Projection device with optimized emission point distribution based on isoline configurations

Abstract

A method is proposed for manufacturing a device (10) for projecting an image onto an eye, comprising an emission surface S. The surface S comprises an array of waveguides, an array of diffraction gratings, and an array of electrodes. Each grating is positioned at the intersection of one of the waveguides and one of the electrodes to form a point of emission of a light wave, the array of waveguides being arranged in a waveguide configuration, and the array of electrodes being arranged in an electrode configuration. The method comprises a design phase of said device and a material fabrication phase of said device thus designed.The design phase includes the step of determining (290) said waveguide configuration or said electrode configuration, said determined configuration being a configuration of non-straight isolines defined in the plane defined by the emission surface S and generating an aperiodic distribution of emission points. Figure for the abstract: [Fig.13].

Description

Title of the invention: Projection device with optimized emission point distribution based on isoline configuration Technical field

[0001] The present invention relates generally to the projection of an image onto an eye, in augmented reality type applications, and in particular to a projection device with optimized distribution of emission points on a discretized emission surface and from isoline configuration, as well as a method for manufacturing such a device.

[0002] A portable optical system for displaying data in augmented reality superimposes, on the vision of the real world of a user of the system, an image containing information intended for the user, such as for example information on his environment, his position, his speed of movement, etc.

[0003] Certain known portable optical systems for displaying data in augmented reality use an image projection device comprising a transparent integrated optical circuit composed of a network of nanometric light guides, a network of electrodes and a holographic film, as described for example in patent applications FR3122929A1 and FR3022642A1. Such an image projection device is implemented without a screen or optical system, which makes it possible to obtain a compact optical system and an extended field of vision for the user. The intersections of the circuit of nanometric light guides with the network of electrodes make it possible to define a set of emission points capable of emitting a light wave directed towards the pupil of the user's eye. The set of emission points is subdivided into different subsets, each subset comprising emission points distributed as randomly as possible.Light waves associated with the same subset of emission points propagate in the same direction to form a single light point at the level of the retina of the user's eye. A light point corresponds to a pixel of an image to be projected.

[0004] However, such a device has the disadvantage of using distributions of emission points defined according to certain spatial periodicities generating diffraction effects during the formation of the image on the retina. Furthermore, it also has a limited density of the emission points of each subset subdivided to form each light point. This results in a light halo associated with the retinal projection which degrades the contrast of the image to be displayed. As a result, the quality of projection of images onto a user's retina is insufficient.

[0005] There is thus a need for an improved image projection device allowing to increase the density of emission points of a subdivided subset while improving the pseudo-random and aperiodic distribution of the emission points. Summary of the invention

[0006] The present invention improves the situation by proposing a method for manufacturing a device for projecting an image onto an eye. The device is defined in an orthogonal frame (X,Y,Z) and comprises an emission surface S extending generally in the plane (X,Y) of the orthogonal frame (X,Y,Z). The emission surface S comprises a stack of elements, the elements comprising a set of Mx waveguides, a set of Mx x My diffraction gratings 1 pv and a set of My electrodes e<^, Mx and My being positive integers whose product Mx x My is strictly greater than 1. Each diffraction grating r'iP is positioned at the intersection of one of the waveguides 8P and one of the electrodes eP so as to form an emission point EP^ of a light wave. The set of M x waveguides gp is arranged in a waveguide configuration, and the set of My electrodes e'i is arranged in an electrode configuration.

[0007] The method comprises a phase of designing the device and a phase of material manufacturing of the device thus designed. The design phase comprises a step consisting of determining the configuration of waveguides or the configuration of electrodes, the determined configuration being a configuration of non-rectilinear isolines defined in the plane (X,Y) defined by the emission surface S and generating an aperiodic distribution of the Mx x My emission points EPpq.

[0008] In embodiments, the configuration determination step may consist of determining the waveguide configuration and the electrode configuration, the determined configurations being a configuration of non-rectilinear isolines defined in the plane (X,Y) defined by the emission surface S and generating an aperiodic distribution of the Mx x My emission points EPpq.

[0009] Advantageously, the design phase may comprise the step of determining an initial geometric curve extending on either side of the emission surface S along an extension axis parallel to the X axis and / or to the Y axis of the orthogonal reference frame (X,Y,Z). The initial geometric curve may be defined randomly while respecting at least one curve construction characteristic.

[0010] The initial geometric curve can be made up of a succession of segments.

[0011] According to some embodiments, the design phase may comprise the step of applying a distance function to the initial geometric curve to generate the isolinear waveguide configuration and / or the isolinear electrode configuration.

[0012] The distance function can be determined by an F ast-Marching type method

[0013] Advantageously, the initial geometric curve can comprise only ordinary points called 'regular points' for which the partial derivatives of the curve along X and Y are not simultaneously zero.

[0014] The initial geometric curve may include at any point of the curve, in the (X,Y) plane, a radius of curvature greater than or equal to a predefined minimum radius of curvature.

[0015] The minimum bending radius may be defined based on the physical properties of the 8P waveguides and / or electrodes and / or manufacturing constraints of the waveguides or electrodes.

[0016] In embodiments, the design phase may comprise the step of discretizing the emission surface S into Lx xLy of elementary emission zones z,y according to a continuous mesh in the (X,Y) plane, each elementary emission zone z'7 comprising a subset of mÿ x mÿ of emission points ■ The step of determining the waveguide configuration and / or the electrode configuration may be applied to each elementary emission zone z'7 so as to form a sub-configuration of non-rectilinear isolinear waveguides and / or a sub-configuration of non-rectilinear isolinear electrodes.

[0017] The sub-configuration of non-rectilinear isolinear waveguides and / or the sub-configuration of non-rectilinear isolinear electrodes can be generated by applying a Fast-Marching type method to an initial geometric curve extending in the plane of the emission surface S on either side of the elementary emission zone z'7 along an extension axis parallel to the X axis and / or to the Y axis.

[0018] Another object of the invention is a device for projecting an image onto an eye. The device is manufactured using the method for manufacturing. The device is defined in an orthogonal coordinate system (X,Y,Z) and comprises an emission surface S extending generally in the plane (X,Y) of the orthogonal coordinate system (X,Y,Z). The emission surface S comprises a stack of elements, the elements comprising a set of Mx waveguides 8^ a set of Mx x My diffraction gratings rP(i and a set of My electrodes, the numbers Mx and My being integers strictly greater than 1. Each diffraction grating is positioned at the intersection of one of the waveguides 8p and one of the electrodes eP so as to form an emission point EPpq of a light wave. The set of Mx waveguides Sp is arranged in a waveguide configuration, and the set of My electrodes is arranged in an electrode configuration.The waveguide configuration or electrode configuration is a configuration of non-rectilinear isolines defined in the (X,Y) plane defined by the emission surface S and generating an aperiodic distribution of the (Mx X Mv) points. EPpq emission.

[0019] The invention also provides a portable optical system for displaying transparent data comprising a device for projecting an image. The system is an augmented reality glasses system or headset.

[0020] The image projection device according to the embodiments of the invention makes it possible to improve the quality, in particular the contrast, of projection of images onto the retina of a user, from an aperiodic distribution of emission points optimized on a discretized emission surface, an increase in the densification of the emission points on an emission surface and isoline configurations.

[0021] Such a device also makes it possible to facilitate and accelerate the process for manufacturing the device. Description of figures

[0022] Other characteristics, details and advantages of the invention will emerge from reading the description given with reference to the appended drawings given by way of example.

[0023] [Fig. la] [Fig. la] is a diagram showing an image projection device, according to embodiments of the invention.

[0024] [Fig. 1b] [Fig. 1b] is a diagram showing an example of an image to be projected onto an eye of a user of a portable optical data display system, according to embodiments of the invention.

[0025] [Fig.2] [Fig.2] is a diagram showing a cross-section of the structure of the emission surface of an image projection device along the plane (X,Z), according to embodiments of the invention.

[0026] [Fig.3] [Fig.3] is a diagram representing a discretization of the surface emission of an image projection device in the (X,Y) plane, according to embodiments of the invention.

[0027] [Fig.4] [Fig.4] is a diagram representing an elementary emission zone of the discretized emission surface of an image projection device in the (X,Y) plane, according to embodiments of the invention.

[0028] [Fig.5] [Fig.5] is a diagram representing a projection in the (X,Y) plane of the angular projection domain associated with an elementary emission zone of the discretized emission surface of an image projection device, according to embodiments of the invention.

[0029] [Fig.6] [Fig.6] is a diagram representing a set of illumination zones on a pupil associated with the elementary emission zones of the discretized emission surface of an image projection device, according to embodiments of the invention.

[0030] [Fig.7a] and [Fig.7b] Figures 7a and 7b are diagrams representing surfaces emission images reproduced in the (X,Y) plane on an image projection device, according to embodiments of the invention.

[0031] [Fig.8a] [Fig.8a] is a graph representing a spatial distribution of the intensity of the percussion response formed by a distribution of emission points of an image projection device, according to embodiments of the invention.

[0032] [Fig.8b] [Fig.8b] is a graph representing an evolution of a power ratio of a spatial distribution of the intensity of the percussion response as a function of a density of the emission points of an image projection device, according to embodiments of the invention.

[0033] [Fig.9] [Fig.9] illustrates the results of simulations of a projected image for dif different comparative dimensions of an image projection device, according to embodiments of the invention.

[0034] [Fig. 10] [Fig. 10] is a flowchart showing steps of a method for manufacturing the image projection device, according to embodiments of the invention.

[0035] [Fig.11a] and [Fig. 11b] Figures 11a and 11b are flowcharts representing sub-steps of a method for manufacturing the image projection device, according to embodiments of the invention.

[0036] [Fig. 12] Figures 12a and 12b are diagrams representing an optical system for recording holograms, according to embodiments of the invention.

[0037] [Fig. 13] [Fig. 13] is a diagram showing the configurations of waveguides and isolinear electrodes of an emission surface of an image projection device, according to embodiments of the invention.

[0038] [Fig. 14] [Fig. 14] is a flowchart showing steps of a method for manufacturing the image projection device, according to embodiments of the invention.

[0039] [Fig. 15] [Fig. 15] is a flowchart showing graphics associated with steps of a method for manufacturing the image projection device, according to embodiments of the invention.

[0040] [Fig. 16a] and [Fig. 16b] Figures 16a and 16b represent simulation results of the spatial distribution of the intensity of the percussion response for “segment translation” configurations and emission surface isoline configurations of an image projection device, according to embodiments of the invention.

[0041] [Fig. 17a] and [Fig. 17b] Figures 17a and 17b represent results of applying criteria for quantifying the quality of projection of an image for configurations by “segment translations” and configurations of surface isolines emission of an image projection device, according to embodiments of the invention.

[0042] [Fig. 18] [Fig. 18] is a diagram representing the configurations of waveguides and isoline electrodes associated with a discretization of the emission surface of an image projection device in the (X,Y) plane, according to embodiments of the invention.

[0043] Identical references are used in the figures to designate identical or similar elements. For reasons of clarity, the elements shown are not to scale. Furthermore, in the remainder of the description, unless otherwise indicated, the terms "substantially" and "generally" mean "within plus or minus 10%". Detailed description

[0044] Figure 1a schematically represents an image projection device 10 defined in an orthogonal reference frame (X,Y,Z) and comprising a set of waveguides 8P comprising a number Mx of waveguides and a set of electrodes e'i comprising My electrodes, each waveguide 8P being non-parallel to each of the electrodes according to embodiments of the invention.

[0045] The numbers Mx and My are positive integers whose product (Mx X My) is strictly greater than 1. The parameter P designates an index associated with the different 8P waveguides, with pe [ L Mx ], and the parameter ? designates an index associated with the different electrodes with qg [ 1, Mv] .

[0046] The image projection device 10 can be used for example in a portable optical data display system, in the field of displaying images in augmented reality (AR) and more generally in the field of virtual reality (VR) and mixed reality (MR). Figure 1a corresponds to such an application. As shown in Figure 1a, the image projection device 10 can be arranged at a distance ZCT. from the pupil P of an eye of a user of the portable optical data display system. The Z axis then corresponds to the optical axis of the gaze of the user of the system and is associated with the position of the eye in the plane (X,Y) of the orthogonal reference frame (X,Y,Z).For example, the Zer distance (also called 'eye relief' or 'eye relief' according to the corresponding Anglo-Saxon expression) can be equal to 20 mm for a portable optical data display system comprising a pair of glasses of the prescription glasses type, or equal to 30 mm for a system comprising an RA or VR / RM immersion headset.

[0047] The image projection device 10 comprises an emission surface S, generally extending in the plane (X,Y) and formed of (Mx X My) emission points, denoted EPpq. Each emission point EPpq corresponds to the intersection of a waveguide 8P and an electrode e <i. Chaque point d'émission EPpq est adapté pour émettre une onde light propagating along a propagation axis (or wave vector) from the emission surface S towards the pupil P which then focuses the light on the retina R of the eye. The emission surface S then corresponds to a retinal projection screen.

[0048] The image projection device 10 is adapted to project an image of size QVv x Nen number of pixels and substantially defined in the projection plane of the retina R of the eye. Figure 1b represents an example of an image to be projected onto an eye of a user. Each projected pixel then corresponds to a point Ruv illuminated on the retina R.

[0049] The numbers Nx and Ny are positive integers whose product (Nx X Ny) is strictly greater than 1. The parameter 11 designates an index associated with the pixels of the image on the X axis, with u E [ 1, Nx ], and the parameter v designates an index associated with the pixels of the image along the Y axis, with yg [ 1, A\] • For example and without limitation, a minimum number of the product of the integers (Nx XA\) can be equal to 100 pixels of an image to be projected.

[0050] The set of (Mx X My) emission points EPpq is then divided into (x Ny) subsets of emission points. These subsets of emission points are also called 'distributions of emission points' and designated by the notation EPDUV. It should be noted that a minimum number of the product of the integers ( Mx X Mv ) can be defined from the maximization of the number of distributions of emission points ( Nx X AA) on the emission surface S (i.e. pixels to be projected) and the number nemiEPD of emission points EPpq per distribution of emission points EPDllv. For example and without limitation, the number nemiEPD can be between 50 and 200, and the minimum number of the product of the integers ( Mx X Mv ) can therefore be equal to 5000 emission points EPpq.

[0051] Each distribution of emission points EPDUV can be formed in such a way that all the emission points EPpq of the same distribution all emit a light wave tuned in phase along the same propagation axis £. It should be noted that a first light wave and a second light wave each having the same propagation axis are tuned in phase if in a plane perpendicular to £ the value of the phase of the second light wave is substantially equal to the value of the phase of the first light wave modulo The light waves emitted by the emission points EPpq of the same distribution can therefore propagate in parallel between the image projection device 10 and the pupil P of the eye over the distance and converge substantially to the same point Rlw on the retina thanks to the crystalline lens C of the eye as shown in FIG. 1a. In this case, the eye perceives the projected image by accommodating to infinity. According to a non-relevant variant shown in the figures, the phase-tuned light waves emitted by the EPpq emission points of the same EPDUV distribution can propagate in substantially divergent directions around a direction that targets the Rltv point (i.e. pixel with indices (u, y)) in the visual field. In this case, the eye perceives the projected image by accommodating at a distance, called the accommodation distance and noted da, which can be a few meters (i.e. da is not equal to infinity). For example and without limitation, the accommodation distance da can be 2 m. The degree of divergence of a direction £ of one and / or more EPpq emission points with respect to the direction can then be determined from the accommodation distance da. The light intensity at the Rliv point results from the contribution of each light wave of the EPpq emission points. The distribution of EPDUV emission points is phase adjusted and forms a resulting light wave.The resulting light wave from EPDUV can be associated with a plane wave or a substantially spherical wave if the accommodation distance da of the image projected onto the retina is not at infinity). Advantageously, the distribution of the emission points EPpq for each distribution of EPDUV emission points can be determined randomly or pseudo-randomly.

[0052] [Fig. 2] schematically represents a cross-section of the structure of the emission surface S of the image projection device 10 along the plane (X,Z). This structure comprises a stack of several layers superimposed in the direction of the Z axis.

[0053] The device 10 comprises a first layer corresponding to the set of Mx 8P waveguides arranged parallel to the emission surface S. Each 8P waveguide extends on either side of the emission surface S, from an initial position to a final position arranged along an extension axis X' substantially parallel to the longitudinal axis X. For the sake of simplification, in FIGS. 1 (a) and 2, the 8P waveguides are shown schematically in a rectilinear shape along the axis X. Nevertheless, those skilled in the art will easily understand that the invention is not limited to such a form of waveguide configuration.

[0054] The 8P waveguides are configured to receive coherent light, emitted by one or more laser sources (not shown in the figures), and to propagate the coherent light along the emission surface S. The Sp waveguides can for example be formed of silicon nitride, and have a width (gy), in the direction of the Y axis in [Fig. 1a], and a thickness (gz), in the direction of the Z axis, both between 100 nm and 600 nm.

[0055] The device also comprises a second layer corresponding to the set of My electrodes arranged parallel to the emission surface S. Each electrode e <? s'étend de part et d'autre de la surface d'émission S, à partir d’une position initiale vers une position finale disposées suivant un axe d’extension Y’ sensiblement parallèle à l'axe transversal Y. A titre de simplification, sur la figure la, les électrodes sont représentés schématiquement sous forme rectiligne dans la direction de l’axe Y. However, those skilled in the art will readily understand that the invention is not limited to such a form of electrode configuration.

[0056] The electrodes e <i sont configurées pour recevoir une tension de polarisation (ou modulation électrique) spécifique, gérée par ou plusieurs alimentations électriques (non représentées sur les figures), et propager la électrique le long surface d'émission s. les électrodes constituées d'un matériau conducteur, exemple des transparentes, ce conducteur peut être l’oxyde d'indium étain. ev peuvent avoir largeur (ex), dans direction l’axe x figure la, comprise entre 2 pm 10 pm, épaisseur (ez ), z, dépendant du conducteur. par exemple, l’épaisseur (ez) d’électrodes d’oxyde étain typiquement 20 nm 100 nm.

[0057] l’ensemble my est superposé à l’ensemble mx guides d’ondes sp, parallèlement chaque électrode ‘croise’ ainsi façon définir intersections croisements). intersection correspond position point eppq, tel qu’utilisé ici, terme ‘intersection’ fait référence superposition guide 8p (i.e. première couche) d'une deuxième non parallèles elles.

[0058] le dispositif comprend en outre troisième couche, correspondant un ensemble (mx my} réseaux diffraction, notés réseau diffraction rw couplé optiquement 8p assemblé ep. tiip formé partir d’une variation périodique d’au moins indice réfraction pouvant modulé appliquant traversant l’électrode

[0059] la couche comprenant constituée structure continue plan (x,y) alternativement, également structures localisées s positionnées sensiblement au niveau formées d'inclusions, définissant motif périodique, l'oxyde silicium. inclusions alors tout ayant électriquement modulable, comme cristal liquide. lorsque longueur d'onde lumière émise source laser 532 nm, période 300 400 un étalé selon plage motifs périodiques (rx) s’étendre (ry). ces grandeurs (ry) définies zones eci. tels que supérieure égale (ex) ep induisant l’indice réfraction. (rz) 500

[0060] comprendre d’éléments d’orientations d’onde lumineuse configurés contrôler onde lumineuse. telle qu’utilisée l’expression « contrôle » (encore appelé ‘manipulation lumineuse’) divers phénomènes liés aux ondes électromagnétiques qui se produire lorsqu'un faisceau optique notamment interagit avec matière objet donné, tel représenté l’élément hpq [fig.2]. comprennent déviation angulaire, modification phase, transmission, réflexion, l'absorption, diffusion, l’onde électromagnétique.

[0061] avantageusement, quatrième h, positionnée lumineuse, h film holographique ( mv ) hologrammes, hp(f. hologramme élément d’orientation (réflectif transmissif) associé z. photopolymère, poly-méthacrylate méthyle, photorésine, d’épaisseur (hz) pm. enregistré codé) s'étendre (hx) (hy) 1 hologrammes jointifs recouvrant deux deux, hp^.^, permettant diminuer distance minimale électrodes.

[0062] l'empilement différentes couches éléments agencé support v. v transparent verre polycarbonate compris lunette visière système portable d’affichage données transparent.

[0063] sur 2, unique sp trois ponctuels fiq+l), rpi rp(q-ï), e(q+1), e,i ^g-l) h^q+y , hp(q^ représentés titre simplification d’exemple limitatif. l’homme métier comprendra aisément l’invention n’est pas limitée nombres configurations d’ondes,

[0064] volume d’émission (c’est-à-dire ep^) superposition, s, sp c; couplés ponctuel associés l’hologramme hpq. pour activer, c’est-à-dire faire émettre eppq, projection d’image configuré transmettre cohérente p moduler polariser) e(k une partie propageant &p extraite modifié flèche.}se propage d’orientations), sorte émet sortie hpq, spatialement ep{xp vecteur déphasage prédéterminés stockés préalablement enregistrés encodés) lumineuses émises différents points eppq même distribution epduv accordées phase unes rapport autres l'ensemble ces forme résultante dont front maîtrisé, plane accommodation l’infini, sphérique d’accommodation da. cette vers pupille p former lumineux ruv rétine r. fonde dépendent l'information enregistrée

[0065] chaque epduv permet donc formation perçu l'utilisateur pixel image. image formée illuminant successivement distributions epduv, comportant nombre élevé pixels. fréquence d'illumination dimensionnée manière puisse ressentir fixe sous l'effet persistance rétinienne, malgré techniquement séquentielle pixels l'image.

[0066] dans premier mode réalisation, discrétisée lx x lv) élémentaires notées zu formant maillage continu (x,y), [fig.3],

[0067] ly entiers positifs produit (lv ly) strictement supérieur 1. paramètre i discrétisation x, e [ 1, ], j y, y ] • sans limitation, minimum (lx à(lx2)ou(2xl) élémentaires.

[0068] zone élémentaire zlj associée g m!j eÿ, 1- illustré [fig.4].

[0069] zv sous-ensemble mÿ ep1^ défini parmi

[0070] z,j fini d'émission, epi%v- epd'^ d’un déterminés uniquement m1^ zlj.

[0071] chaque z'â répartition ep^pq d'émissionepd^ déterminée aléatoire pseudo-aléatoire. ep^q peut sélection pseudo-aléatoire my indices m' 9 respectivement gv pi constituant zv-

[0072] epd\{v projeté r, noté ^v. ainsi, epd'jv zi; égal (nÿ n’y) l’image projeter projetés r 7 «7 définis res-

[0073]

[0074]

[0075]

[0076]

[0077]

[0078]

[0079]

[0080]

[0081]

[0082] pectivement n'j 2va jÿ aa]- identiques nÿ total (aa nv) discrétisant équations (01) (02) suivantes : 4 (400 200) 16 12) zu. cas, z'7 (25 17) r) (25x epd2v correspondantes. cas où, zà «( identiques, u1’ v1’ correspondent intervalles d’indices j) nx ny expressions (03) (04) uv € -1) " + 1; v*j e^j-ljx^ l; jx-jr z'à epd^. -pi contenu domaine gammes angulaires zfv’, (05) (06) précédentes, 3x 3y projections y cône visualisation 6 (appelé encore champ vision yonjield ofview anglo-saxonne correspondante) œil, figures 3 5.

[0083] en particulier, z'7', angulaire (dç>, Ay) can be directed along a central vector oriented from the elementary emission zone z'7 to a central impact point fd in the (X,Y) plane at the pupil P. The angular direction of the central vector in the (X,Y,Z) plane is characterized by angles qp and ^7 determined with respect to the optical axis Z in the (X,Z) and (Y,Z) planes respectively. As an illustration, the projection K'j in the (X,Z) plane of the central vector ^7 of the angular domain of the elementary emission zone Zz7 is shown in Figure 5. The angle ^ / 7 corresponds to the angle between the optical axis Z and the direction of the projection k1] of the elementary emission zone zZ7 to the coordinate fi of the central impact point fi on the X axis, as shown in Figure 5.The angles (pj and associated with the central vector ^7 can be defined from the angular domain, the total number of elementary zones (Lx x discretizing the emission surface S and the position of the elementary emission zone z!7 on the emission surface S (represented for example by the indices ( / , j ) of the zone z'7)-The angles çp and y / j associated with the central vector can thus be defined according to the following equations (07) and (08): .

[0084] =

[0085] = ($0)

[0086] Advantageously, for each emission point gp1^ corresponding to the intersection of a waveguide Sp and an electrode e,i of a distribution of emission points EPI% of an elementary emission zone z'7 (as shown in FIG. 4), the angular direction of the wave vector -flJ in the plane (X,Y,Z) can be characterized by angles (p1^ and yPv determined with respect to the optical axis Z in the planes (X,Z) and (Y,Z) respectively. In particular, each hologram hpq associated with an emission point EP'^ of the distribution of emission points EPDljv can encode the angular directions and yPv defined according to the following equations (09) and (10):

[0087] ^7 = (W7-1) x 0-^(09) T uv xz 2

[0088]

[0089] In the preceding equations (09) and (10), the quantity 5 corresponds to the angular resolution of the image and depends on the level of sharpness desired on the display. For example, an angular resolution 5 considered as a visual limit value of angular resolution (i.e. a separable limit) corresponds to an angle of one minute of arc equal to the angle of 1 / 60 - 0.02 0 (and converted to radians in the calculations of the previous equations). It should be noted that the previous equations (09) and (10) are therefore obtained with an approximation of the small angles (i.e. tanô « ô). In this case, it can be considered that below this value the eye cannot perceive the pixelation of an image to be projected onto the retina R.

[0090] For each elementary emission zone z'7, the beams emitted by the emission points can impact the plane of the pupil P around the central point of impact pi over an impact extent WJ\ For example and without limitation, the impact extent wlj can be considered uniform for all the elementary emission zones z'7 of the emission surface S. The extent W' of the impacts of the emission points EP'^ in the plane of the pupil P can in particular take into account the extent of the angular domain of the elementary emission zone z'7 and the contribution of a diffraction of the emission points.

[0091] The coordinates / 7 and of the central impact point fl on the X axis and the Y axis respectively, in the plane of the pupil P, can be expressed according to the following equations (11) and (12):

[0092] ( (11)

[0093]

[0094] In the preceding equations (11) and (12), the magnitudes d']mx ct d'Jmv correspond to the size of the elementary emission zone z'7 along the X axis and the Y axis respectively, as shown in Figure 5. The sizes dl?mx ct dgmv May depend on the dimensions associated with the waveguides and the electrodes, their distribution in the emission surface plane S, as well as the numbers mÿ of waveguides Sp and m1^ of electrodes of the elementary zone z'7-

[0095] The discretization of the emission surface S into elementary emission zones z'7 can be uniform, the quantities d1^ and then being values ​​of constant sizes for all the elementary zones z'À and being able to be defined from the size of the emission surface S and the total number of elementary zones (Lv x Ly) discretizing the surface S, according to the following equations (13) and (14):

[0096] dj -2^(13) u emx ""

[0097] — (14)

[0098] For example and without limitation, the quantities d1^ and demy may be equivalent to the same quantity equal to 0.5 mm. In the preceding equations (13) and (14), the quantities Dcmx and Demy correspond to the size of the emission surface S along the X axis and the Y axis respectively, as shown in Figures 3 and 5. In particular, the size of the emission surface S can be determined as a function of the diameter Dpup of the pupil of the eye (typically not negligible and chosen to be equal to 4 mm typically). Indeed, the size of the emission surface S can be determined so as to be greater than the size of an area covered only by taking into account the viewing cone 0, that is to say by taking into account the diameter Dpiip of the pupil. The quantities Demx and Demy can then be defined, with a small angle approximation, according to the following equations (15) and (16):

[0099] Demx = (Zer x ô x Nx) + Dpup (15)

[0100] Demy = (Zer x <5 x A\.) + Dpup (16)

[0101] The discretization of the emission surface S into elementary emission zones z! / can also be uniform, the number of waveguides &P in an elementary zone and the number / nÿ of electrodes in an elementary zone then being constant values ​​for all the elementary zones zZJ and being able to be defined from the total number of ( Mx X My ) emission points of the emission surface S and the total number of elementary zones (Lx X Ly) discretizing the surface S, according to the following equations (17) and (18):

[0102] m0 = ^(17)

[0103] wv = ^(18) > Ly

[0104] Alternatively, the discretization of the emission surface S into elementary zones may be non-uniform, the number mÿ of waveguides &P and the number mÿ of electrodes being defined for example according to increasing and / or decreasing functions as a function of the X and Y axes respectively. The non-uniformity of the discretization of the emission surface S into elementary zones may similarly result in non-homogeneous quantities d1^ and demv as a function of the X and Y axes respectively.

[0105] In particular, it should be noted that the quantities associated with the emission surface S and the different elementary zones are expressed, in the preceding equations, using the small angle approximation as a simplification. This approximation is applicable for a small emission surface size S (i.e. small viewing cone 6) and / or an emission surface S centered with respect to the optical axis Z of the gaze of the user of the system. Alternatively, if the small angle approximation can no longer be applicable (i.e. tan5 / 5) the formulas of the preceding equations, in particular (09), (10), (15) and (16), are more complex. In this case, peripheral elementary zones of an emission surface S can be widened in the (X,Y) plane for high viewing angles with respect to the so-called central elementary zones with respect to the optical axis Z of this same emission surface S.

[0106] Figure 6 represents a diagram of all the impact extents (i.e. illumination zones on the pupil) at the positions of the central impact point fl for an emission surface S discretized for example into (16x 12) elementary emission zones of dimension <b égale à 0.5 mm et selon un champ de vision défini par des grandeurs 0x gy égales (12°,6° )• dans cet exemple, les points d'émission ep pq ont tailles similaires sur toute la surface d’émission s, définies diamètre 4 pm. ainsi, pour chaque zone élémentaire z'à dimension l’étendue d’impact w' (c’est-à-dire ‘tache’ diffraction au niveau l’œil) est 850

[0107] As shown in Figure 6, depending on the dimensioning of the device 10 and generally of the portable optical system for displaying transparent data, certain coordinates fl and f / of the central point of impact f J of an elementary emission zone z'7 can be positioned outside the pupil P on the (X,Y) plane. In this case, at least part of the pixels associated with the indices ifl and v,j is not projected onto the retina R of the eye because at least part of the light energy associated with the emission points pp1^ of the distributions of emission points EPD'2 does not penetrate the pupil represented by the circle P in [Fig.6].

[0108] However, this effect is partly compensated by the orientation of the eye in its orbit leading to a displacement of the position of the pupil P. Indeed, as illustrated by the dotted circle P' in Figure 6, the position of the pupil can be shifted in the (X,Y) plane when the eye targets a direction different from the direction of the optical axis Z. In Figure 6, and according to the dimensioning presented, the shift of the position of the pupil P' represents a target in a direction -6°: +3°, i.e. the direction concerned by the emission of a light signal coming from the elementary emission zone z01,01- In this case, the associated pixels can be projected onto the retina R of the eye because the light energy associated with the emission points of the distributions of emission points EPD'2 penetrates the pupil P'. The movement of the eye in its orbit therefore makes it possible to manage the dispersion of signals in the plane of the pupil.

[0109] Advantageously, to extend the viewing cone G of the image projection device 10 by one eye, i.e. the angular range of image projection, the emission surface S can be reproduced in the (X,Y) plane. In particular, the image projection device 10 can comprise a finite number of distinct emission surfaces Sxy distributed in the (X,Y) plane. For each emission surface S x\ each hologram hpq associated with an emission point pp'^ of a distribution of emission points EPD'2 can encode the determined angular directions and yA with respect to an optical axis rotated in the planes (X,Z) and (Y,Z) and centered with respect to the point Prxv corresponding to the position of the center of the pupil of the eye after rotation in its orbit and according to the angular direction of the area of ​​the emission surface Sxy concerned. The image projection device 10 shown diagrammatically in [Fig.7a] represents the emission surfaces S1, S2 and S3 positioned consecutively along the X axis and having different angular cone projections along the optical axis Z*=Z and the rotated optical axes Z2 and Z3 centered respectively on the points Pr1, Pr2 and Pr3 of the projection of the center of the pupil P on the X axis.

[0110] In particular, in the case of optical axes Zxy rotated relative to the optical axis Z of the user's gaze of the system, the small angle approximation may no longer be applicable (i.e. tanô * ô) for high viewing angles. In this case, the emission surfaces Sxy may be enlarged in the (X,Y) plane for high viewing angles relative to a so-called central emission surface relative to the optical axis Z. In [Fig.7a], the emission surfaces S2 and S3, associated with the rotated optical axes Z2 and Z3, are larger than the central emission surface S1 associated with the optical axis Z=Z*.

[0111] Furthermore, to extend the size of the "eye box", i.e. the area in which the user (and therefore his eye) can move in front of the image projection device 10 while continuing to see the entire image, the emission surface S can also be duplicated in the (X,Y) plane. In this case, the image projection device 10 can comprise a finite number of the same emission surface S distributed in the (X,Y) plane. For each distributed emission surface S, each hologram hpg associated with an emission point of a distribution of emission points EPD't'lv can encode the angular directions (p1-^ and ^0, determined with respect to the optical axis Z in the (X,Z) and (Y,Z) planes and centered with respect to the point Ptxy corresponding to the position of the center of the pupil of the eye after translation in the (X,Y) plane.The image projection device 10 shown diagrammatically in Figure 7b represents three times the same emission surface S positioned consecutively along the X axis and having the same three 6X projections of angular cone along the optical axis Z centered respectively on the points Pt1, Pt2 and Pt3 of the projection of the center of the pupil P translated on the X axis.

[0112] For example and without limitation, in the cases shown diagrammatically in Figures 7a and 7b, to duplicate or reproduce emission surfaces, the length of the waveguides Sp can be extended along the X axis, and sets of My electrodes €ci can be duplicated (in particular here three times) in the longitudinal direction of the guides to form the emission surfaces S and / or S1, S2 and S3. Similarly, the length of the electrodes can be extended along the Y axis, and sets of Mx of guides Sp waves can be duplicated in the transverse direction of the guides to form new S and / or Sxy emission surfaces.

[0113] The discretization of the emission surface S into elementary emission zones z'7 makes it possible to significantly increase the densification of the emission points of the same distribution of emission points EPD'^ ■ In particular, the density P of the emission points can be defined according to the following expression (19): 101141

[0115] The number nemtEPD qc emission points EPpq of the same distribution of emission points can be approximated from the size of the emission surface S, the surface sem of the emission point EPpq, and the number of pixels (Nx x Nv ) of the image to project, as defined by the following expression (20):

[0116] n EmlEPD

[0117] For example and without limitation, the number nem!EPD can be approximately equal to 50. For an image to be projected of (400 X 200) pixels and a discretization of the surface emission S in (16x 12) elementary emission zones z'7, the density P of points of EfP emission of the same distribution of emission points EPD„V May be equal according to the previous equation (19) to about 0.24%. In the previous equation (20), the area of ​​the emission point EPpq can be defined from the diameter of the emission point EPpq according to the following expression (21):

[0118] _æX2£(21)

[0119] It should be noted that there is at least one criterion for quantifying the quality of the image formed on the retina R by self-focusing of the emission points EPm of the same distribution of emission points EPDl(V. In particular, to quantify the effectiveness of self-focusing, a power ratio noted F can be defined from the spatial distribution of the intensity of the percussion response (designated by the acronym PSF, or Point Spread Function according to the corresponding English expression), formed by a distribution of emission points EPDUV at a point Ruv on the retina R of the eye, as shown in FIG. 8a. The power ratio Y can be expressed according to the following expression (22):

[0120] (22) 101og

[0121] In the previous equation (22), the quantity Py corresponds to the power of the central intensity peak formed by the distribution of emission points EPDltv, i.e. make the central peak represented in light gray in figure 8a. P^ + Pz denotes the total light power, the quantity Pz corresponding to the noise power intensity formed by the distribution of EPDUV emission points, i.e. the peripheral area of ​​the central peak shown in dark gray of the central peak shown in gray.

[0122] The power ratio Y is optimal for a zero power A of the intensity noise formed by the emission point (value of Y tending towards 0). Figure 8b represents the evolution of the power ratio Y according to the variation of the density P of the emission points EPpq of the same distribution of emission points EPDUV, and according to different characteristic parameters (for example characteristic dimensions) of the portable optical system for transparent data display. The criterion Y for quantifying the image quality is therefore well linked to the surface density of the distribution of the emission points EPm on the emission surface S, as illustrated in Figure 8b. However, the discretization of the emission surface S into elementary emission zones z'7 makes it possible to significantly increase the densification of the emission points of the same distribution of emission points EPD'^ ce 9ui Thus allows a significant gain (of at least 9 dB) on the value of the criterion Y compared to a configuration without discretization of the surface (i.e. for — 0.001%).

[0123] Figure 9 illustrates the results of image simulations for different comparative dimensions of the image projection device 10. In particular, the simulation results are carried out as a function of the discretization of the emission surface S via the number (Lx x Ly) of elementary zones zz / on the one hand and as a function of the size of the emission points EPpq defined by the diameter on the other hand, and for an image to be projected of (400 X 200) pixels representing the words “CEA leti Retinal Projection”. In particular, the variation of the number (Lx X Ly) of elementary zones zQ induces a variation of the size of the emission zone (0 or more generally the quantities ct demJ ct thus a variation of the density P of the emission points EPpq for each distribution of emission points EPDUV.

[0124] As illustrated by the results of image simulations in Figure 9, the implementation of the discretization of the emission surface S, then the reduction of the size of the emission zone (i.e. increase in density) allows an improvement in the contrast of the image to be projected. However, the increase in the size of the emission zone induces a reduction in the blur on the image to be projected. It should also be noted that the reduction in the size of the emission points (i.e. low diameter value wi) induces an increase in the proportion of the extent of the intensity noise formed by the emission points compared to the central emission peak and thus a reduction in the impact of this noise on the construction of the image on the retina R. Consequently, the results of image simulations, illustrated by Figure 9, make it possible to indicate a dimension Optimal operation possible with emission points between 2 pm and 4 pm in diameter and an elementary emission zone size z'7 between 200 pm and 800 pm.

[0125] The manufacture of a device for projecting an image 10 is implemented using a method comprising a design phase and a phase of physical manufacture of the device (10) thus designed.

[0126] [Fig. 10] is a flowchart representing steps of the design phase of the image projection device 10, according to embodiments of the invention.

[0127] The design phase of the image projection device can, for example, be implemented by computer.

[0128] In step 120, a discretization of the emission surface S into (Lx X Lv) of elementary emission zones z'7 can be applied so that each elementary emission zone z'7 comprises a subset of (x mÿ) of emission points rpO.

[0129] In step 140, a distribution of the subset of (mÿ x mÿ) of emission points EP'^q according to (nÿ X n'^ distributions of emission points EPD1^, is carried out.

[0130] In step 160, for each elementary emission zone z'7, the (nÿ x «y) distributions of emission points EPD'h, are assigned (i.e. associated or attributed) to (nÿ x « <!--) de pixels de l’image à projeter.<br-->

[0131] In step 180, for each distribution of emission points EPD1^ the direction of the wave vector of the resulting light wave emitted by the emission points of the distribution of emission points EPO^v is determined so that the direction -?'i is contained in an angular domain directed according to a central vector ^9 defined from the number (LK X Lv) of elementary emission zones z'7 discretizing the emission surface S and the position of the elementary emission zone z!J on the emission surface S.

[0132] [Fig.11a] is a flowchart representing examples of sub-steps of step 120 of discretization of the emission surface S in the design phase of the image projection device 10, according to embodiments of the invention.

[0133] In substep 102, the resolution of the image to be projected can be determined. For example, the resolution (Nx x Nv ) of the image to be projected can be chosen equal to (400 X 200) • In substep 104, the size of the field of view can be determined. For example, the magnitudes Bx and By of the field of view can be chosen equal to 12° and 6° respectively. In substep 106, the eye relief distance can be determined. For example, the eye relief distance ZCT can be chosen equal to 20 mm. In substep 108, the size of the emitting surface S can be calculated from the values ​​set in substeps 102 to 106. According to equations (15) and (16), the magnitudes Demx and DeMy of the emitting surface S can be equal to approximately 8 mm and 6 mm respectively.

[0134] In sub-step 110, the size of the elementary emission zones zv can be determined. For example, the magnitudes of ^ and d^mv of the elementary emission zones can be chosen equal to 500 pm. In sub-step 112, the number of elementary emission zones can be calculated from the quantities Demx and Demy and the values ​​set in sub-step 110. According to equations (13) and (14), the number (Lx x Ly) of elementary emission zones z'7 can be equal to approximately ( 19 X 12) • The number of pixels defined by elementary emission zones z'7 can then be deduced from the results of sub-step 110 and the values ​​set in sub-step 102. For example, the number ( nÿ x nÿ) of pixels, i.e. distribution of emission points, to be encoded by elementary emission zones can be equal to approximately 350.

[0135] Advantageously, in sub-step 114, widths ex and gy and configurations of the waveguides Sp and the electrodes ei can be determined. In sub-step 116, the numbers (my x my') of waveguides and electrodes per elementary emission zone can be calculated from the results of sub-step 114. And in sub-step 118, the maximum number nem / EPD emission points per distribution of emission points can be calculated from the numbers of waveguides Sp and electrodes Cti, and the resolution of the image to be projected (or number ("ÿ x "( / ) of pixels, i.e. distribution of emission points, per elementary emission zones z'7 for example).

[0136] Figure 11b is a flowchart representing examples of sub-steps of step 140 of distributing the emission points of an EPD^ distribution in the design phase of the image projection device 10, according to embodiments of the invention.

[0137] In sub-step 122, the number of waveguides Sp and the number of electrodes A may be determined by random and / or pseudo-random selections, in particular to obtain (or approach) the number n«miEPD emission points. For example, the maximum number nem!EPD emission points per distribution of emission points calculated in sub-step 118 may be equal to 50. The number of waveguides may be equal to 10 and the number of electrodes may be equal to 5.

[0138] In sub-step 124, the position of the waveguides 8P and the position of the electrodes €ci among the (x m1 / ) of waveguides Sp and electrodes e'{ of ​​the emission zone elementary z*7 can be determined by random and / or pseudo-random draws.

[0139] Sub-steps 122 and 124 can be applied for each distribution of emission points EPDÿv sequentially for example. The pseudo-random draws carried out for a distribution EPDljv can take into account the prior results of sub-steps 122 and 124 of one or more other distributions of emission points of the same elementary emission zone z'7- such that a pair of a waveguide Sp and an electrode eci representing a specific emission point EPpq can only be associated with a single distribution of emission points epi^e

[0140] The material manufacturing phase of the device (10) takes into account the results of the previously implemented design phase. The material manufacturing phase may comprise steps of manufacturing waveguides, diffraction gratings and electrodes, by conventional techniques, such as for example and without limitations chemical or physical deposition techniques. The material manufacturing phase may also comprise steps of assembling layers.

[0141] In addition, the material manufacturing phase may comprise a step of recording the holograms hpq, each hologram being associated with an emission point EPpq,

[0142] In particular, the step of recording the holograms hm can take into account the results of the step 180 of determining, for each distribution of emission points EPD^, the direction of the wave vector -pi of the light wave re- result emitted by the emission points EP^q of the distribution of emission points EPtfE

[0143] The recording of holograms is based on the interference of two groups of light beams on the holographic film. The generated interference fringes are stored physically or chemically on the holographic film resulting in a variation of the refractive index. The two groups of beams are conventionally called 'reference beams' and 'object beams'. For the encoding of hpq holograms of a distribution of EPDUV emission points (or the Reference beams correspond to the light waves extracted from the diffraction gratings optically coupled to the waveguides Sp and the electrodes eP selected to form the emission points EPpq (or EP^ of the distribution EPDuy (or EPDu^- The hologram recording step requires an optical recording system consisting of the formation of the object beams having an encoding angle with the emission surface S (or specifically the elementary emission zone zM)- The encoding angle / î then induces the direction of the wave vector (or

[0144]

[0145]

[0146]

[0147] yd). Object beams, coming from a coherent light source not shown on the figures, are obtained from the projection of a given mask through a convergence lens (La or Lb), as shown in Figure 12. Advantageously, for each distribution of emission points EPDUV (or EPD1^) different zones of the emission surface S (or specifically the elementary emission zone zv) are sequentially exposed to different object beams of specific encoding angle P, so as to form emission points associated with distinct emission directions. The discretization of the emission surface S according to the embodiments of the invention makes it possible in particular to facilitate and accelerate the step of recording the holograms during the material manufacturing phase of the device 10 compared to a non-discretized surface S. Indeed, the emission points EP^pq of the same distribution of emission points EPD\{v being contained in a single elementary emission zone Z'7 of a size smaller than the overall size of the emission surface S, the recording (or encoding) of the direction of the wave vector in the ho-kUv lograms associated with the emission points EP^ allows the coding of angles on larger values. As represented by the optical systems for recording holograms in Figure 12, the encoding of an area (z / j in Figure 12b) smaller than a surface (S in Figure 12a), induces an increase in the cone of the encoding angle A / j. This reduction in the encoding area also induces a reduction in the constraints on the temporal coherence of the light source generating the object beams. Figures 1, 2 and 4 show sets of waveguides Bp and electrodes of rectilinear configurations along the X axis and the Y axis respectively, each waveguide Sp being arranged perpendicular to the electrodes to obtain emission points EPpq regularly distributed according to a spatially periodic grid on the emission surface S. Alternatively, the set of Mx waveguides Sp of the image projection device 10 may be arranged in a configuration of non-rectilinear isolines in the plane defined by the emission surface S, as shown in [Fig. 13]. As used herein, the expression "non-rectilinear isolines" refers to a set of at least two geometric curves fl(X,Y) and f2(X,Y), defined in a plane (X,Y), such that the distance between a point pl belonging to the geometric curve fl(X,Y) and the geometric curve f2(X,Y) is the same regardless of the point pl, said distance being defined by the length of the shortest path between the point pl and a point of the geometric curve f2(XY) according to a reference metric dref. In payment, the distance between a point p2 belonging to the geometric curve f2(X,Y) and the geometric curve fl(X,Y) is the same regardless of the point p2, said distance being defined by the length of the shortest path between the point p2 and a point of the geometric curve fl(XY) according to the reference metric dref. Thus, non-rectilinear isolines correspond to a set of geometric curves, such that the curves of the set are always separated two by two by a constant distance. Such geometric curves can be parametric curves defined in the plane of the emission surface S. For a given geometric curve, an associated isoline can take on a very variable geometric curve appearance depending on the value of the metric dref considered.

[0148] Advantageously, the configuration of the electrodes on the emission surface S may be similar to the configuration of the waveguides. Thus, as shown in FIG. 13, the set of My electrodes of the image projection device 10 may be arranged according to a configuration of non-rectilinear isolines in the plane (X,Y) defined by the emission surface S.

[0149] In the case where the configuration of isolinear waveguides in the (X,Y) plane is defined according to a constant reference metric value dref\ between each of the 8P waveguides, the configuration may be a configuration of so-called regular isolinear waveguides. In the same way, in the case where the configuration of isolinear electrodes in the (X,Y) plane is defined according to a constant reference metric value dref between each of the electrodes, the configuration may be a configuration of so-called regular isolinear electrodes.

[0150] The reference metric values ​​drefv and dref can be variable between different waveguides Sp and different electrodes e <i respectivement. les variabilités de valeurs métrique référence permettent notamment d’améliorer la suppression des effets diffraction liés à d’éventuelles périodicités générées dans les configurations guides d’onde et d’électrodes.

[0151] Advantageously, the configuration of isolinear waveguides and / or the configuration of isolinear electrodes can be implemented using a method of the type "fast march" (or FMM, Fast-Marching Method according to the corresponding Anglo-Saxon expression) or "level lines" type (or LSM, Level-Set Method according to the corresponding Anglo-Saxon expression).

[0152] It should be noted that the FMM and LSM methods (James Sethian, Level Set Methods and Fast Marching Methods Evolving Interfaces in Computational Geometry, Fluid Mechanics, Computer Vision and Materials Science, Cambridge University Press, 1999) are numerical algorithms for solving boundary value problems of the Eikonal equation in order to follow the evolution of interfaces.

[0153] Figure 14 is a flowchart representing steps of the design phase of the method for manufacturing the device 10 and implemented to generate configurations of the set of waveguides and / or the set of electrodes on the emission surface S, according to embodiments of the invention.

[0154] In step 210, a calculation matrix is ​​initialized. The calculation matrix corresponds to the emission surface S, the size Demx and Demy, and is characterized according to predetermined calculation steps dcai and dcaiv. The initialized calculation matrix can then be a matrix comprising a number nbcai X ilbcaiY of points defined according to the following equations (23) and (24):

[0155]

[0156]

[0157] The calculation steps can be defined for example as a function of the width (gy) of the 8P waveguides and the width (ex) of the electrodes C(i, and / or as a function of the reference metric values ​​drefy and dref of the 8P waveguides and the electrodes En In particular, the calculation steps can be defined in such a way as to minimize the time required to generate these configurations while optimizing the calculation resolution. For example and without limitation, the calculation steps dcai and dcajY can be equal to the same quantity dcal between 1.5 pm and 250 nm.

[0158] In step 230, for the configuration of isolinear waveguides, an initial geometric curve of waveguide g0(X,Y) is determined and implemented in the initialized calculation matrix, as illustrated by the implementation example of [Fig. 15] (i.e. step 230). Equivalently, in step 230, for the configuration of isolinear electrodes, an initial geometric curve of electrode e0(X,Y) can be determined and implemented in the initialized calculation matrix.

[0159] The initial geometric curve of waveguide and / or electrode f0(X,Y), also called 'initial geometric curve', has a plurality of construction features. For example and without limitation, construction features may include: - an initial geometric curve f0(X,Y) defined by a non-rectilinear geometric function and extending on either side of the emission surface S, from an initial position (X0,Y0) to a final position (Xf,Yf) in the initialized calculation matrix, the initial position and the final position being arranged along an extension axis X' or Y', substantially parallel to the X or Y axis; in particular, as illustrated in Figure 15 (i.e. step 230), the initial geometric waveguide curve g0(X,Y) can be defined from an initial position (0,Y0) to a final position (nbcaj,Yf); in the same way, the initial geometric electrode curve e0(X,Y) can be defined from an initial position (XO, 0) to an end position (Xf, nbcaQ; - an initial geometric curve fO(X,Y) comprising only ordinary points called 'regular points' for which the partial derivatives of the curve along X and Y are not simultaneously zero, that is to say in this case the initial geometric curve fO(X,Y) does not include any stationary point called 'singular point' for which the partial derivatives of the curve along X and Y at this point are zero, the tangent at such a point then appearing undetermined; in particular, the initial geometric curve fO(X,Y) then does not include any point called 'double point' for which the curve "crosses back" on itself (or intersects) in the (X,Y) plane, nor any point called 'cusp point' for which the curve "returns" in the (X,Y) plane; - an initial geometric curve fO(X,Y) comprising at any point in the plane (X,Y) a radius of curvature greater than or equal to a predefined minimum radius of curvature rc.In particular, a minimum curvature radius ? cp associated with the Sp waveguides can be defined based on the physical properties of the Sp waveguides, and / or their respective manufacturing constraints. For example and without limitation, the minimum curvature radius zcp associated with the Sp waveguides can be equal to 20 pm; . - an initial geometric curve fO(X,Y) contained in a predetermined area of ​​the initialized calculation matrix; for example and without limitation, a so-called “content area” can be determined as a function of a number of points nbmaXj. or nbmaxy defined relative to the initial position (X0,Y0) and / or the final position (Xf,Yf) of the curve.

[0160] Advantageously, the initial geometric curve fO(X,Y) can be defined randomly while respecting one or more of the preceding construction characteristics.

[0161] The initial geometric curve f0(X,Y) may for example consist of a succession of segments. In particular, the initial geometric curve f0(X,Y) may comprise segments inclined at an angle of inclination b(X, Y) relative to the extension axis X' or Y' substantially parallel to the X or Y axis. The angle of inclination P(X, Y) may correspond to a constant quantity fi or to a variable function in the plane (X,Y), for example increasing, decreasing or any, and defined randomly, pseudo-randomly and / or as a function of the construction characteristics. Along the extension axis X' or Y', the angle of inclination / 5(X, Y) may be positive or negative. For example, the angle of inclination may be constant fi, and alternately positive and negative. The size of each inclined segment may be determined randomly, pseudo-randomly and / or as a function of the construction characteristics.The segments of the initial geometric curve f0(X,Y) can then be connected to each other by curved segments with a radius of curvature greater than or equal to the radius of . minimum curvature.

[0162] The initial positions (0,Y0) and (X0,0), and / or similarly the “content area” of the initial geometric curve, can be defined randomly, pseudo-randomly and / or defined for example so as to be centered in the initialized calculation matrix respectively along the Y axis and the X axis.

[0163] In step 250, a distance function may be applied to the initial geometric curve fO(X,Y) to generate the isoline waveguide configuration and / or the isoline electrode configuration. Such a distance function corresponds to a succession of steps making it possible to apply one or more reference metric values ​​dref between different curves forming a non-rectilinear isoline configuration, from the initial geometric curve fO(X,Y). For example and without limitation, a distance function used in step 250 may be the Fast-Marching type method.

[0164] By way of illustration, in step 250 for the configuration of isolinear waveguides, a distance map (or distance map according to the corresponding English expression) can be calculated from the Fast-Marching (or Level-Set) type method applied to the initial geometric curve f0(X,Y) in the calculation matrix. For each of the llbcai X nbcap points of the calculation matrix, the arbitrary isotropic distance metric is then calculated with respect to the closest point of the initial geometric curve f0(X,Y) implemented in the calculation matrix.

[0165] [Fig. 15] (i.e. step 250) illustrates the result of calculating a distance mapping for each point of the matrix with respect to an initial geometric waveguide curve g0(X,Y) determined in the previous step 230.

[0166] In step 270, the value of a minimum distance dmin can be determined.

[0167] In particular, in the case of determining the configuration of waveguides isolines, the minimum distance dmin corresponds to the constant and fixed spacing distance dmin.p between two 8P waveguides on the emission surface S. The value of dmitlrp can be defined with respect to the width (gy) of the waveguides Sp, for example according to the following expression (25):

[0168] ^w^ = 3xgy(25)

[0169] In the case of determining the configuration of isolinear electrodes, the minimum distance dmin corresponds to the constant and fixed spacing distance dmjn^ between electrodes on the emission surface S. The value of dmift-q can be defined in relation to the width (ex) of the electrodes e <i, par exemple selon l’expression (26) suivante :

[0170] d,wM=l,2xex(26)

[0171] Also in step 270, one or more multiples c of the value of the distance minimum dmin can be determined. The determination of these multiples can depend in particular on the size Demx and Demy of the emission surface S, and on the initial position (X0,Y0) of the initial geometric curve fO(X,Y).

[0172] For example and without limitation, in the case where the initial position (0,Y0) is fixed so as to be centered in the initialized calculation matrix respectively along the Y axis, the multiples cx dmin can be determined for the indices c of interval defined according to the following expression (27):

[0173] — [ p I 11 (27) [ ' 1 | j

[0174] In step 290, the set of isolines (also called isoline paths) of a configuration can be determined from the distance mapping and as a function of the multiples c of the value of the minimum distance dmill. In particular, for each order c, an isoline associated with the initial geometric curve fO(X,Y) implemented corresponds to the position of the points of the calculation matrix associated with the value closest to the distance cx dmin. Thus, in step 290, the waveguide configuration and / or said electrode configuration are determined.

[0175] Figure 15 (ie step 290) illustrates the result of a set of isolines associated with the initial geometric curve of waveguide g0(X,Y) corresponding to the isoline of order c = 0. The set of isolines determined can thus be obtained for the isolines defined from the orders c — [ - 7; - 1] and c — [ 1; 7] ■ In this example, each waveguide Sp of order P then corresponds to an isoline of order c.

[0176] The steps of generating the configurations of the set of waveguides Sp and of the set of electrodes on the emission surface S of the design phase of the method for manufacturing the device for projecting an image 10 make it possible in particular to minimize the degradations of the projected image linked to spatial periodicities between the emission points EPtxl and to optimize the number of emission points EPpq on the emission surface S to maximize the resolution of the image formed on the retina R, to obtain for example an optimized contrast.

[0177] In particular, the implementation of Fast-Marching (or Level-Set) type methods for calculating the distance maps at step 250 makes it possible to promote an aperiodic distribution of the intersections between the SP waveguides and the electrodes and therefore the aperiodic distribution of the different EPpq emission points, which leads to an improvement in the quality of the images.

[0178] Figure 16a represents the result of a simulation of the spatial distribution of the intensity of the percussion response (i.e. PSF) on the retina R formed by illumination points associated with distributions of emission points on the emission surface S defined from configurations of waveguides and electrodes obtained by translation of a segmented curve along the Y axis and the X axis respectively. tively. These configurations are called "segment translation" configurations hereinafter. Although such configurations improve the aperiodicity of the distribution of the different EPpq emission points compared to rectilinear configurations, such configurations do not allow to avoid diffraction effects during the formation of the image on the retina R since, as illustrated in the simulation in Figure 16a, the intensity of the perceptual response presents lines in the noise generated by the alignment of certain EPpq emission points.

[0179] Conversely, [Fig. 16b] represents the result of a simulation of the spatial distribution of the intensity of the percussion response (i.e. PSF) on the retina R formed by illumination points associated with distributions of emission points on the emission surface S defined from configurations of waveguides and isolinear electrodes. As illustrated in [Fig. 16b], such configurations make it possible to limit the diffraction effects during the formation of the image on the retina R and thus to obtain a more uniform noise.

[0180] Furthermore, the calculation of the number of intersections between the waveguides °P and the electrodes and therefore of emission points EPpq, for an emission surface S associated with isolinear waveguide and electrode configurations increases (in particular by more than 14%) compared to an emission surface S associated with waveguide and electrode configurations by “segment translations”, each emission surface S comprising in this case, the same numbers Mx of waveguides 8P and My of electrodes. This significant increase in emission points EPpq induces an increase in the density and therefore an increase in the possible number of pixels of the image to be projected and / or an increase in the intensity of the illuminated points on the retina R, while improving the aperiodic distribution of the different emission points EPpq,

[0181] Figures 17a and 17b represent the results of the comparison of image quality quantification criteria generated from two distinct emission surfaces S each associated with a random draw of a number of 25,000 emission points EPpq. The first emission surface S is associated with configurations of waveguides and electrodes by “segment translations”, and the second emission surface S is associated with configurations of waveguides and isolinear electrodes.

[0182] In particular, Figure 17a represents the results of the comparison of the power ratios F for each of the distinct emission surfaces S. As illustrated, the implementation of isolines makes it possible to reduce the power ratio F compared to configurations by “segment translations”, which means that this induces a reduction in the power of the intensity noise formed by the distribution of emission points compared to the power of the central intensity peak and thus an improvement

[0183]

[0184] of the quality of the projected image. Figure 17b shows the results of the comparison of the signal-to-noise ratios (SNR) for each of the distinct emission surfaces S. It should be noted that the SNR criterion, like the power ratio Y, allows to quantify the efficiency of the self-focusing of a distribution of emission points EPDUV to a point Ruv on the retina R of the eye. As illustrated in [Fig.8a], the SNR criterion can be evaluated from the difference between the height of the central intensity peak formed by the distribution of emission points and the height of the so-called 'secondary peak'. As shown in [Fig.17b], the implementation of isolines allows to increase the signal-to-noise ratio compared to configurations by 'segment translations', which also means an improvement in the quality of the projected image. According to certain embodiments, the emission surface S can be discretized into (Lx X Lv) elementary emission zones z'7 in the (X,Y) plane, as illustrated by Figure 18. Each elementary emission zone z" can be associated with a number mÿ of waveguides g'J and a number of electrodes eÿ, so as to generate a subset of (mÿ x inÿ ) emission points pp'^ defined among the (Mv x My) emission points EPpq of the emission surface S. Each elementary emission zone z'7 can then be associated with a finite number of emission point distributions EPD'^- Each emission point distribution EPDltiv of a zone elementary emission zone z'7 is then made up of one or more emission points EPp(f determined solely among the ( mÿ x mÿ ) emission points EPpq of the zone elementary emission zone z'7- For each elementary emission zone z'7, the sub- set of my waveguides g'J can be arranged according to a sub-configuration of non-rectilinear isolinear waveguides, and the subset of electrodes eÿ can be arranged in a sub-configuration of non-rectilinear isolinear electrodes.

[0185] Advantageously, for each elementary emission zone zA the sub-configuration of isolinear waveguides and / or the sub-configuration of isolinear electrodes can be generated from the application of the Fast-Marching (or Level-Sef) type method as described in particular by the steps of the design phase for the generation of the configurations.

[0186] For example and without limitation, for an elementary emission zone z'Â a sub-configuration of non-rectilinear isolines can be formed from an initial geometric curve extending in the plane of the emission surface S on either side of the elementary emission zone z'7, from an intermediate initial position to an intermediate final position arranged along an extension axis X' or Y' sen- possibly parallel to the X axis or the Y axis.

[0187] In the embodiments previously described, each emission point EPpq is formed from a hologram hm inscribed in a holographic film H and illuminated by light extracted from an 8p light guide. In these cases, each hologram hPq allows the phase control and the angle control of the extracted light wave at the level of the diffraction grating 1 p^- Other alternatives for forming (or configuring) the set of orientation elements and therefore the emission points can be envisaged.

[0188] In particular, the fourth layer (i.e. the holographic film H) can be replaced by a spatial light modulator (or SLM for Spatial Light Modulator according to the corresponding English expression) arranged opposite the diffraction grating. A SLM located at an emission point can be formed by liquid crystals for example. The SLM can then be configured (i.e. manufactured) to control the phase of the light wave extracted at the diffraction grating. In this case, each diffraction grating can be configured (i.e. manufactured) to control the angle of the light wave in addition to the extraction of the light wave. Alternatively, the fourth layer can additionally comprise a holographic film hologram hpq configured (i.e. manufactured) to control only the angle of the light wave.

[0189] It should be noted that certain features of the invention may have advantages when considered separately.

[0190] Those skilled in the art will easily understand that certain steps and sub-steps of the method described above can be carried out simultaneously and / or in a different order, for example in an order defined from the characteristics of the image projection device 10.

[0191] The device and methods described above according to the embodiments of the invention or sub-elements of this system can be implemented in various ways by hardware, software, or a combination of hardware and software, in particular in the form of program code which can be distributed in the form of a program product, in various forms.

[0192] The invention is not limited to the embodiments described above by way of non-limiting example. It encompasses all the variant embodiments that may be envisaged by those skilled in the art. In particular, those skilled in the art will understand that the invention is not limited to the different elementary emission zones and the different isoline configurations of the image projection device described by way of non-limiting example. In particular, certain embodiments of the invention may be combined.

Claims

Claims

1. Method for manufacturing a device for projecting an image (10) onto an eye, the device being defined in an orthogonal reference frame (X,Y,Z) and comprising an emission surface S extending generally in the plane (X,Y) of said orthogonal reference frame (X,Y,Z), the emission surface S comprising a stack of elements, said elements comprising a set of Mx waveguides Sp, a set of M vx My diffraction gratings and a set of My electrodes, Mx and My being positive integers whose product Mx x My is strictly greater than 1, each diffraction grating r <iP étant positionné à l’intersection d’un desdits guides d’ondes Sp et d’une desdites électrodes eP de manière à former un point d'émission EP!> q of a light wave, said set of Mx waveguides Sp being arranged in a waveguide configuration, and said set of My electrodes e <i étant agencé selon une configuration d’électrodes, le procédé comprenant phase de conception dudit dispositif (10) et fabrication matérielle ainsi conçu, caractérisé en ce que ladite comprend étape consistant à déterminer (290) guides d’ondes ou déterminée d’isolignes non rectilignes définie dans plan (x,y) défini par la surface d'émission s générant répartition apériodique des mx x my points eppq. [revendication 2] procédé pour d’un projection d'une image revendication 1, lequel l’étape détermination consiste lesdites configurations déterminées définies 3] l’une revendications 1 2, (230) courbe géométrique initiale s'étendant part d'autre un axe d'extension parallèle l'axe x y repère orthogonal (x,y,z), aléatoirement respectant au moins caractéristique construction courbe. 4] 3, est constituée d’une succession segments. 5] 3 4, appliquer (250) fonction distance générer isolignes d’électrodes isolignes. 6] 5, méthode type fast-marching. 7] 6, uniquement ordinaires dits ‘points réguliers’ lesquels les dérivées partielles ne sont pas simultanément nulles. 8] 7, tout point courbe, (x,y), rayon courbure supérieur égale minimum prédéfini. 9] 8, ledit propriétés physiques sp électrodes contraintes desdits desdites électrodes. 10] précédentes, discrétiser (120) lx ly zones d’émission élémentaires maillage continu chaque zone élémentaire z" sous-ensemble m(       ladite confi guration appliquée z'7 sorte former 11] 12] 13] sous-configuration rectilignes. 10, générées appliquant fast-marching y. dispositif sur œil, fabriqué partir du l'une quelconque 8 (x,y,z) s’étendant généralement empilement d’éléments, lesdits éléments ensemble 8p, réseaux diffraction m nombres entiers strictement supérieurs réseau positionné l’intersection ep manière epl>q of a light wave, said set of Mx waveguides Sp being arranged in a waveguide configuration, and said set of My electrodes e <i est agencé selon une configuration d’électrodes, ladite de guides d’ondes ou d’électrodes étant d’isolignes non rectilignes définie dans le plan (x,y) défini par la surface d'émission s et générant répartition apériodique desX My) EPpq emission points. A portable optical transparent data display system comprising an image projection device (10) according to the preceding claim, wherein said system is a system of glasses or an augmented reality headset.

Images