Method for designing an imaging device having high geometric distortion

Abstract

The invention relates to a method for designing an optical system comprising modifying an optical transfer function of the optical system so as to minimise differences between the optical transfer function of the optical system and a target optical transfer function, the target optical transfer function being characterised, for each light source point from a plurality of light source points A located upstream of the optical system and for each wavelength from at least one target wavelength, by: - a convergence of light through the optical system along its own axis U at a point Bu and with a focal length and / or a refractive power and / or an optical center position Cu along this axis U, this convergence varying according to the point A, this convergence being defined by: o at least one of Bu and Cu, and o at least one other parameter, different from that of the preceding bullet point, from Bu, Cu, the focal length and the refractive power.

Description

[0001] DESCRIPTION

[0002] TITLE: Method for designing an imaging device with high geometric distortion

[0003] Technical field

[0004] The present invention relates to a method of designing an imaging device.

[0005] State of the prior art

[0006] Smartphone camera modules are often multiple within a single smartphone. In fact, several camera modules with different focal lengths are often combined to create a very wide-angle module, a wide-angle module, and a narrower-angle module within the same smartphone. The images from the three modules can be digitally merged. This has the disadvantage of multiplying the number of camera modules, which is more expensive.

[0007] Similarly, a periscope-type module can be used, which allows a large displacement of a lens group in order to ensure a fairly significant optical zoom function in amplification. These modules cost more than a camera module simply equipped, if necessary, with an adjustment of the lens focusing distance.

[0008] In another field, anamorphic lenses are known, which are devices that do not project the image with the same compression ratio along two perpendicular axes. These lenses are used in cinema, in particular to convert a width / length ratio of the image taken according to the ratio of another support (for example 24x36 to 9x16). These devices have an inhomogeneous magnification ratio along the two axes X and Y, but homogeneous along one of these axes. The present invention cannot use this type of lens to achieve the intended objective, because it aims to obtain a different magnification in at least one or some areas of the image, in order to associate a zoom function with it.

[0009] The aim of the present invention is to propose a method for designing an imaging device, preferably an imaging device with high geometric distortion or minimal non-zero geometric distortion, relaxed and / or modulated, preferably with a view to obtaining a device allowing a zoom function while obtaining all or part of the technical advantages below:

[0010] - reduction in manufacturing cost, and / or

[0011] - better compactness, and / or

[0012] - mechanical simplification, and / or

[0013] - obtaining better sharpness for each of the three detected colors compared to the state of the art.

[0014] Statement of the invention

[0015] This objective is achieved with a method of designing an optical system, implemented by technical means, comprising:

[0016] 1) a modification of an optical transfer function of said optical system so as to minimize deviations between the optical transfer function of the optical system and a target optical transfer function, said target optical transfer function being characterized, for each point among several light source points A located upstream of the optical system and for each wavelength among at least one target wavelength, by:

[0017] - a convergence of light through the optical system along a proper axis U at a point Bu and with a focal length along the axis U and / or a curvature power along the axis U and / or an optical center position Cu along this axis U, this convergence varying along the point A, this convergence being defined by: o at least one among Bu and Cu, and o at least one other parameter different from that of the preceding paragraph among Bu, Cu, the focal length along the axis U and the curvature power along the axis U. and / or

[0018] 2) a minimization between the desired value and the value obtained of at least two parameters among:

[0019] - at least one of Bu and Cu, and

[0020] - at least one other parameter different from that of the previous paragraph among Bu, Cu, the focal length along the U axis and the curvature power along the U axis In addition:

[0021] - Said target optical transfer function can be characterized, for each point among several points A light sources located upstream of the optical system and for each wavelength among at least one target wavelength, by a convergence of the light through the optical system along a proper axis V different from the axis U, at a point Bv and with a focal length along the axis V and / or a curvature power along the axis V and / or an optical center position Cv along this axis V, this convergence varying according to the point A, and / or

[0022] - the minimization between the desired value and the value obtained of at least two parameters may comprise a minimization between the desired value and the value obtained of at least four parameters from: o at least one from Bu and Cu, and o at least one other parameter different from that of the previous paragraph from Bu, Cu, the focal length along the U axis and the curvature power along the U axis o at least one from Bv and Cv, and o at least one other parameter different from that of the previous paragraph from Bv, Cv, the focal length along the V axis and the curvature power along the V axis.

[0023] The convergence of light through the optical system along a proper axis U may be different from the convergence of light through the optical system along a proper axis V.

[0024] For each point among the several light source points A and each wavelength among the at least one target wavelength, the target optical transfer function and / or the desired value of at least two parameters may have points Bu and Bv merged or substantially merged.

[0025] For each point among the several light source points A and each wavelength among the at least one target wavelength, the target optical transfer function and / or the desired value of at least two parameters may have different focal lengths and / or curvature powers and / or optical center positions along the U axis and along the V axis.

[0026] The U and V axes can be perpendicular. The U axis can be a sagittal axis of the optical system.

[0027] The V axis can be a tangential axis of the optical system.

[0028] The optical system can be centrosymmetric.

[0029] The method according to the invention may comprise, before minimizing the deviations between the optical transfer function of the optical system and the target optical transfer function and / or minimizing between the desired value and the value obtained of at least two parameters, a definition of a material and / or an optical index of at least one lens making up the optical system.

[0030] The optical system may include at least two lenses, including:

[0031] - a first converging lens located closer to the upstream than to the downstream of the optical system, and

[0032] - a final converging or diverging lens, located closer to the downstream than the upstream end of the optical system.

[0033] The optical system may comprise an optical axis, said method further comprising taking into account an image sensor located downstream of the optical system, such that the optical transfer function of the optical system and / or the target optical transfer function and / or the desired value of at least two parameters and / or the obtained value of at least two parameters is arranged so that:

[0034] - the optical system receives light beams from the source points A and directs them towards the sensor, and preferably concentrates or converges them at the level of said sensor,

[0035] - the optical system has a variable magnification on the sensor, magnification which depends on the orientation and / or position of each optical beam incident on the optical system, and

[0036] - preferably, the magnification of the optical system comprises a minimum value on the sensor and a maximum value on the sensor, the ratio of the maximum of the magnification of the optical system to the minimum of the magnification of the optical system preferably being equal to at least 1.2.

[0037] The optical transfer function of the optical system and / or the target optical transfer function and / or the desired value of at least two parameters and / or the obtained value of at least two parameters can be arranged so that: - different light beams corresponding to a decrease in the magnification of the optical system are imaged in different parts of the sensor moving away from a point or segment or surface on the sensor, and / or

[0038] - that the magnification of the optical system decreases when the angle of the light beam incident on the optical system increases relative to the optical axis, and / or

[0039] - that the optical system comprises two lenses including: o A first lens, or a first face of the same single lens, preferably with constant or substantially constant magnification, and o A second lens, or a second face of the single lens, located respectively between the first lens or the first face of the single lens and the sensor, responsible at least in part for the variability of the magnification of the optical system and / or

[0040] - that the second lens is at a distance from the first lens equal to the focal length (preferably along the U axis) of the first lens plus or minus 20% of the value of this focal length of the first lens, and / or

[0041] - that the magnification of the second lens decreases when the angle of the light beam incident on the second lens increases relative to the optical axis, and / or

[0042] - that the magnification of the second lens includes a minimum value and a maximum value, the ratio of the maximum magnification of the second lens to the minimum magnification of the second lens being equal to at least 2, and / or

[0043] - that the optical system is symmetrical around the optical axis, and / or

[0044] - that the magnification of the optical system is maximum on a central or substantially central part of the sensor, and / or

[0045] - that the magnification of the optical system is minimal on a peripheral part of the sensor, and / or

[0046] - that the magnification of the optical system is minimal on a part of the sensor elongated in one direction. According to another aspect of the invention, a computer program is provided comprising instructions which, when executed in a computer, implement the steps of the method according to the invention.

[0047] According to another aspect of the invention, there is provided a computer-readable storage medium comprising instructions which, when executed by a computer, cause the computer to perform the steps of the method according to the invention.

[0048] Description of figures and embodiments

[0049] Other advantages and particularities of the invention will appear on reading the detailed description of implementations and embodiments which are in no way limiting, and the following appended drawings:

[0050] [Fig. 1] Figure 1 is a schematic side view of a first embodiment of an imaging device 1 according to the invention obtained by an embodiment of a method according to the invention, which is the preferred embodiment of the invention,

[0051] [Fig. 2a] Figure 2a is a schematic side view of an optical system illustrating the concept of focal length,

[0052] [Fig. 2b] Figure 2b is a schematic side view of the first embodiment of imaging device 1 according to the invention,

[0053] [Fig. 3] Figure 3 illustrates the variation in focal length of the optical system 3 of the first embodiment of the imaging device according to the invention, [Fig- 4] Figure 4 illustrates the distance to equivalent optical center, for the optical system 3 of the first embodiment of the imaging device 1 according to the invention, for a sharpness at 2m,

[0054] [Fig. 5] Figure 5 is a schematic side view of the first embodiment of imaging device 1 according to the invention,

[0055] [Fig. 6] Figure 6 is a schematic side view of the first embodiment of imaging device 1 according to the invention,

[0056] [Fig. 7] Figure 7 illustrates, in the plane of the sensor 2, different areas or parts of different magnifications of the optical system 3 on the sensor 2 in the first embodiment of the imaging device 1 according to the invention, [Fig. 8] Figure 8 illustrates, in the plane of the sensor 2, different areas or parts of different magnifications of the optical system 3 on the sensor 2 in a second embodiment of the imaging device according to the invention obtained by an embodiment of the method according to the invention,

[0057] [Fig. 9] Figure 9 illustrates, in the plane of the sensor 2, different areas or parts of different magnifications of the optical system 3 on the sensor 2 in a third embodiment of an imaging device according to the invention obtained by an embodiment of a method according to the invention,

[0058] [Fig. 10] Figure 10 illustrates, in the plane of the sensor 2, different areas or parts of different magnifications of the optical system 3 on the sensor 2 in a fourth embodiment of an imaging device according to the invention obtained by an embodiment of a method according to the invention, and

[0059] [Fig. 11] Figure 11 illustrates, in the plane of the sensor 2, different areas or parts of different magnifications of the optical system 3 on the sensor 2 in a fifth embodiment of an imaging device according to the invention obtained by an embodiment of a method according to the invention.

[0060] These embodiments being in no way limiting, it will be possible in particular to consider variants of the invention comprising only a selection of characteristics described or illustrated subsequently isolated from the other characteristics described or illustrated (even if this selection is isolated within a sentence comprising these other characteristics), if this selection of characteristics is sufficient to confer a technical advantage or to differentiate the invention compared to the state of the prior art. This selection comprises at least one preferably functional characteristic without structural details, and / or with only a part of the structural details if this part only is sufficient to confer a technical advantage or to differentiate the invention compared to the state of the prior art.

[0061] We will first describe, with reference to figures 1 to 7, an embodiment of the method according to the invention.

[0062] This embodiment of a method for designing an optical system 3, implemented by technical means 10, comprises a modification of an optical transfer function of said optical system 3 so as to minimize deviations between the optical transfer function of the optical system 3 and a target or desired optical transfer function, said target optical transfer function being characterized (or the optical system being characterized), for each point among several light source points A located upstream of the optical system and for each wavelength among at least one target wavelength, by:

[0063] - a convergence of light through the optical system 3 along a proper axis U at a point Bu and with a focal length Fu and / or a curvature power PC rbu = 1 / Fu and / or an optical center position Cu along this axis U, this convergence varying according to point A, this convergence being defined by: o at least one among Bu and Cu, and o at least one other parameter different from that of the previous paragraph among Bu, Cu, the focal length along U and the curvature power along U.

[0064] Said target optical transfer function (or the optical system) is preferably further characterized, for each point among several light source points A located upstream of the optical system 3 and for each wavelength among at least one target wavelength, by a convergence of the light through the optical system 3 along a proper axis V different from the axis U, at a point Bv and with a focal length Fv and / or a curvature power PC rbv = 1 / Fv and / or an optical center position Cv along this axis V, this convergence varying according to point A.

[0065] Cu, Cv, Bu, and Bv are for example illustrated in Figures 5 and 6.

[0066] In the absence of the proper axis V different from the axis U, the point Bv, the focal length Fv, the curvature power 1 / Fv and / or the optical center position Cv along this axis V, the proper axis U, the point Bu, the focal length Fu, the curvature power 1 / Fu and / or the optical center position Cu along this axis U may be called respectively proper axis, point B, focal length F or f, curvature power 1 / F and / or optical center position C in the remainder of the description.

[0067] The means 10 are calculation or simulation means. These means 10 comprise at least one computer, a central or calculation unit, an analog electronic circuit (preferably dedicated), a digital electronic circuit (preferably dedicated), and / or a microprocessor (preferably dedicated), and / or software means. This minimization of the differences between the optical transfer function of the optical system 3 and a target optical transfer function can be carried out from CAD (computer-aided design) software according to a known basis of the state of the art but improved so as to take into account the new constraints linked to the present invention.

[0068] There are many known examples of optical system CAD software (such as, but not limited to, the commercial software OpticStudio from Zemax).

[0069] Such optical system CAD software, including within the scope of the present invention and implemented by the means 10, comprises a step of describing the objective 3 to be designed, and the input light beams. Certain parameters are specified through an interface:

[0070] - the number of lenses 31, 32, at least equal to one, preferably greater than or equal to two, this parameter subsequently being preferably considered as fixed and therefore not free;

[0071] - the shape (convex, concave, free) of each surface of each lens 31, 32, this parameter subsequently being preferably considered as fixed and therefore not free except in the case of a “free” value;

[0072] - the equation of each lens surface, i.e. for example 9 coefficients with the conicity K and the coefficients A k of order 2 to 16 (for k=1 to 8), these parameters being subsequently preferably considered as free to allow the method according to the invention to optimize them, but could also be at least partly fixed, at least for certain faces of certain lenses.

[0073] - the material and / or optical index (as a function of the wavelength, in particular for each wavelength among the at least one target wavelength) of each lens 31, 32, these parameters subsequently being preferably considered as fixed and therefore not free;

[0074] - the distance between each lens, this parameter being subsequently preferably considered free;

[0075] - the thickness of each lens, this parameter being subsequently preferably considered as free;

[0076] - the possible central symmetry of the system 3, each lens 31, 32 being preferably perpendicular(s) to an axis 4 which we will also call Z or ZZ', which also constitutes the axis of symmetry (by rotation) of the lenses 31, 32 and of the objective 3 which results from it by assembly of said lenses 31, 32, this parameter being subsequently preferably considered as fixed and therefore not free.

[0077] Also, another interface allows you to specify a distance where the points A of the scene are located which illuminate the objective 3, and a list of angles or coordinates (X and Y coordinates) in a plane perpendicular to the ZZ' axis.

[0078] For each previously defined light entry point A, the CAD software and / or method embodiment according to the invention calculates (by the means 10) the propagation of a set of light beams coming from this point.

[0079] There are two possible representations of propagating light: either we consider a propagating wavefront, or we consider a set of propagating lines or beams:

[0080] The light beam can thus be described as a wavefront propagating in a continuous medium, or discontinuous through interfaces between two different materials, in particular whose refractive index, which generally depends on the wavelength, changes abruptly in the vicinity of said interface. In particular, we know the Snell-Descartes law to model the propagation of a wavefront initially locally before the interface, to obtain it after the interface in a medium with a different refractive index.

[0081] Alternatively, it is also possible to consider the normals to said wavefront, these normals drawing as many light beams. It is possible to characterize these normals by their local angular variation, for example along two principal axes, along which two coefficients of derivative of the angle of said normal change with respect to a displacement in a certain direction with respect to said normal. A second parameter to consider is the optical power density distribution of all said beams. This power density evolves by propagation according to the conservation of radiated energy, (along a cone delimited by a set of beams, for example) or its attenuation if the material has a non-zero attenuation, or even an amplification if the material has excited atoms ready to de-excite in the case of a light-emitting medium.

[0082] Of course, in either case, the propagation is calculated by respecting Maxwell's equations in a differential form, or an integrated form, describing the propagation of an electromagnetic wave.

[0083] The CAD software and / or method embodiment according to the invention comprises a calculation core modeling in one form or another (wavefront or beam) the propagation of light energy, in the materials and through the optical interfaces encountered.

[0084] An assembly of light elements, such as a set of lenses 31, 32 forming a lens 3, can be described by a set of surface equations representing each face of each lens 31, 32. For lenses 31, 32 with circular symmetry, said surfaces can be modeled by the rotation of a curve, around the axis normal to said surface, over 360°. The curve is generally described by a term of order 2 (coefficient times the distance r to the normal axis, squared). Such a term is divided by a term taking into account the conical, spherical, parabolic character of said surface, through a coefficient K called conicity.Finally, this equation is added by N terms in r to the power of 2*k, k being a natural integer, generally ranging from 2 to 8, making it possible to deviate from the said surface from its shape previously defined in the vicinity of the axis of symmetry by the first term of order 2 and conicity K, so as to obtain any shape, in particular for large values ​​of r.

[0085] In such an equation, we therefore have a coefficient A2 in front of the term r 2, the term K, the set of coefficients A?k in front of the terms of order 2k, which are all parameters defining the shape of the curve, and therefore of the surface of each lens of the system 3. Then, some or all of these terms can be declared as “free” i.e. free to be optimized by the CAD software and / or method embodiment according to the invention, in order to optimize the shape of the lenses 31, 32. The equation of each face of the lens can therefore be written as a function z( ​​) returning the distance between a plane perpendicular to the optical axis tangent to the apex of the considered face of the lens as a function of the distance r of a point distant from r to the optical axis of the lens according to: Likewise, the refractive indices of the materials of the lenses 31, 32, the thicknesses at the center of the lenses 31, 32, the distances between the lenses 31, 32, also called air thickness, or air gap in English, are all “free” parameters capable of being optimized by the CAD software.

[0086] From these elements, for each point A of the scene from which the light comes, or for a set of such points, the CAD software and / or method embodiment according to the invention can calculate a so-called optical transfer function, the aim of which is to know how a radiated light intensity is concentrated on the sensor 2 where the image is expected to be formed, both in intensity ratio and in spatial spread.

[0087] Such a function depends, among other things, on where it is considered. In particular, there is a distance Z (for the optical axis associated with the coordinate ZZ' of a reference frame), called the convergence distance, for which the function has the smallest spread of light power density. Ideally, the distance Z obtained is that where the sensor for detecting the image formed from the scene is placed. In the case where the sensor 2 is a plane, which is not essential, the CAD software and / or method embodiment according to the invention must ensure that for several light points in the scene, the respective convergence points are with a Z close to each other, which makes it possible to place the sensor 2 at this common Z. The CAD software and / or method embodiment according to the invention can even automatically calculate such a Z from the transfer functions resulting from the different points in the scene.

[0088] For this common Z, as for any other Z, it is possible to calculate what is called a transfer function, which characterizes the power ratio between the power at the sensor 2, and that coming from the light spot, and the spread of this power in a plane perpendicular to the ZZ' axis. This spread is called PSF or "Point Spread Function". Ideally, this function should not spread over a distance greater than the size of a pixel of the detector 2, so as to obtain maximum sharpness for the image formed, that is to say that for example, 95% of the light power received should be collected by a single pixel (or at most 2 pixels in X and 2 pixels in Y if the center of the light spot is between these 4 pixels), the remaining 5% being collected mainly by a few pixels near said convergence pixel.From these design elements, the implementation of the method embodiment according to the invention can be done by adding the following elements to the optical system CAD software previously described.

[0089] It is appropriate to decompose the set of propagated beams (or the propagated wavefront, depending on the representation chosen), at the output of the set of optical elements 3, 31, 32 crossed by the system 3, according to two axes U and V called principal axes. In the case of circular symmetry, which is only a possible particular case of implementation of the invention, the radial axis can advantageously be considered as being the U axis, and the so-called tangential axis V also perpendicular to U, which are naturally the two principal axes in the case of circular symmetry.

[0090] From these U and V axes, we can search, for U, for the best Z of beam convergence, that is to say a plane of coordinate Z (generally perpendicular to the principal or average axis of propagation of the beams) where the PSF function (for Point Spread Function) is the least spread in terms of optical power in the direction U. We define this Z as being a distance called Z u .

[0091] Similarly, we can look for a better Z distance according to which the distribution of power along the V axis is the least spread. We call this Z distance v .

[0092] In addition, we can define a point called B u whose Z coordinate is Z u and which is the center of the light spot obtained for Z u . The same applies to point Bv.

[0093] For a succession of several source points A close enough together, we obtain a succession of several convergence points Bu in the direction of the axis U for each of these points (the axis U can evolve between the points A). We can then define a so-called focal distance along U, and local to one of these points, which is linked to a notion of local magnification.

[0094] The magnification G is related to the focal length f by the relation:

[0095] Let dr be the thickness of a point in the scene located at a distance z from the lens.

[0096] The angle related to this thickness is da = dr / z. This angle corresponds to a distance of on sensor 2, as being da = dc / f. Thus, da = dr / z = dc / f therefore dc / dr = f / z. Calling this term G dc / dr or local magnification factor (much less than 1 because a scene of several meters is projected onto a few tens of millimeters). The larger this term, the more the scene is magnified on sensor 2.

[0097] So G = f / z. G is therefore proportional to the local magnification of the image. The focal length is also proportional. The magnification is inversely proportional to the distance z from the lens.

[0098] Thus, the focal length Fu is proportional to the local magnification factor along the U axis.

[0099] For a so-called main beam connecting a point A to point Bu, it is possible to define a point Cu at a distance Fu from point Bu, on this so-called main beam (defined above as, for example, the beam of maximum power density).

[0100] Similarly, for a so-called main beam connecting a point A to point Bv, it is possible to define a point Cv at a distance Fv from point Bv, on this so-called main beam (defined above as, for example, the beam of maximum power density).

[0101] From these elements, consider for example angles a, and © describing the angle a to the optical axis of the lens, and ® the rotation from 0 to 360° of this axis connecting point A of the scene to the lens. It is then possible to obtain several functions of Alpha, giving:

[0102] • Point Bu as a function of a,

[0103] • The focal length Fu in fct of a,

[0104] • The Cu point as a function of a.

[0105] • Same for Bv, Fv, Cv.

[0106] We may then wish to design an objective 3 including:

[0107] • The Bu points have a neighboring Z coordinate, ideally equal to a common Z value, because the optical sensors are generally planar, planar placed at a particular Z distance from the objective 3, which allows the Bu points to belong to the surface of the sensor. Or more generally, the Bu points belong to a common surface equal to that of the optical sensor 2, for the more general case where it would not be planar.

[0108] • A certain profile of Fu according to the angle a, and for example independent of 0 for a circular symmetry. Fu can also depend on the angle © in a more general case.

[0109] Then the CAD software and / or method embodiment according to the invention performs a calculation (via the means 10) of propagation providing a set of propagation functions through the lens. A calculation module linked to the invention performs the extraction of the points Bu, distance Fu for each point A(a).

[0110] Concerning the minimization of the deviations between the optical transfer function of the optical system 3 and the target optical transfer function, we set up a deviation indicator IE, that is to say a number, relating the deviation between the different quantities representative of the transfer function that we seek to control. Thus, we can choose to control the deviation between the curve Fuo(a) and the curve Fus(a) as well as the deviation between the projections of the points Bu(a) and Bv(a) on the z axis, that is to say the deviation between their coordinates along Z, named Buo.z(a) and Bvo.z(a). The IE indicator then contains elements of the Fuo(a) curve and the Fus(a) curve, respectively the curve obtained at a state of progress of the optimization calculation of the objective, and the Fus curve provided as input as the desired curve, and respectively the z coordinates of the Buo and Buv points of convergence along the U axis and along the V axis.

[0111] Such a number IE should ideally be zero when the curves are identical, and all the larger and more positive as there are more deviations between the curves, at more points. It is sufficient to sum the squared deviations of (Fuo(a) - Fus(a)) divided by a normalization term NFu, on a discrete set of a to obtain a first part of IE, as well as the squared deviations of (Buo.z(a) -Bvo.z(a)) on the same discrete set of alpha, divided by a normalization term NBu to obtain the second part of IE. If one wishes to establish the optimization on more terms, the calculation of IE can be extended to these other terms. Thus, we can still add for IE a third term of comparison of Buo.z(a) to Bus.z(a), with Bus.z(a) possibly constant with respect to a according to the same principle, to obtain the convergence on the same plane where the plane of the sensor is located, for all angles a, and for the axes U and V.

[0112] The several normalization terms (e.g. NFu and NBu) can take the value 1, or preferably selected values ​​of possible range of the different parameters, respectively. Their main purpose is to make IE sensitive in a comparable way to the different variations of each term, and also to mask possible effects of different choices of units between the different optimized quantities which would make the different contributions to IE unbalanced. IE can be derived with respect to the set of parameters left free for the description of objective 3.

[0113] Then, the CAD software and / or method embodiment according to the invention can modify the parameters left free (such as the coefficients Ak,j (k even index from 2 to 16, j index of the face encountered, for example from 1 to 4), the Kj conicity of each face encountered, the E£ thickness of the lens £ (£ = 31 or 32 in our example) which each represent a term of the polynomial development which defines the generator which defines the shape of the face j considered of one of the lenses (j = 1 or 2 for the first lens 31, j=3 or 4 for the second lens 32), the AGw air space or “Air gap” of index w (w being a positive integer numbering the air gap, knowing that the number of air gaps is equal to the number of lenses minus one) measured at the center of the lenses between the respective apices of the lens n face 2 of exit and the lens n + 1 face 1 input (for example between lens 31 and lens 32), the optical indices or group index Ne of each lens £.), for example using a so-called gradient method (iterative gradient method, with simultaneous variation of different free parameters) or any other technique, in order to evolve said parameters left free to guide the design towards obtaining an objective that best approaches the smallest possible IE indicator, and therefore a Fuo(a) curve close to the Fus(a) curve that we wish to obtain, as well as a Buo.z(a) curve close to Bvo.z(a) and fairly constant (therefore also close to a desired Bus.z(a) curve, constructed as being constant).

[0114] As it will generally not be possible to perfectly obtain the desired Fus(a) curve, for example, it is appropriate to provide the CAD software and / or method embodiment according to the invention with several modes of representing the curves as a function of a for (X, Y and Z forming three axes of an orthogonal or orthonormal reference frame, Z preferably being the same as the ZZ' or 4 axis):

[0115] Fus(a) desired local focal length (near angle a) of the optical system 3 for the U axis,

[0116] Fuo(a) local focal length (near angle a) obtained from optical system 3 for axis U,

[0117] Bus.x(a) the desired coordinate along the X axis of the point Bu of convergence along the U axis of the beam emitted by a point A(a), Bus.y(a) the desired coordinate along the Y axis of the point Bu of convergence along the U axis of the beam emitted by a point A(a),

[0118] Bus.z(a) the desired coordinate along the Z axis of the convergence point Bu along the U axis of the beam emitted by a point A(a),

[0119] Buo.x(a) the coordinate obtained along the X axis of the point Bu of convergence along the U axis of the beam emitted by a point A(a),

[0120] Buo.y(a) the coordinate obtained along the Y axis of the point Bu of convergence along the U axis of the beam emitted by a point A(a),

[0121] Buo.z(a) the coordinate obtained along the Z axis of the point Bu of convergence along the U axis of the beam emitted by a point A(a),

[0122] Cus.x(a) the desired coordinate along the X axis of the optical center C along the U axis of the optical system 3,

[0123] Cus.y(a) the desired coordinate along the Y axis of the optical center C along the U axis of the optical system 3,

[0124] Cus.z(a) the desired coordinate along the Z axis of the optical center C along the U axis of the optical system 3,

[0125] Cuo.x(a) the coordinate obtained along the X axis of the optical center C along the U axis of the optical system 3,

[0126] Cuo.y(a) the coordinate obtained along the Y axis of the optical center C along the U axis of the optical system 3,

[0127] Cuo.z(a) the coordinate obtained along the Z axis of the optical center C along the U axis of the optical system 3, and the same on the V axis (with .x, .y and .z for the x, y and z coordinates of the points Bu and Cu), in order to be able to visualize and compare the desired profiles, those obtained, and also visualize them simultaneously on a common graph. We can also visualize the difference between the index s and the index o for each or some of these calculated parameters.

[0128] This provides the user of the software implementing the invention with a set of visualization elements in order to qualify the quality and relevance of the results obtained, and where appropriate to modify his wishes as a function of the desired index quantities, to take into account the physical constraints of production.

[0129] Thus, in this embodiment of the method according to the invention, the minimization implemented between the desired value (target and fixed value) and the obtained value (which is variable and preferably obtained iteratively) of at least two parameters among: - at least one among Bu and Cu, and

[0130] - at least one other parameter different from that of the previous paragraph among Bu, Cu, the focal distance along U and the curvature power along U preferably:

[0131] - the minimization between the desired value Fus(a) of the focal length and the obtained value Fuo(a) of the focal length, and the minimization between the desired value Bus(a) (preferably only Bus.z(a) which is sufficient) of the position Buo(a) of the point Bu and the obtained value of the point Bu (preferably only Buo.z(a) which is sufficient), corresponds to minimizing the deviations between the optical transfer function of the optical system 3 and the target optical transfer function, since each of the optical transfer function of the optical system 3 and the target transfer function is characterized, for each point among several points A light sources located upstream of the optical system and for each wavelength among at least one target wavelength, by a convergence of light through the optical system 3 along a proper axis U at a point Bu and with a focal length Fu and / or a curvature power P bu = 1 / Fu and / or an optical center position Cu along this axis U, this convergence varying according to the point A, this convergence being defined by: o at least one among Bu and Cu, and o at least one other parameter different from that of the preceding paragraph among Bu, Cu, the focal length along U and the curvature power along U.

[0132] In this embodiment of the method according to the invention, the minimization between the desired value (target and fixed value) and the obtained value (which is variable and preferably obtained iteratively until this minimization is finalized below a predefined threshold) of at least two of:

[0133] - at least one of Bu and Cu, and

[0134] - at least one other parameter different from that of the preceding paragraph among Bu, Cu, the focal length along U and the curvature power along U is preferably done by minimizing an indicator of global deviation on the set of these at least two parameters which is preferably equal to the sum (but one could take a polynomial, a multiplication, the square root of the sum of the squares, or any combination of the preceding examples or any mathematical function taking into account the influence of these parameters possibly with different weights) of individual indicators of different parameters,each individual indicator of a given parameter being equal or dependent or proportional or representative of a deviation or difference between the desired value (target and fixed value) of this parameter and the obtained value (which is variable and preferably obtained iteratively until this minimization is finalized below a predefined threshold) of this parameter, these different parameters including,

[0135] - at least one of Bu and Cu, and

[0136] - at least one other parameter different from that of the previous paragraph among Bu, Cu, the focal length along U and the curvature power along U.

[0137] Likewise, in this embodiment of the method according to the invention, the minimization implemented between the desired value (target and fixed value) and the obtained value (which is variable and preferably obtained iteratively) of at least four parameters among:

[0138] - at least one of Bu and Cu, and

[0139] - at least one other parameter different from that of the previous paragraph among Bu, Cu, the focal length along U and the curvature power along U, and

[0140] - at least one of Bv and Cv, and

[0141] - at least one other parameter different from that of the previous paragraph among Bv, Cv, the focal distance according to V and the curvature power according to V preferably:

[0142] - the minimization between the desired value Fus(a) of the focal length and the obtained value Fuo(a) of the focal length, and the minimization between the desired value Bus(a) (preferably only Bus.z(a) which is sufficient) of the position Buo(a) of the point Bu and the obtained value of the point Bu (preferably only Buo.z(a) which is sufficient), and

[0143] - the minimization between the desired value Fvs(a) of the focal length and the obtained value Fvo(a) of the focal length, and the minimization between the desired value Bvs(a) (preferably only Bvs.z(a) which is sufficient) of the position Bvo(a) of the point Bv and the obtained value of the point Bv (preferably only Bvo.z(a) which is sufficient), corresponds to minimizing the deviations between the optical transfer function of the optical system 3 and the target optical transfer function, since each of the optical transfer function of the optical system 3 and the target transfer function is characterized, for each point among several points A light sources located upstream of the optical system and for each wavelength among at least one target wavelength, by a convergence of light through the optical system 3:

[0144] - along a proper axis U at a point Bu and with a focal length Fu and / or a curvature power PC rbu= 1 / Fu and / or an optical center position Cu along this axis U, this convergence varying according to point A, this convergence being defined by: o at least one among Bu and Cu, and o at least one other parameter different from that of the previous paragraph among Bu, Cu, the focal length along U and the curvature power along U

[0145] - along a proper axis at a point Bv and with a focal length Fv and / or a curvature power pc rbv = 1 / Fv and / or an optical center position Cv along this axis V, this convergence varying according to point A, this convergence being defined by: o at least one among Bv and Cv, and o at least one other parameter different from that of the previous paragraph among Bv, Cv, the focal length along V and the curvature power along V

[0146] In this embodiment of the method according to the invention, the minimization between the desired value (target and fixed value) and the obtained value (which is variable and preferably obtained iteratively until this minimization is finalized below a predefined threshold) of at least four parameters among:

[0147] - at least one of Bu and Cu, and

[0148] - at least one other parameter different from that of the previous paragraph among Bu, Cu, the focal length along U and the curvature power along U, and - at least one among Bv and Cv, and

[0149] - at least one other parameter different from that of the preceding paragraph among Bvu, Cv, the focal length according to V and the curvature power according to V, is preferably done by minimizing an indicator of global deviation on the set of these at least four parameters which is preferably equal to the sum (but one could take a polynomial, a multiplication, the square root of the sum of the squares, or any combination of the preceding examples or any mathematical function taking into account the influence of these parameters possibly with different weights) of individual indicators of different parameters,each individual indicator of a given parameter being equal or dependent or proportional or representative of a deviation or difference between the desired value (target and fixed value) of this parameter and the obtained value (which is variable and preferably obtained iteratively until this minimization is finalized below a predefined threshold) of this parameter, these different parameters including,

[0150] - at least one of Bu and Cu, and

[0151] - at least one other parameter different from that of the previous paragraph among Bu, Cu, the focal length along U and the curvature power along U, and

[0152] - at least one of Bv and Cv, and

[0153] - at least one other parameter different from that of the previous paragraph among Bv, Cv, the focal length according to V and the curvature power according to V.

[0154] Let R be the distance equal to the projection, along the U axis, of the vector connecting point A to Cu and being equal to 0 when A is located on the Z axis.

[0155] R can be defined in the plane of sensor 2, a is the angle of light. R and a are related: there is preferably a bijection between the two. Different parameters, including Cu, Cv, Bu, Bv, , PCrbu, PG-bv, can be described as a function of one or the other. R is more intrinsic to the plane of sensor 2 for a certain sensor-lens distance setting, a is more intrinsic to the lens 3. Between R and a, one can replace the other as a scanning parameter.

[0156] For a centrosymmetric design, we can simply generate the parameters of the local points Cu, Cv, Bu, Bv, and the curvature power PCrbu, PCrbv as a function of the distance R to the center of the optical system 3 from point A of the input light. From these parameters, it is possible and desirable to plot several curves as a function of the input data, for example R. What will also be interesting is to plot the gaps between Bu and Bv, or between the positions respectively z(Bu) and z(Bv) of each of these points along the Z axis, which we seek to minimize or ideally cancel to have the same convergence along the 2 axes U and V, and therefore the simultaneous sharpness in the plane of the screen at a distance z(Bu) = z(Bv).Furthermore, we seek to have a z(Bu) independent of R, so that the convergence takes place in the same plane for all R, therefore all the entry points A with all possible angles (and preferably for a choice of a particular fixed distance z(A), therefore an entry plane).

[0157] From PCrbu, PCrbv, we can calculate Fu = 1 / PCrbu and Fv = 1 / PCrbv the focal lengths along each axis.

[0158] For a non-centrosymmetric design, it is necessary to consider the variations of all these variables (Fus, Bus, Cus, Fuo, Buo, Cuo, for the .x, .y, .z components) according to, for example, the x and y coordinates (perpendicular to each other and to the Z axis) of point A, or R and 0 of the entry point (0 being the angle which, with R, allows to describe a plane perpendicular to the plane of figure 6, for example in the plane of the sensor). This makes surfaces and not curves to be represented, which can be visualized in different ways, including sections along particular axes to return to 2D representable on a sheet or a screen.

[0159] Of course, these different visualization modes are to be considered according to the wavelength, or wavelength band, which will correspond for example to the R, G, B detection bands of the Bayer matrices of the optical sensor 2 which will be associated with the designed objective 3. We can also represent differences according to 2 wavelength bands or two wavelengths, to visualize the chromatic effects on the current design.

[0160] The invention provides tools for understanding the design of optical systems 3 for which the geometric distortion is deliberately relaxed, and even modulated to obtain a particular optical function.

[0161] Likewise, the invention provides tools for understanding the design of optical systems 3 for which voluntarily the minimization of chromatic aberrations is relaxed, in order to use the degrees of freedom gained to better preserve sharpness as a function of R or a, by modulating where appropriate the magnification and therefore the geometric distortion, at a particular intensity of modulation thus controlled.

[0162] These tools according to the invention allow in particular a synthetic representation of the 2 natural modes of convergence of the light according to the 2 natural axes, in each region of the field, in order to make them coincide, in the same plane which becomes the sensor plane 2 or a plane parallel to the latter.

[0163] They also allow a synthetic representation of the local magnification factor, or according to another representation, of the local focal length, therefore of the local optical performances in the field of the sensor.

[0164] The CAD software and / or method embodiment according to the invention may include different variants which can be considered alone or in combination:

[0165] - the convergence of light through the optical system 3 along the proper axis U may be different from or identical to the convergence of light through the optical system along the proper axis V, and / or

[0166] - for each point among the several light source points A and each wavelength among the at least one target wavelength, the target optical transfer function may have either points Bu and Bv which are merged or substantially merged or points Bu and Bv which are distinct, and / or

[0167] - for each point among the several light source points A and each wavelength among the at least one target wavelength, the target optical transfer function can have:

[0168] • either different focal lengths along the U axis and along the V axis, or identical or substantially identical focal lengths along the U axis and along the V axis, and / or

[0169] • either different curvature powers along the U axis and along the V axis, or identical or substantially identical curvature powers along the U axis and along the V axis, and / or

[0170] • either positions of different optical centers along the U axis and along the V axis, or positions of identical or substantially identical optical centers along the U axis and along the V axis

[0171] , and / or

[0172] - the U and V axes may be perpendicular or oblique, and / or - the U axis may be a sagittal axis of the optical system, and / or the V axis may be a tangential axis of the optical system, and / or

[0173] - the optical system 3 can be centrosymmetric or not.

[0174] The optical system 3 preferably comprises at least two lenses 31, 32, including:

[0175] - a first converging lens 31 located on the upstream side of the optical system 3, i.e. closer to the upstream than to the downstream side of the optical system 3, and

[0176] - a final converging or diverging lens 32, located on the downstream side of the optical system 3, i.e. closer to the downstream than the upstream side of the optical system 3.

[0177] The optical system 3 thus designed comprises the optical axis 4, and this embodiment of the method according to the invention therefore preferably comprises taking into account an image sensor 2 located downstream of the optical system 3, so that the optical transfer function of the optical system and / or the target optical transfer function is preferably arranged so that:

[0178] - the optical system 3 receives light beams from the source points A and directs them towards the sensor, and preferably concentrates them or converges them at the level of said sensor,

[0179] - the optical system 3 has a variable magnification on the sensor, magnification which depends on the orientation and / or the position of each optical beam incident on the optical system, and

[0180] - preferably, and for information purposes, the magnification of the optical system 3 comprises a minimum value on the sensor and a maximum value on the sensor, the ratio of the maximum magnification of the optical system to the minimum magnification of the optical system preferably being equal to at least 1.2.

[0181] Preferably, the optical transfer function of the optical system and / or the target optical transfer function:

[0182] - is arranged so that different light beams corresponding to a decrease in the magnification of the optical system are imaged in different parts 51, 52, 53 of the sensor moving away from a point 6 or a segment 7 or a surface on the sensor, and / or - is arranged so that the magnification of the optical system decreases when the angle of the light beam incident on the optical system increases relative to the optical axis, and / or

[0183] - is arranged so that the optical system comprises two lenses including: o A first lens 31, or a first face of the same single lens, preferably with constant or substantially constant magnification, and o A second lens 32, or a second face of the single lens, located respectively between the first lens or the first face of the single lens and the sensor, responsible at least in part for the variability of the magnification of the optical system. and / or

[0184] - is arranged so that the second lens 32 is at a distance from the first lens 31 equal to the focal length (along the U axis) of the first lens at plus or minus 20% of the value of this focal length of the first lens 31, and / or

[0185] - is arranged so that the magnification of the second lens 32 decreases when the angle of the light beam incident on the second lens increases relative to the optical axis, and / or

[0186] - is arranged so that the magnification of the second lens 32 comprises a minimum value and a maximum value, the ratio of the maximum of the magnification of the second lens 32 to the minimum of the magnification of the second lens 32 being equal to at least 2, and / or

[0187] - is arranged so that the optical system 3 is symmetrical around the optical axis 4, and / or

[0188] - is arranged so that the magnification of the optical system 3 is maximum on a central or substantially central part of the sensor 2, and / or

[0189] - is arranged so that the magnification of the optical system 3 is minimal on a peripheral part of the sensor 2, and / or

[0190] - is arranged so that the magnification of the optical system is minimal on a part of the sensor elongated in one direction. This embodiment of the method according to the invention is implemented by:

[0191] - A computer program embodiment according to the invention, comprising instructions which, when executed in a computer, implement the steps of the method embodiment according to the invention, and / or

[0192] - A computer-readable storage medium embodiment of the invention comprising instructions which, when executed by a computer, cause the computer to perform the steps of the method embodiment of the invention.

[0193] Finally, it is noted that the method according to the invention makes it possible to obtain better sharpness for each of the three detected colors, the design method according to the invention making it possible to tolerate more geometric distortion and / or chromatic aberrations, in particular lateral ones (i.e. in the plane of the sensor 2) in order to reduce constraints in order to obtain better sharpness.

[0194] We will now describe, with reference to Figures 1 to 7, a first embodiment of imaging device 1 obtained by the design method according to the invention.

[0195] The first embodiment of imaging device 1 (also called camera 1) comprises:

[0196] - A 2 image sensor,

[0197] - An optical system 3 (also called lens 3) comprising an optical axis 4 also called Z axis (said system 3 being preferably centered on this axis 4) and arranged to receive light beams and direct them at least partly parallel to the optical axis (in order to use the lens preferentially in a configuration of good light capture because its projected surface is generally larger when the beams enter around an axis of optical symmetry of a lens, and this also minimizes certain possible optical aberrations) towards the sensor 2, more precisely to focus them on the sensor 2.

[0198] The light beams of the present description are light beams which can range from X-rays to infrared, and having a wavelength between 10 nm and 3000 nm, preferably between 400 nm and 700 nm.

[0199] The optical system 3 has (preferably considering a constant wavelength between 10 nm and 3000 nm, preferably between 400 nm and 700 nm) a magnification as well as a focal length:

[0200] - variable depending on the regions on sensor 2, and

[0201] - which depends on the orientation (or angle a) and / or the position of each optical beam incident on the optical system 3 relative to the optical axis 4.

[0202] In the present description, a magnification (which is a dimensionless quantity) of the system 3 on the sensor 2 for a given incident beam coming from a point of an object is defined as the ratio between the angle 3, relative to the axis 4, under which the image of this point formed by the optical system 3 is seen (i.e. at the output of the system 3, preferably for the image formed on the sensor 2) and the angle a, relative to the axis 4, under which this point of the object is seen "with the naked eye" i.e. at the input of the optical system 3. These angles are oriented angles. If the image is inverted, the magnification is negative (but its absolute value can also be considered for convenience, the latter being the one which is described as increasing or decreasing in this method).

[0203] The magnification of the optical system 3 comprises a minimum magnification value on the sensor 2 and a maximum magnification value on the sensor 2, the ratio of the maximum magnification of the optical system 3 to the minimum magnification of the optical system 3 being equal (typically for an angle a varying from 0° to 45°) to at least 1.2 in absolute value, preferably at least 1.5 in absolute value, preferably at least 2 in absolute value, preferably at least 4 in absolute value, preferably at least 5 in absolute value, preferably at least 6 in absolute value.

[0204] Different light beams corresponding to a decrease in the magnification of the optical system are imaged in different parts 51, 52, 53 (illustrated in Figure 7) of the sensor 2 moving away from a point 6 or a segment 7 or a surface on the sensor 2.

[0205] In the case of Figure 7, the system 3 has a symmetry around the axis 4, and different light beams corresponding to a decrease in the magnification of the optical system 3 are imaged in different parts 51, 52, 53 of the sensor 2 moving away from a point 6 on the sensor 2, so that there is, on the sensor 2, a central zone 51 (circular or substantially circular) of high magnification (which decreases progressively moving away from the point 6), a first peripheral zone 52 (circular or substantially circular) of lower magnification (which decreases progressively moving away from the point 6) and located around the central zone 51, and a second peripheral zone 53 of even lower magnification (which decreases progressively moving away from the point 6) and located around the first peripheral zone 52.

[0206] The magnification of the optical system 3 decreases when the angle α of the light beam incident on the optical system 3 increases relative to the optical axis 4.

[0207] The optical system 3 includes two lenses including:

[0208] - A first lens 31 (also noted Ll), which ensures a first spatial concentration of an incident beam, and at constant or substantially constant magnification and focal length depending on the angle a, and

[0209] - A second lens 32 (also noted L2), located between the first lens 31 and the sensor 2, and responsible at least in part (and even totally in the first embodiment) for the variability of the magnification and the focal length of the optical system 3.

[0210] The second lens 32 is for example at a distance from the first lens 31 equal to the focal length (along the U axis) of the first lens 31 at plus or minus 20% of the value of this focal length of the first lens 31.

[0211] The magnification of the second lens 32 decreases when the angle α of the light beam incident on the second lens 32 increases relative to the optical axis 4.

[0212] The magnification of the second lens 32 comprises a minimum value and a maximum value, the ratio of the maximum of the magnification of the second lens to the minimum of the magnification of the second lens being equal (typically for an angle a varying from 0° to 45°) to at least 2 in absolute value, preferably at least 6 in absolute value, preferably at least 8 in absolute value, preferably at least 10 in absolute value. The optical system 3 is centrosymmetric, that is to say symmetrical around the optical axis 4.

[0213] The magnification of the optical system 3 is maximum on a central or substantially central part 51 of the sensor 2.

[0214] The magnification of the optical system is minimum on a peripheral part 53 of the sensor 2.

[0215] In the present description, the variability of magnification of the system 3 and of the lens 32 (or its constant character for the lens 31) corresponds to a variability of focal length of the system 3 and of the lens 32 (or its constant character for the lens 31), and vice versa, and the terms magnification and focal length can therefore be exchanged in the present description of the device 1 and of its implemented method while remaining within the scope of the invention.

[0216] Thus, an example of an embodiment obtained from the invention, the local curvature of the 2 èmelens varies greatly in the peripheral region, to modulate the equivalent focal length of the optical beam arriving on the lateral zones of the sensor 2.

[0217] According to the state of the art, the focal length of an optical system (lens or a stack of lenses forming an objective) is defined as the distance CF between:

[0218] - the optical center C of the optical system and

[0219] - the average point B or also noted F where the cone of light leaving the system focuses.

[0220] With reference to figure 2a, for the invention, we will call focal length the distance between the projection on axis called ZZ' (which corresponds to axis Z or 4), of respectively points C and F at two points called Cz and Fz (also noted C' and F' in the rest of this description), ZZ' being an axis perpendicular to the image plane considered, this image plane corresponding in figure 2a to the plane of the sensor (whose plane contains F and Fz, for an image coming from infinity at the input of the optical system).

[0221] This definition allows us to directly compare the focal lengths f for the different possible angles formed by the incident beams for different points on the sensor, especially if it is flat. Thus, equality of magnification between two different parts of the image means that the distances f are also equal. According to the state of the art also, for two different angles of incidence a, the focal points have the same projection F on the optical axis, because we are trying to obtain an image appearing in the same plane perpendicular to the optical axis. This defines the focal length of the optical system, which is the one at which to place the image plane, for a set of optical beams coming from a point "at infinity", or very far from the lens, to obtain a sharp image of a plane of the scene at infinity.

[0222] According to the invention, for the optical system 3, this definition is retained but considering as illustrated in figures 2b and 3 that the focal length C'F' of the system 3 depends on the angle a of incidence of all the light beams incident on the optical system 3. Thus, this makes it possible to design an optical system 3 such that the focal length C'F'(a) becomes a non-constant function according to a.

[0223] As illustrated in figures 2b and 3, it is possible to have a ratio for example of at least 1.2 in absolute value or at least 1.5 in absolute value or at least 2 in absolute value or at least 4 in absolute value or at least 5 in absolute value or at least 6 in absolute value between the equivalent focal length C'Fa' at the center of the optical system 3 for a=0 and the equivalent focal length C'Fb' on the edges of the optical system 3 for example for a=45°.

[0224] For example, referring to Figure 2b, C'Fa'= 120 mm in the center and C'Fb'=20 mm on the edges.

[0225] When designing the system 3 or each lens 31, 32, the distance (Z) between the plane of the sensor 2 and the equivalent optical center C or Cl or C2 respectively of the system 3 or of the lens 31 or of the lens 32 can also be taken into account.

[0226] In fact, a lens has an equivalent point called the center through which all the rays would appear to pass if the light arrived as a single light ray, this center possibly being physically outside the lens.

[0227] Here, each angle a corresponds to a focal length, therefore to an equivalent pseudo lens. Each of these lenses has a particular position of its pseudo center. When this pseudo center is placed at the focal length of the sensor, the image is sharp at infinity (and the rays entering the focusing device considered 3, 31 or 32 are all parallel). We have the conjugation relation: 1 / f = 1 / dl + l / d2 where f is the focal length, dl is the distance from the pseudo center C or Cl or C2 to the sensor 2, and d2 the distance from the pseudo center C to the point of the scene.

[0228] Given the chosen definition of the focal length as a projection onto the ZZ' axis, it is convenient to define the conjugation relation with dl, d2, f, as also being the three distances projected onto the ZZ' axis. Of course, this amounts to factoring the projection factor and making it disappear with respect to the usual conjugation relation, which therefore continues to be true (to a certain approximation).

[0229] It is appropriate to modulate the distance between the optical center C and sensor 2, for a given angle a. This optical center of variable position as a function of a will be noted C(a). For example, it is a question of simultaneously obtaining all the zones of sensor 2 in focus for all the points of a particular plane of the scene, located at a distance Z0 along the Z axis. This scene plane - sensor plane relationship is not obligatory. We can also choose that the surface of sensor 2 corresponds to a surface shape other than a plane.

[0230] Furthermore, when system 3 is moved longitudinally (parallel to axis 4 also noted ZZ') above sensor 2 to change its focusing distance, it is likely that the new selected sharpness surface will no longer be a plane.

[0231] Thus, in use, the distance between the sensor 2 and the lens 3 can be modified (by the movement means 8, 81 described below) to move this zone of sharpness on the scene, which will then be deformed.

[0232] Figure 4 illustrates the equivalent optical center distance, for sharpness at 2m, obtained from the same lens 3 as in Figure 3 having a focal length of 120mm at the center and 20mm at the edge, set for a sharp image at infinity, whose lens is moved to tune it at a distance of 2m. If sharpness is desired across the entire image, at 2m, it would then be necessary to modify the projected distance on the Z axis by 7.45mm, between the optical centers for the central part at high magnification Cl and for the lateral part C2.

[0233] We notice in fact a difference of (127.65-20.20) - (120 -20) = 7.45 mm between the equivalent optical centers, between the 2 settings at infinity and at 2m (using the conjugation formula).

[0234] In practice, as the beam is more parallel (to the optical axis) in the center, we can concentrate on the sharpness at the edge, that in the center will be relatively little degraded. But the image to be observed does not necessarily need sharpness everywhere, we can also maintain maximum sharpness in the center even if it means degrading that at the edge when changing focal length.

[0235] Typically, a longer focal length means less focused beam. Therefore, it is appropriate to modulate the low angle areas a by giving the lens 32 little curvature near the 4 axis.

[0236] Thus, according to an objective that can be designed from the invention, the device comprises for the system 3 and / or for the lens 32 a focal distance (the highest) in its central zone (i.e. for its parts closest to the axis 4), a lower focal distance for each lateral zone (i.e. for its parts furthest from the axis 4), and a continuous variation of the focal distance between the central zone and each lateral zone.

[0237] This allows for a fairly magnified image in the central area, and a much less magnified image in the lateral areas.

[0238] Thus, the device 1 designed from the invention consists of a camera module 1 capable of having a high resolution in a central part 51, medium resolution around 52, and low resolution on the edges 53. Other arrangements are possible, but the latter is more natural because physiologically, we generally focus our attention in the center of the field of view where human vision is also more resolved.

[0239] Along with this higher resolution, the magnification power is modulated strongly depending on the angle: more magnification in the center 51 where the resolution is higher.

[0240] The magnification of a lens is linked to its so-called focal length. The greater the latter, the more the image produced downstream of the lens (here constituted by system 3, that is to say by lenses 31 and 32) will be dilated by geometric effect.

[0241] Thus, in an equivalent manner, the focal length of each angular sector of the objective 3 produced is made strongly dependent on the angle a: we ensure that we have less focal length when the angle a of the incident beam increases in relation to the optical axis 4 of the objective 3.

[0242] This gives an image capture system 1, also called camera module 1, using an optic 3 with high geometric distortion. This high geometric distortion is designed to increase the density of points preferentially in the central part of the field of the sensor 2. This allows the demand to obtain more resolution such as:

[0243] - an optical zoom, or

[0244] - switching to another camera module with a narrower angle, or in a lower resolution mode to provide a wider angle image.

[0245] So from the same device 1, device 1 makes it possible to create the equivalent of a mechanically more complex device with optical zoom or multiple camera modules.

[0246] With reference to Figure 1, in a concrete case of the first non-limiting embodiment of the device 1 designed according to the invention:

[0247] - the first lens 31 ensures a focal length substantially equal to 240 mm in all its directions (i.e. according to all angles a of the beams)

[0248] - for the 2 ème lens 32: o in the center: the 2 èmeLens 32 contains a moderate curvature at its center (i.e. for its parts closest to axis 4), which also gives it a focal length of 240mm at its center. The 2 lenses 31, 32 add their focusing power (1 / f), which leads to a total focal length of system 3 of 120mm at the center of system 3 (i.e. for its parts closest to axis 4). o At the edges: On the extreme angles a around lens 32 (i.e. for its parts furthest from axis 4), the 2 ème lens 32 is highly curved to locally achieve at its ends an equivalent focal length of 22mm. The sum of the two focal lengths of the two lenses 31, 32 then makes a total focal length of system 3 of 20mm for the highest values ​​of a.

[0249] We therefore have a magnification or focal length ratio, between its maximum value and its minimum value:

[0250] - From 240 / 22 = 10.9 for the 32 lens

[0251] - From 120 / 20 = 6 for system 3

[0252] The optical center Cl for the central part (i.e. closest to axis 4 or corresponding to the lowest angles α) of the system 3 can be located quite far from the 2 lenses 31, 32 on the left, to be approximately 120 mm from the sensor 2. The optical center C2 of the extreme parts (i.e. furthest from axis 4 or corresponding to the highest angles α) of the system 3 can be located approximately inside the end of the 2nd lens 32 to be approximately 20 mm from the sensor 2.

[0253] Typically, each lens 31, 32 is made of plastic, for example polycarbonate, and has a diameter, perpendicular to the axis 4, of approximately 10 mm for example.

[0254] The device 1 comprises means 8, 81 for longitudinal movement (typically comprising one or more micrometric motors along 1 axis, or along 2 or 3 distinct axes, preferably perpendicular to each other, preferably a single axis parallel to the axis 4) arranged to move the relative position of the sensor 2 (parallel to the optical axis 4) with respect to the optical system 3 so as to modify the distance between the optical system 3 and the sensor 2 and thus modify the magnification values ​​of the optical system 3 on the sensor 2.

[0255] The device 1 comprises means 8, 82 for lateral movement (typically comprising one or more micrometric motors along 1 axis, or along 2 or 3 distinct axes preferably perpendicular to each other, preferably 2 axes perpendicular to each other and to the axis 4) arranged to move the relative position of the sensor 2 (perpendicular to the optical axis 4) with respect to the optical system 3 so as to shift the sensor 2 with respect to the optical axis 4 and thus modify the part of the sensor 2 on which the beam having the maximum value of the magnification of the optical system is imaged.

[0256] The lateral displacement means 8, 82 are positionally controlled to the restitution of a stable image by a processing unit 9 despite the movements of the device 1.

[0257] Ideally, two moving groups are required to modulate both the focal length and the distance to the equivalent optical center if one wants to keep the image sharp when zooming, and / or work with a large depth of field. For example, the distance of lens 31 and that of lens 32 can be modulated separately. This can provide the degree of freedom to change both focus and magnification.

[0258] At least one so-called liquid lens can also be used to modify at least one curvature, which can modulate the focusing distance and / or the magnifications of the objective. The processing unit 9 comprises at least one computer, a central or calculation unit, an analog electronic circuit (preferably dedicated), a digital electronic circuit (preferably dedicated), and / or a microprocessor (preferably dedicated), and / or software means.

[0259] The image sensor 2 comprises an array of pixels, said pixels all having the same dimension.

[0260] The image sensor 2 comprises a Bayer matrix (i.e. a matrix of light-detecting patterns), the patterns of which all have the same dimension and are possibly interspersed with so-called dual pixel patterns (for adjusting the focusing distance).

[0261] This strong modulation of the equivalent focal length of system 3 and / or lens 32 (located between lens 31 of system 3 and sensor 2) is accompanied by several effects:

[0262] (a) The projection of a grid occupying the major part of the field of camera 1 (square mesh grid) of the scene is no longer a grid (square mesh): there is a strong geometric dispersion, called cousin, because the corners of the grid appear strongly inside the expected design of the grid, because of the compression of the distances which moves towards the center the projected points of sensor 2, for the points located angularly at high values ​​of a.

[0263] (b) The central part of the image is less exposed, in terms of brightness received per unit area, than the peripheral parts. Indeed, the high magnification in the center selects for each pixel a narrower cone of light, therefore less light. For a factor of 6 like the example in figure 1, the ratio in area, therefore in brightness, is 6 2 = 36 so quite strong between the high angle parts and in the center.

[0264] (c) Finally, the chromatic dispersion can also be accentuated, with a fixed number of lenses 31, 32 making up the system 3. By increasing the number of lenses 31, 32, etc. of the system 3, the chromatic distortion can be reduced.

[0265] (d) When the nominal focusing distance of the camera module 1 is changed, there is a modulation of the shape of the surface of the sharpness point. From flat, it can become curved. The device 1 further comprises correction means (forming part of the processing unit 9) arranged and / or programmed to correct an image captured by the image sensor, comprising in particular:

[0266] - (a) means for correcting or compensating for geometric distortion of the image; the correction means are arranged and / or programmed to process the images produced by the sensor 2 with the aim of displaying them without distortion, with high compensation for geometric distortion, according to known state-of-the-art methods. This amounts to moving parts of the image outwards, therefore translating and / or interpolating pixels in the high angle areas a. Symmetrically, pixels can also be decimated in the centre according to the current representation of the image, if this allows not needing the full resolution of the sensor 2. It is also possible to advantageously use sampling that is too high compared to the current one of the display, for a particular representation of the contours of the imaged objects. As a general rule, this geometric distortion can depend on the distance between the point of the object in the scene and the camera module 1.This effect is generally accentuated for objects close to camera 1. In this case, the compensation may benefit from receiving an estimate of this distance as input, in order to be more precise; and / or.

[0267] - (b) means of correcting or compensating for vignetting or inhomogeneity of brightness of the image. This phenomenon compensates and even overcompensates for the vignetting effect. This effect is as follows: when the light arrives at an angle a, the captured angular surface decreases because the lens 31 at the entrance is seen at an oblique angle which reduces its apparent section and therefore its light collection.The correction means use for this a Bayer Matrix (part of the sensor 2) with a higher exposure time in the center of the sensor 2 to compensate for the loss of brightness, and a lower exposure time around to avoid saturation; It is also possible to use a first lens 31 with a fairly concave shape, to increase vignetting: as such a shape is contradictory with the expected convergence effect of the beam, it is then necessary to produce more convex curvature on the second face of the lens 31, or on at least one other lens 32 located between the input lens 31 and the sensor 2. It is possible, if the pixel pitch is not already at the minimum possible for the technology, to use a more resolved Bayer matrix in the lateral parts to increase the saturation threshold.It is also possible to modulate the gain of an analog-to-digital converter at the output of sensor 2, for example by dividing the photocurrent received from sensor 2 before sampling it, for example by a capacitive divider at the input of a blocking sampler, to limit the charge to be taken into account in the Analog / Digital conversion. It is also possible to use a technology with 'active' sampling time individualized according to the region of sensor 2: this makes it possible to limit the time elapsed between the resetting (RAZ) of the capacitors accumulating the charge generated by the photodiodes of sensor 2, and the action of the blocking sampler, where less electrical gain is desired; and / or.

[0268] - (c) means for correcting or compensating for chromatic distortion of the image. The correction means are arranged and / or programmed to compensate for chromatic distortion by processing the pixel displacements of the 3 colors differently; and / or

[0269] - (d) means for correcting or compensating for image sharpness. The correction means are arranged and / or programmed to correct the effects of loss of sharpness, in particular to enhance sharpness, by taking into account the PSF transfer function (Point spread function), or an estimate of its effects.

[0270] The device 1 further comprises zoom means forming part of the processing unit 9) arranged and / or programmed to zoom in on an image, coming from the image sensor 2, by selecting a sub-part of the image sensor 2 so that an increase in the zoom on the image coming from the image sensor 2 corresponds to an increase in the average value of the magnification of the optical system 3 on the selected sub-part of the sensor 2. For example, by default the device 1 can record and / or display a zoomed-out image corresponding to the assembly of the zones 51, 52, and 53, and a user can zoom in progressively by removing at least a part of the zone 53, then all of the zone 53 and at least a part of the zone 52, then all of the zone 53 and all of the zone 52 and at least a part of the zone 51, so as to increase the average value of the magnification on the recorded and / or displayed image as and when increasing the zoom.

[0271] In the preceding description, the variable magnification or focal length must be understood as at least one variable magnification or focal length, because as will be seen later (with reference to figures 5 and 6) the system 3 or the lens 32 can comprise, in variants described in a more detailed or complex manner, several magnifications or focal lengths which vary (typically two magnifications or focal lengths, along two different axes for example U and V, which preferably vary continuously as a function of a).

[0272] The focal length of a lens is a distance, which would be the distance at which a parallel beam entering the said lens is supposed to focus on a single point at the exit of the lens (if it is converging), or at the entrance (virtual point) if the lens is diverging, relative to the "optical center" of the said lens.

[0273] This focal length therefore characterizes the bending power of the light entering the lens.

[0274] From the previous focal point, on the axis connecting the point from which the input light appears to come (point A) to the point where the output light appears to converge (point B), by subtracting the focal length on this axis, we arrive at a point which appears to be the one by which the light does not appear to be deflected. This point is called the optical center of the lens.

[0275] Applying these two notions to the concept of multi-magnification in an objective, we obtain:

[0276] - For each point of origin of the light entering the optical system (lens or objective stack of lenses), there is a cone of light, defined by a source point (virtual or real) (A) from which the light seems to come, a point possibly at infinity.

[0277] At the exit of the lens, this cone of light is supposed to focus in a fairly punctual zone at the exit (B), possibly located at infinity if the wavefront is very flat, or if the light rays are very parallel.

[0278] When one of the points is at infinity, point A or point B, we can continue to define the optical center by considering the point at a finite distance, and the optical direction defined on the side of the point at infinity (because the beam is then parallel). With reference to figures 5 and 6, consider for example the two centrosymmetric lenses 31, 32, stacked.

[0279] - The first lens 31 is assumed to be convex and relatively close to a “classical ideal” lens, in the sense that over a wide angular aperture, the image of a point at infinity is formed in a plane without significant geometric distortion.

[0280] - The second lens 32 has 2 concave faces, its second face on the sensor 2 side being very curved at the edges.

[0281] SO :

[0282] In an axis inclined relative to the central axis 4, we observe in this second face of this second lens 32 a radius of curvature very different according to the local radial axis, also called sagittal (corresponding to the proper axis U of the description of the embodiment of the method according to the invention), and the so-called tangential axis (corresponding to the proper axis V of the description of the embodiment of the method according to the invention). According to U, the curvature is significant. According to V, the curvature is very small, or even zero.

[0283] Thus, let us assume the incident beam is still spherically symmetrical as it exits lens 31. When it exits lens 32, it undergoes a much greater curvature along the U axis. This causes this component to converge at a shorter distance than the curvature along the other axis, the two focusing distances being different along the U axis and along the V axis.

[0284] The proper axes are defined according to the combination of the effects of the different lenses 31, 32 of the system 3.

[0285] Generally, one of the two U and V axes is the sagittal axis, the other is the tangential axis, in a centrosymmetric optical system.

[0286] The 4-axis (also called the Z-axis) is chosen as the rotational symmetry axis of a centrosymmetric 3-axis system. The X and Y axes define any plane perpendicular to this 4-axis.

[0287] We then have, with reference to figures 5 and 6:

[0288] A: point from which the beams of light appear to originate before reaching system 3.

[0289] Bu or Bu: the point (already mentioned during the description of the embodiment of the method according to the invention) where the light beams seem to converge along the proper axis U, at the exit of the system 3. Bv or Bv: the point (already mentioned during the description of the embodiment of the method according to the invention) where the light beams seem to converge along the proper axis V, at the exit of the system 3.

[0290] Cu or Cu: Optical center (already mentioned during the description of the method embodiment according to the invention) of system 3 for the U axis.

[0291] Cv or Cv: Optical center (already mentioned during the description of the method embodiment according to the invention) of system 3 for axis V.

[0292] Fu or Fu is the focal distance (already mentioned during the description of the method embodiment according to the invention) defined between Cu and Bu when A is at infinity (on the left) of figure 1, 2b, 5 or 6.

[0293] Fv or Fv is the focal length (already mentioned during the description of the method embodiment according to the invention) defined between Cv and Bv when A is at infinity (on the left) of figure 1, 2b, 5 or 6.

[0294] But the image obtained may not be sharp whatever the position of the sensor 2, which is not desirable. To remedy this fact, it is then appropriate to move, by designing the shapes of the lenses 31 and 32, the "optical centers" along the U and V axes, by placing the Cv at a z coordinate (along the Z or 4 axis) lower than the z coordinate (along the Z or 4 axis) of Cu, more precisely at the difference between the focal lengths Fu and Fv, which then means that, for a source point at infinity, the focalizations become merged again in a certain plane P (which preferably corresponds to the plane of the sensor 2 thanks to the means 81). It is then ensured that this plane P is also that for the other incident angles of the beam, including in particular the zero angle a at the center. Thus, it is possible to obtain a sharp image everywhere, despite a different magnification ratio depending on the distance R from the optical axis, and a different ratio between the local axes U and V.

[0295] Thus, still for a non-zero angle of incidence a, we ensure that the curvature of the lens 32 along the axis V is negligible, which is naturally the case with a lens that is very curved in the radial axis. We can then consider that the center Cv would be substantially that of the passage of light in the lens 31, for example.

[0296] And we make it so that the focal length Fv is substantially that defined by the lens 31 also, with the offset linked to the refraction in the lens 32 close, to be included in the design. This defines the position of the convergence point in the other axis V. On the other axis U, we introduce a significant difference in curvature between the two faces of the lens 3, so that each beam is very modified in curvature by the lens 32 in this direction when passing from the first face of the lens 32 (on the side of the lens 31) to the second face of the lens 32 (on the side of the sensor 2). Thus, the distance Fu is much less than Fv, and the center of curvature Cu can be moved to the level of the lens 32, for example, or at least towards the lens 32, compared to Cv which remains towards the lens 31. Thus, by adjusting the curvatures of the faces of the lens 32, it is ensured that the point of convergence is also in the plane P of convergence along the axis V.

[0297] Figures 5 and 6 are not to scale: point A is generally quite far away, its position here is not to scale, for example the distance between A and lens 31 is of the order of a meter, the distance between lens 31 and sensor 2 is of the order of 5 to 10mm.

[0298] In Figure 6, the beam 41 in the center and the beam 42 on the edge do not travel through the same area of ​​the lens 32, which makes it possible to benefit from quite different beam curvature powers between these two areas (here central and peripheral). Of course, there is a continuous modulation, with overlapping areas between the two.

[0299] If the design is to predominantly use a source point located at a finite distance A from the objective, it is appropriate to ensure equality of the convergence points for this distance and not for a source point located at infinity.

[0300] In the previous design reasoning, the local focal length, along each axis, must be replaced by that such as: l / (distance between A and Cu) + l / (distance between Cu and Bu) = 1 / Fu l / (distance between A and Cv) + l / (distance between Cv and Bv) = 1 / Fv

[0301] We then make sure that Bu is close to Bv along the Z axis, in order to arrive at the same result, but in this case, it is the distance between Cu and Bu and the distance between Cv and Bv that should be added to Z(Cu) (the position of Cu along the Z axis, i.e. axis 4) and Z(Cv) (the position of Cv along the Z axis, i.e. axis 4), to obtain a z(Bu) (the position of Bu along the Z axis, i.e. axis 4) and z(Bv) (the position of Bv along the Z axis, i.e. axis 4) that are close.

[0302] Since there is also a risk of obtaining a position difference according to AR = (R(Bu) - R(Bv) ), we can also usefully seek to compensate for this aspect, by keeping AR close to zero. For example, we can also introduce a geometric distortion defect from lens 31 to obtain this result.

[0303] Whatever its variant or embodiment, the imaging device 1 designed using the design method according to the invention can be integrated into numerous forms of system or device, and it is possible, for example, to have, in a non-limiting manner:

[0304] - A smartphone comprising an imaging device 1, and / or

[0305] - a vehicle comprising an imaging device 1, preferably integrated into a driving assistance system and / or for providing images for the automatic piloting of said vehicle; this vehicle may for example be an automobile, or an aircraft such as a drone or an airplane; and / or

[0306] - an alarm and / or video surveillance system, comprising an imaging device 1; and / or

[0307] - a medical imaging device, comprising an imaging device 1. This medical imaging device may comprise a diagnostic system or a surgical operation system.

[0308] Device 1 (and its implemented process) offers numerous technical advantages.

[0309] Device 1 allows for improved night vision: the central part of sensor 2 receives less light. It will therefore be possible to expose for longer periods in the center if the technology of sensor 2 allows it; the disadvantage being greater sensitivity to movements, either of device 1, or of the observed scene, or of both.

[0310] Compared to state-of-the-art devices combining several camera modules including a wide-angle module and a moderate-angle module, the device 1 allows a wider-angle camera module to be removed. Thus, two state-of-the-art camera modules can be replaced by the device 1 providing vision at these wider angles, but also fine resolution at smaller angles. The device 1 obtained by the design method according to the invention allows, for example, several state-of-the-art cameras to be merged into a single one, with a more resolved area in the center, the image of which can be selected exclusively if one is interested in a magnification in relation to this part of the field.In the prior art, it is even common to associate for example 3 cameras, one small angle, one medium angle, one wide angle (for example a smartphone having 3 separate camera lenses), only one of the cameras providing the image to be obtained according to the area to be viewed. The device 1 obtained by the design method according to the invention makes it possible to merge these 3 modules into a single one.

[0311] The device 1 obtained by the design method according to the invention also makes it possible to remove the optical zoom function from a state-of-the-art camera module specializing in reduced angles. For example, if a user wishes to have more details in the central part of the image:

[0312] - in the prior art, it requires an optical zoom function, i.e. a modification of the lens-sensor distance, over a relatively large distance. At the same time, it may be necessary to modify the optical center-sensor distance to keep it consistent with the observed part. There may possibly be 2 mobile optical groups, with a second distance modification.

[0313] - The device 1 obtained by the design method according to the invention makes it possible to eliminate this or these mobile optical groups. It replaces this optical zoom function with a more framed image restitution on an interior zone such as what the optical zoom would have done. As the pixel density is greater, the better resolved information is present in the image at the sensor 2.

[0314] That being said, a device 1 obtained by the design method according to the invention can furthermore, if necessary, integrate another camera module with a wider angle than that corresponding to the optical system 3. This other module can provide the missing information of the image in the peripheral zones, compared to a restitution of straight edges, which causes loss of information in the less dilated lateral zones, compared to those dilated. This other camera module with even wider angles than the system 3 can provide the missing data during these image distortion compensations.

[0315] The embodiment of the imaging method implemented by the device 1 typically comprises: - a reception, by the optical system 3, of light beams so as to direct them at least partly parallel to the optical axis 4 towards the sensor 2, including different light beams having between them different orientations and / or positions of incidence on the optical system relative to the optical axis, the optical system having a variable magnification on the sensor which depends on this orientation and / or position of each optical beam incident on the optical system relative to the optical axis, so that the different beams are imaged on the sensor with different magnifications of the optical system on the sensor, the magnification of the optical system comprises a minimum value on the sensor and a maximum value on the sensor, the ratio of the maximum of the magnification of the optical system to the minimum of the magnification of the optical system,preferably for the different beams imaged on the sensor, being equal to at least 1.2.,

[0316] This method embodiment may further comprise all or part of the steps below:

[0317] - a displacement, by the longitudinal displacement means 8, 81, of the position of the sensor relative to the optical system so as to modify the distance between the optical system and the sensor and thus modify the different values ​​of the magnification of the optical system on the sensor; and / or

[0318] - a displacement, by the lateral displacement means 8, 82, of the position of the sensor relative to the optical system so as to shift the sensor relative to the optical axis and thus modify the part of the sensor having the maximum value of the magnification of the optical system; and / or

[0319] - a correction, by the correction means 9, of an image captured by the image sensor, comprising in particular: o a correction or compensation of geometric distortion of the image, and / or o a correction or compensation of vignetting or inhomogeneity of brightness of the image, and / or o a correction or compensation of chromatic distortion of the image, and / or o a correction or compensation of sharpness of the image; and / or - a zoom, by the zoom means 9, on an image from the image sensor by selecting a sub-part of the image sensor so that an increase in the zoom on the image from the image sensor corresponds to an increase in the average value of the magnification of the optical system on the selected sub-part of the sensor.

[0320] The method embodiment can be implemented within a smartphone, within a vehicle, within an alarm and / or video surveillance system, and / or within a medical imaging device.

[0321] Of course, the invention is not limited to the examples which have just been described and numerous adjustments can be made to these examples without departing from the scope of the invention. For example, in one or more variants or alternative embodiments considered solely for their differences with respect to the device 1 previously described:

[0322] - the system 3 can be made up of o more than 2 lenses. In this case, the first lens 31 of the system, i.e. the furthest from the sensor 2, preferably has a constant focal length and magnification as previously described, or o a single lens, with a slightly curved central zone and increasingly curved lateral or peripheral zones, for example on the second face on the sensor side, the first face on the imaged object side then having a fairly uniform curvature over the entire lens, which ensures the concentration of the beam in a reduced space on the second face.However, such a variant is quite disadvantageous compared to the case comprising at least two lenses, because it comprises fewer degrees of freedom, typically the shapes of only two surfaces instead of those of four surfaces, typically also, which is related, the presence of air between the two lenses which modifies the possible deflection of the beams, which makes it more difficult to obtain all of the constraints mentioned during the embodiments; and / or compared to the embodiment of device 1 described with reference to figures 1 to 7, and in particular compared to the zones 51, 52, 53 illustrated in figure 7: o with reference to figure 8, different light beams corresponding to a decrease in the magnification of the optical system are imaged in different parts 51, 52, 53 of the sensor moving away from a point 6 on the sensor 2. The magnification of the optical system is minimum on a part of the sensor elongated in one direction.This can for example be achieved by using lenses 31, 32 and / or a system 3 having a section plane equivalent to that of Figure 1 but (instead of symmetry around axis 4) stretching with a slight curvature perpendicular to the plane of Figure 1; and / or o with reference to Figure 9, different light beams corresponding to a decrease in the magnification of the optical system are imaged at different parts 51, 52, 53 of the sensor moving away from a point 6 on the sensor 2. This can for example be achieved by using lenses 31, 32 and / or a system 3 similar to that shown in Figure 1 by shifting the center of the lenses 31 and 32 to the edge 6 of the sensor 2; and / or o with reference to figure 10, different light beams corresponding to a decrease in the magnification of the optical system are imaged in different parts 51, 52, 53 of the sensor moving away from a segment 7 on the sensor 2.The magnification of the optical system is minimum on a part of the sensor elongated in one direction. This can for example be obtained by using lenses 31, 32 and / or a system 3 derived in part from FIG. 1 (on one side or the other of the axis 4), recentered on the sensor 2, and stretching without any curvature perpendicular to the plane of FIG. 1, this stretching axis being made parallel to the axis 7 and superimposed on the axis 7 in the direction ZZ'; and / or o with reference to FIG. 11, different light beams corresponding to a decrease in the magnification of the optical system are imaged in different parts 51, 52, 53 of the sensor moving away from a segment 7 on the sensor 2. The magnification of the optical system is minimum on a part of the sensor elongated in one direction.This can for example be obtained by using lenses 31, 32 and / or a system 3 having a section plane equivalent to that of figure 1 but (instead of a symmetry around the axis 4) stretching without any curvature or with a very slight curvature perpendicular to the plane of figure 1 (as for the system 3 of figure 10); and / or o the point 6 can be replaced by the contour of a surface (for example a circle) and therefore be replaced in the previous description by this surface contour located on the sensor.

[0323] Of course, the various features, forms, variants and embodiments of the invention may be combined with each other in various combinations provided that they are not incompatible or mutually exclusive. In particular, all the variants and embodiments described above may be combined with each other.

Claims

CLAIMS 1. Method for designing an optical system, implemented by technical means, comprising: a) a modification of an optical transfer function of said optical system so as to minimize deviations between the optical transfer function of the optical system and a target optical transfer function, said target optical transfer function being characterized, for each point among several light source points A located upstream of the optical system and for each wavelength among at least one target wavelength, by: - a convergence of light through the optical system along a proper axis U at a point Bu and with a focal length along the axis U and / or a curvature power along the axis U and / or an optical center position Cu along this axis U, this convergence varying according to point A, this convergence being defined by: o at least one among Bu and Cu, and o at least one other parameter different from that of the preceding paragraph among Bu, Cu, the focal length along the axis U and the curvature power along the axis U. and / or b) a minimization between the desired value and the value obtained of at least two parameters among: - at least one of Bu and Cu, and - at least one other parameter different from that of the previous paragraph among Bu, Cu, the focal length along the U axis and the curvature power along the U axis.

2. Method according to claim 1, characterized in that: - said target optical transfer function is characterized, for each point among several points A light sources located in upstream of the optical system and for each wavelength among at least one target wavelength, by a convergence of the light through the optical system along a proper axis V different from the axis U, at a point Bv and with a focal length along the axis V and / or a curvature power along the axis V and / or an optical center position Cv along this axis V, this convergence varying according to the point A, and / or - the minimization between the desired value and the value obtained of at least two parameters may comprise a minimization between the desired value and the value obtained of at least four parameters from: o at least one from Bu and Cu, and o at least one other parameter different from that of the previous paragraph from Bu, Cu, the focal length along the U axis and the curvature power along the U axis o at least one from Bv and Cv, and o at least one other parameter different from that of the previous paragraph from Bv, Cv, the focal length along the V axis and the curvature power along the V axis.

3. Method according to claim 2, characterized in that the convergence of light through the optical system along a proper axis U is different from the convergence of light through the optical system along a proper axis V.

4. Method according to claim 2 or 3, characterized in that for each point among the several points A light sources and each wavelength among the at least one target wavelength, the target optical transfer function and / or the desired value of at least two parameters has points Bu and Bv merged or substantially merged.

5. Method according to any one of claims 2 to 4, characterized in that for each point among the several light source points A and each wavelength among the at least one target wavelength, the target optical transfer function and / or the value desired of at least two parameters has different focal lengths and / or curvature powers and / or optical center positions along the U axis and along the V axis.

6. Method according to any one of claims 2 to 5, characterized in that the axes U and V are perpendicular.

7. Method according to any one of claims 2 to 6, characterized in that the U axis is a sagittal axis of the optical system, and / or the V axis is a tangential axis of the optical system.

8. Method according to any one of the preceding claims, characterized in that the optical system is centrosymmetric.

9. Method according to any one of the preceding claims, characterized in that it comprises, before minimizing the deviations between the optical transfer function of the optical system and the target optical transfer function and / or minimizing between the desired value and the value obtained of at least two parameters, a definition of a material and / or an optical index of at least one lens making up the optical system.

10. Method according to any one of the preceding claims, characterized in that the optical system comprises at least two lenses, including: - a first converging lens located closer to the upstream than to the downstream of the optical system, and - a final converging or diverging lens, located closer to the downstream than the upstream of the optical system.

11. Method according to any one of the preceding claims, characterized in that the optical system (3) comprises an optical axis (4), said method further comprising taking into account an image sensor (2) located downstream of the optical system, so that the optical transfer function of the optical system and / or the function of target optical transfer and / or the desired value of at least two parameters and / or the obtained value of at least two parameters is arranged so that: - the optical system receives light beams from the source points A and directs them towards the sensor, - the optical system has a variable magnification on the sensor, magnification which depends on the orientation and / or position of each optical beam incident on the optical system, and - the magnification of the optical system includes a minimum value on the sensor and a maximum value on the sensor, the ratio of the maximum magnification of the optical system to the minimum magnification of the optical system being at least 1.

2.

12. Method according to claim 11, characterized in that the optical transfer function of the optical system and / or the target optical transfer function and / or the desired value of at least two parameters and / or the obtained value of at least two parameters is arranged so that different light beams corresponding to a decrease in the magnification of the optical system are imaged in different parts (51, 52, 53) of the sensor moving away from a point (6) or a segment (7) or a surface on the sensor.

13. Method according to any one of claims 11 to 12, characterized in that the optical transfer function of the optical system and / or the target optical transfer function is arranged so that the magnification of the optical system decreases when the angle of the light beam incident on the optical system increases relative to the optical axis.

14. Method according to any one of claims 11 to 13, characterized in that the optical transfer function of the optical system and / or the target optical transfer function is arranged so that the optical system comprises two lenses including: - A first lens (31), or a first face of the same single lens, preferably with constant or substantially constant magnification, and - A second lens (32), or a second face of the single lens, located respectively between the first lens or the first face of the single lens and the sensor, responsible at least in part for the variability of the magnification of the optical system.

15. Method according to claim 14, characterized in that the optical transfer function of the optical system and / or the target optical transfer function is arranged so that the second lens is at a distance from the first lens equal to the focal length of the first lens plus or minus 20% of the value of this focal length of the first lens.

16. Method according to claim 14 or 15, characterized in that the optical transfer function of the optical system and / or the target optical transfer function is arranged so that the magnification of the second lens decreases when the angle of the light beam incident on the second lens increases relative to the optical axis.

17. Method according to any one of claims 14 to 16, characterized in that the optical transfer function of the optical system and / or the target optical transfer function is arranged so that the magnification of the second lens comprises a minimum value and a maximum value, the ratio of the maximum of the magnification of the second lens to the minimum of the magnification of the second lens being equal to at least 2.

18. Method according to any one of claims 11 to 17, characterized in that the optical transfer function of the optical system and / or the target optical transfer function is arranged so that the optical system is symmetrical around the optical axis.

19. Method according to any one of claims 11 to 18, characterized in that the optical transfer function of the system optical and / or the target optical transfer function is arranged so that the magnification of the optical system is maximum on a central or substantially central part of the sensor.

20. Method according to any one of claims 11 to 19, characterized in that the optical transfer function of the optical system and / or the target optical transfer function is arranged so that the magnification of the optical system is minimal on a peripheral part of the sensor.

21. Method according to any one of claims 11 to 20, characterized in that the optical transfer function of the optical system and / or the target optical transfer function is arranged so that the magnification of the optical system is minimal on a part of the sensor elongated in one direction.

22. A computer program comprising instructions which, when executed in a computer, implement the steps of the method according to any one of claims 1 to 21.

23. A computer-readable storage medium comprising instructions which, when executed by a computer, cause the computer to perform the steps of the method according to any one of claims 1 to 21.

Images