and in particular its scanning speed, are synchronized, and more precisely coordinated, with the PWM system acting on the light sources 36, 37, and therefore on the illumination frequency of the LEDs 38 that constitute them. To this end, the camera 22 uses a CMOS sensor of the type better known by the English term "rolling shutter," and as illustrated, for example, at the internet link https: / / en.wikipedia.org / wiki / Rolling_shutter.

[0074] This moiré (or aliasing) effect results in bands on the acquired image, appearing on the individual's face ([Fig. 7]) and on the background. The width of these bands depends, on the one hand, on the pulse period of the LEDs 38, knowing that this period is linked to the frequency of the PWM system and the PWM duty cycle (i.e., the variation in intensity of the LEDs 38), and on the other hand, on the reading time of the lines of the digital camera 22. The properties of these bands can vary in intensity and width over certain areas of the image. These areas include, in particular, light-emitting surfaces that may be located opposite the camera 22. Thus, typically, an electronic display such as a smartphone or tablet, or even an illuminated surface, such as surface 20, when the latter is equipped with a light source 50, affects the image captured by said camera. This can be observed in [Fig.[8], that the area limited to the smartphone screen, i.e. the light-emitting area, does not exhibit aliasing. In fact, the central processing unit is able to differentiate the presence of a real individual from an artifact arising from the presentation of such a screen, and more generally of a light source, in front of the camera 22.

[0075] According to a third embodiment of the invention, illustrated in relation to [Fig. 9], this differentiated lighting consists of partially illuminating the individual's face under a spectrum of wavelengths other than visible, and for example under infrared. To this end, at least one of the LEDs 38 of at least one, or even both, of the light sources 36, 37 emits in the infrared (the two upper arrows in [Fig. 8]).

[0076] This partial infrared illumination makes it possible to detect the presence of a real face, and more specifically, the skin of that face. Indeed, the reflectivity of a surface, and for example of skin, varies with wavelength, thus allowing the skin to be distinguished from a more neutral surface, such as a sheet of paper or a smartphone or tablet screen, by artificial intelligence at the processing unit level. Thus, typically, the reflectance of photographic paper, which could represent a face for the purpose of identity theft, is greater than that of a real face.

[0077] According to a fourth embodiment of the invention, not shown, this differentiated lighting consists of a variation in the light intensity generated by the two light sources 36, 37. Typically, if the light intensity of said sources is doubled, then this doubling appears on the subject's face and is detected by the central processing unit. Conversely, this doubling of intensity does not appear on an electronic display such as a smartphone or tablet held by said subject due to the light intensity generated by said display. The processing unit makes it possible to detect the presence or absence of this proportionality, and consequently to identify an artifact.

[0078] According to a fifth embodiment of the invention illustrated in relation to figures 10 to 13, this differentiated lighting is constituted by the formation of a moiré pattern (aliasing) at the level of the bottom 20 of the cabin, when the latter is equipped with a lighting source 50, of the type previously mentioned.

[0079] Thus, by adjusting the scanning frequency of the digital camera 22, again of the "rolling shutter" type, and the illumination frequency of the LEDs 51 of the background light source 50 20, the acquired image exhibits a flickering effect (moiré) on the background (Figures 10 and 11). Therefore, if the subject holds up a printed sheet of paper ([Fig. 13]) or an electronic display ([Fig. 12]), the background captured locally around the subject does not exhibit aliasing. Thus, in Figures 12 and 13, it can be observed that the area behind the face displayed on the screen or on the sheet of paper is free of any aliasing. In this way, the central processing unit is able to distinguish the individual actually present within the display setup of tablets or smartphones.

[0080] According to the invention, these different modes of differentiated lighting can be implemented alone or in combination. Such a combination optimizes the central processing unit's ability to discriminate artifacts, and therefore, as a corollary, to combat any attempt at impersonation even more effectively.

[0081] For example, in relation to figures 14 to 17, the combined implementation of the second and fifth embodiments described previously has been described, that is to say, the two aliasing modes described.

[0082] Thus, on [Fig. 14], we observe double aliasing, that is to say with different beat widths between the individual and the rest of the acquisition.

[0083] In [Fig. 15], the intensity of the light source 50 on the background wall 20 is adjusted to saturate the digital camera 22. This results, for example, in [Fig. 15], in aliasing only on the individual, the rest of the acquired image being white. In this same configuration, i.e., with the digital camera 22 saturated by the intensity of the light source 50, one can observe in [Fig. 16], illustrating The acquisition of a printed sheet of paper reveals that the entire sheet exhibits aliasing, including areas outside the printed image. Similarly, and in the same configuration, we can observe in [Fig. 17] that the smartphone only displays aliasing around the edges of the screen itself, the screen being capable of causing saturation of the digital camera 22.

[0084] With this combination of differentiated lighting modes, the discrimination performed by the central processing unit is thus optimized, and therefore, consequently, the risks of identity theft.

[0085] The full interest of the present invention in the fight against identity theft, an extremely sensitive subject for security services, whether state or private, can be understood.

Claims

Demands

1. A method for acquiring photographic portraits of an individual within an essentially enclosed space, consisting of: - to acquire a biometric image of the individual's face using at least one digital camera (22), - concomitantly with this acquisition, to illuminate said face of the individual by at least two distinct light sources (36, 37), not centered on said face, and positioned laterally in relation to the field of vision (23) of said camera, said light sources (36, 37) being made up of light-emitting diodes (LEDs) (38) managed by a PWM system; the PWM system and the operation of the digital camera (22) being managed by a central control and processing unit; method in which at least two acquisitions of said face are carried out, respectively: - according to standard lighting, that is to say according to which said at least two light sources (36, 37) are activated simultaneously so as to illuminate said face uniformly, in order to allow the biometric acquisition of said face by said at least one camera (22); - according to differentiated lighting, that is to say according to which said at least two light sources (36, 37) generate a lighting effect at the level of said face, also acquired by said at least one camera; these at least two acquisitions being carried out not simultaneously and randomly over a short period of time; the central command and processing unit being capable of processing said acquisitions, and of determining after processing the acquisition according to differentiated lighting the presence or absence of an artifact.

2. A method for acquiring photographic portraits of an individual within an essentially enclosed volume according to claim 1, wherein the differentiated lighting results, alone or in combination: - of a lateralization by weighting of the lighting of the face; - a moiré effect achieved by synchronization, and in particular by the coordination and adaptation of the parameters of the digital camera (22) and the PWM system; - partial illumination by at least one of the aforementioned at least two light sources (36, 37) in a spectrum of wavelengths other that the visible, and in particular in the infrared; - or a variation in the luminous intensity generated by said at least two light sources (36, 37).

3. A method for acquiring photographic portraits of an individual within an essentially enclosed volume according to claim 2, wherein the acquisition under differentiated lighting of the individual's face is implemented by weighting said lighting, and in particular by lateralizing said lighting, resulting from the management of the two light sources (36, 37) using the PWM system, and generating at the level of said face, an illumination of different intensity on the two parts of said face of the individual located substantially on either side of a median plane passing through the center located between the two eyes and the arch of the nose.

4. Method of acquiring photographic portraits of an individual within an essentially enclosed volume according to claim 2, wherein the acquisition under differentiated lighting of the face of the individual results from the formation of a moiré effect, detectable by the digital acquisition camera (22), the formation of this moiré effect resulting from the synchronization of the parameters, and in particular by the coordination and adaptation of the operating parameters of said camera with the PWM system acting on the light sources (36, 37), and therefore on the lighting frequency of the LEDs (38) which constitute them.

5. Method for acquiring photographic portraits of an individual within an essentially enclosed volume according to claim 2, wherein the acquisition under differentiated lighting of the individual's face uses partial infrared lighting, at least one of the LEDs (38) constituting each of the two light sources (36, 37) emitting in the infrared.

6. Method of acquiring photographic portraits of an individual within an essentially enclosed volume according to claim 2, wherein the acquisition under differentiated lighting of the individual's face is based on a variation of the light intensity generated by said at least two light sources (36, 37).

7. A method for acquiring photographic portraits of an individual within a substantially enclosed volume according to claim 1, wherein the installation further comprises an additional lighting source (50) also consisting of LEDs (51) facing the digital acquisition camera (22) and positioned beyond the object focal plane

8. (26) of said camera, this additional light source (50) also being managed by the central management and processing unit, and in which the acquisition under differentiated lighting of the face of the individual results from the formation of moiré at the level of the background of the image acquired by the digital camera (22), the formation of this moiré effect resulting from the synchronization, and in particular from the coordination and adaptation of the parameters of said camera with the PWM system acting on the LEDs (51) of said additional light source (50), and therefore on the lighting frequency of the LEDs (51) which constitute it. Installation for acquiring a photographic portrait of an individual for implementing the method according to any one of claims 1 to 7, consisting of a booth (10) delimiting an interior space (12), within which said individual (30) may position themselves, and comprising: - a first wall (14) delimiting a part of the interior space of the cabin, at least part of which is transparent; - a camera (22) arranged behind said transparent part of said wall (14), and comprising a field of vision (23) directed towards the interior of the cabin and whose object focal plane (26) is located in the cabin; - at least two light sources (36, 37) illuminating the interior of the cabin, arranged so that they laterally illuminate the face (31) of the individual, and being located outside the field of vision (23) of the acquisition camera (22), characterized: - in that the lighting sources (36, 37) consist of light-emitting diodes (LEDs) (38) managed by a PWM system, - in that the acquisition camera (22) is of the digital type, - in that the management of the illumination of the light-emitting diodes (38) and of the acquisition of the camera (22) is managed by a central control unit which also ensures the processing of the images acquired by said camera, said light sources (36, 37) being activated simultaneously in order to allow the biometric acquisition of said portrait by the camera (22) on the one hand, and said light sources (36, 37) generating differentiated lighting in order to allow another acquisition of the individual's portrait by the camera (22) on the other hand, said acquisitions not being carried out simultaneously, but randomly for a duration of less than 1 second.

9. Installation for acquiring the photographic portrait of an individual according to claim 8, wherein the light sources of illumination (36, 37) consist of two LED strips (38), positioned at the ceiling (24) of the installation between said first wall (14) and the object focal plane (26) of the digital camera (22), and wherein said light sources (36, 37) illuminate the side walls (16, 18) of the installation, intersecting with said first wall (14).

10. Installation for acquiring the photographic portrait of an individual according to claim 9, wherein at least one of said side walls (16, 18) is constituted by a retractable shutter curtain or an occulting support.

11. Installation for acquiring the photographic portrait of an individual according to any one of claims 8 to 10, further comprising a background wall (20), opposite said first wall (14) integrating the acquisition camera (22), and positioned beyond the object focal plane (26) of said camera, said background wall (20) being provided with an additional lighting light source (50), consisting of LEDs (51) and managed by the PWM system also ensuring the management of the light sources (36, 37).

12. Installation for acquiring the photographic portrait of an individual according to any one of claims 8 to 11, wherein at least one of the walls (16, 18, 20) is transparent, and is endowed with variable occultation property, the change of state between transparent state and occulting state being managed by the central control unit.

13. Installation for acquiring the photographic portrait of an individual according to claim 12, in which at least one wall is coated with a liquid crystal film, which switches from an occulting state to a transparent state under the application of an electrical pulse synchronized by the central control unit.

14. Installation for acquiring the photographic portrait of an individual according to any one of claims 8 to 11, wherein the booth incorporates an additional camera intended to transmit remotely and in real time the contents of said booth.