Imaging system

Abstract

Hyperspectral Imaging System The invention relates to an imaging system for detecting an image of an object, the imaging system comprising: an imager configured to receive a light beam from the object; a slit configured to receive the light beam from the imager; the imaging system being characterized in that it comprises: a collimator array comprising collimators placed in a first plane and configured to collimate the light beam from the slit into a plurality of sub-beams; a metasurface array comprising metasurfaces placed in a second plane, the second plane being parallel to the first plane, one of the metasurfaces of the metasurface array being configured to receive one of the sub-beams from an associated collimator of the collimator array, each metasurface comprising a substrate and nano-elements; and one or more detectors;the nano-elements of one of the metasurfaces of the metasurface matrix being arranged so as to focus and disperse the associated sub-beam in a plane of the detector(s).

Description

Title of the invention: Imaging system Technical field

[0001] The invention relates to imaging systems and in particular, to hyperspectral imaging systems comprising metasurfaces.

[0002] BACKGROUND

[0003] The miniaturization of imaging systems is ubiquitous in emerging technologies, such as microlens systems on mobile phones, compact cameras, or even industrial cameras. Reducing the size and weight of the system are often the main motivations, and allow the integration of new functionalities while respecting the volume and mass constraints of imaging systems.

[0004] The miniaturization of components is accompanied by an increased requirement for the optical performance of the imaging system. For example, some cameras can image scenes of interest at 360°, at different spectral bands from visible to infrared for day and night vision. This multiplicity of functionalities (i.e. wide-field imaging and broadband imaging) increases the complexity of imaging systems. This complexity is generally addressed by separating the functionalities into multiple acquisition channels. Thus, two distinct optical systems are dedicated separately to the visible and infrared bands, at the cost of an increase in the size of the system.

[0005] Among miniaturized optical components, metasurfaces have recently demonstrated wide-field or multi-spectral imaging functions. For example, in SHALAGINOV, Mikhail Y., AN, Sensong, YANG, Fan, et al. A single-layer panoramic metalens with > 170 diffraction-limited field of view. Preprint at arXiv https: / / arxiv.org / abs / 1908.03626, 2019, a metasurface with a field of view (Field Of View) larger than 170° was demonstrated. Metasurfaces are optical components consisting of nano-elements with a size smaller than the operating wavelength and which can introduce effective refractive index and dispersion, which are dictated mainly by their geometric parameters and the arrangement of the nano-elements rather than the material composition. Additionally, US10634557B2 also exploits the dispersive nature of metasurfaces to separate spectral components in a focal plane.

[0006] However, in SHALAGINOV et al. cited above, the functionality is only effective at a single wavelength, reducing the possibility of imaging on one or more spectral bands. As for US10634557B2, a metasurface separating several wavelengths in a single-field focal plane is disclosed, limiting its imaging potential.

[0007] US20190178714 and US20200348176 disclose the use of metasurfaces optics for performing hyperspectral imaging. In particular, these inventions make it possible to perform a spectrometer function, i.e. a separation of the wavelengths of an incident beam on a detector. The spectrometer thus produced images an incidence (or a field point) on the detector. However, these documents do not disclose the separation of wavelengths at the different field points necessary to produce a spectro-imager.

[0008] In addition, US 10634557 also discloses a spectrometer comprising meta Optical surfaces. This document describes the implementation of a hyperspectral imaging function. The metasurfaces are arranged in cascade on either side of a substrate on which a thin layer of metal is deposited, allowing multiple reflections within the architecture before focusing on a detector. However, the different metasurfaces used in this architecture simultaneously address all the incident fields and associated wavelengths. In fact, each metasurface must correctly reflect the different fields at each wavelength in the right direction. This increase in constraints at the scale of a metasurface actually reduces the spectral or angular range addressable by all the structured components of the architecture.

[0009] SUMMARY

[0010] To address the problems encountered in the state of the art, the invention relates to an imaging system in which the integration of multiple functionalities, wide-field imaging with several spectral bands, is addressed by an optical architecture comprising a matrix of metasurfaces, each of the metasurfaces addressing a specific spectral range over a small range of incident field. The entire field of vision is then reconstructed by collecting the optical response of each metasurface taken individually. The system makes it possible to lighten and compact the imaging system, while meeting multifunctional needs.

[0011] In particular, the invention improves the situation by proposing an imaging system for detecting an image of an object, the imaging system comprising: an imager configured to receive a light beam from the object; a slit configured to receive the light beam from the imager; the imaging system being characterized in that it comprises: a collimator array comprising N x M collimators placed in a first plane and configured to collimate the light beam from the slit into a plurality of N x M sub-beams; a metasurface array comprising N x M metasurfaces placed in a second plane, the second plane being parallel to the first plane, one of the metasurfaces of the metasurface array being configured to receiving one of the sub-beams from an associated collimator of the collimator array, each metasurface comprising a substrate and nano-elements; and one or more detectors; the nano-elements of one of the metasurfaces of the meta-surface array being arranged to focus and disperse the associated sub-beam in a plane of the one or more detectors.

[0012] Thus in the present invention, by dividing the field of view into multiple sub-fields individually addressed by a collimator and an associated metasurface, the complexity of the optical response expected by a single component is diluted over a set of sub-components. The use of metasurfaces in a multifunctional optical architecture makes it possible to obtain a lighter system than conventional systems. In addition, since metasurfaces allow multiple degrees of freedom (by changing the architecture of the metasurfaces), it is possible to control the dispersion arbitrarily, and thus to generate a different optical function at each wavelength over a spectral range of interest. This control of the optical properties at each wavelength then offers a wider choice of architectures that can meet many applications.Finally, the flux perceived at the detector level can be kept constant despite the miniaturization of the components, and this by reducing the focal length of the metasurfaces by the same factor.

[0013] The principle of metasurface architecture is described for example in CHEN, Wei Ting, ZHU, Alexander Y., and CAPASSO, Federico. Fiat optics with dispersion-engineered metasurfaces. Nature Reviews Materials, 2020, vol. 5, no. 8, pp. 604-620.

[0014] In one embodiment, the system comprises a matrix of N x M detectors.

[0015] Advantageously, detectors having different characteristics can be adapted to the associated metasurfaces. In addition, it is possible to have detectors operating at different wavelengths (for example, in the visible and in the infrared). In addition, detectors having a large detection area are more expensive. Thus, by having an array of detectors having a smaller detection area, it is possible to reduce the costs of the system.

[0016] In one embodiment, the system comprises a dispersive element.

[0017] In one embodiment, two of the metasurfaces of the metasurface array are different.

[0018] In one embodiment, the detectors are placed in several planes.

[0019] In one embodiment, at least two of the planes in which the detectors are not parallel.

[0020] Advantageously, it is possible to adapt the position of the detectors according to the sub-beam to be detected. In addition, by being placed on several planes, it is possible to optimize the space used by the detectors.

[0021] In one embodiment, the substrate of at least one of the metasurfaces is curve.

[0022] In one embodiment, a predefined distance separates each nano-element from the metasurfaces, the predefined distance being a variable distance.

[0023] In one embodiment, at least two of the nano-elements of the metasurfaces have a different thickness and / or a different height.

[0024] In one embodiment, the nano-elements are cylindrical in shape.

[0025] Advantageously, by varying the parameters of the architecture of the metasurfaces, such as the shape of the substrate and the nano-elements, and the distance between the nano-elements, it is possible to meet numerous applications and the multifunctionality required in emerging imaging systems.

[0026] In one embodiment, the metasurfaces are fabricated by 3D printing.

[0027] Advantageously, 3D printing makes it possible to facilitate manufacturing and to produce metasurface designs which are expensive and / or difficult to produce with other conventional manufacturing techniques.

[0028] In one embodiment, the imager is a telecentric imager.

[0029] Advantageously, telecentricity allows the light beams to have a normal incidence in the plane of the slit. At the collimator exit, the sub-beams thus arrive with the same incidence on the matrix of metasurfaces, which reduces the number of different designs to be processed. Brief description of the drawings

[0030] The invention will be better understood and other advantages will appear on reading the description which follows, given without limitation and thanks to the figures among which:

[0031] [Fig.l] [Fig.l] illustrates an example of a conventional imaging system;

[0032] [Fig.2] [Fig.2] illustrates an example of an imaging system according to the invention;

[0033] [Fig.3] [Fig.3] illustrates an example of an imaging system according to the invention;

[0034] [Fig.4] [Fig.4] illustrates an example of sub-beam impact points with the detector;

[0035] [Fig.5a] [Fig.5a] illustrates an example of metasurfaces;

[0036] [Fig.5b] [Fig.5b] illustrates an example of metasurfaces;

[0037] [Fig.5c] [Fig.5c] illustrates an example of metasurfaces;

[0038] [Fig.5d] [Fig.5d] illustrates an example of metasurfaces; and

[0039] [Fig.6] [Fig.6] represents an example of a telemetric system according to an example. DETAILED DESCRIPTION

[0040] [Fig.l] illustrates an example of a conventional hyperspectral imager 1000 of the state of the art. An object 102 is imaged by a first imager 104 on an entrance slit 106, the dimensions of which obscure a portion of the fields coming from the object 102. beams from the slit, with a field angle of 0min to 0max, are collimated by a collimator 1002. Spectral components of a region of interest are then separated by a dispersive element 1004, such as a prism or a diffraction grating, then imaged on a detector 112 by a second imager 1006 (for example, a system of multiple lenses). Each field is distributed along columns of a detector 1006 (x-axis), at each wavelength along the lines (y-axis) of the detector 1006. A scan of the object 102 is necessary to image the entirety of the object 102. Indeed, the system is in a “push-broom” configuration, i.e. a horizontal section of the object 102 is imaged at a time, as illustrated in [Fig.l], and a scan is carried out, for example in the s direction, a vertical direction, in order to image several sections of the object 102.

[0041] [Fig. 2] illustrates an exemplary imaging system 100 for detecting an image of an object 102 according to the present invention. Preferably, the imaging system is a hyperspectral imaging system. In the present invention, the combination of the collimator 1002, the dispersive element 1004 and the imager 1006 of the conventional hyperspectral imaging system 1000 described in [Fig. 1] has been replaced by a collimator array 108 and a meta-surface array 110 allowing miniaturization of the hyperspectral imaging system.

[0042] The object 102 may be, for example, any object studied in applications of agriculture, astronomy, surveillance, etc. The object 102 is in a scene of interest.

[0043] The hyperspectral imaging system 100 comprises an imager 104 configured to receive a light beam 114 coming from the object 102. The imager 104 is for example a telescope or a camera.

[0044] The hyperspectral imaging system 100 also comprises a slit 106 configured to receive the light beam 114 coming from the imager 104. The slit 106 is placed in the image plane of the imager 104 (i.e. in the focal plane for an object in a scene at infinity). The slit 106 makes it possible to separate the different fields having different angles of incidence. The slit also makes it possible to suppress stray beams coming from the object 102.

[0045] The hyperspectral imaging system 100 also comprises a collimator array 108 comprising N x M collimators 108 placed in a first plane a and configured to collimate the light beam 114 coming from the slit 106 into a plurality of N x M sub-beams. The plane a is a plane one of whose sides is parallel to a y axis representing the wavelengths and one of whose sides is parallel to an x ​​axis representing the field. In one embodiment, the x and y axes are perpendicular. The plane a is also orthogonal to the optical axis z, the latter being perpendicular to the x and y axes. The different sub-fields separated by the slit 106 are received by different collimators 108 of the collimator matrix 108, which thus spatially separates portions of the incident field. The collimators may be, for example, microlenses.

[0046] The hyperspectral imaging system 100 also includes a metasurface array 110 comprising NXM metasurfaces 110 placed in a second b-plane. The second b-plane is parallel to the first a-plane. One of the metasurfaces of the metasurface array is configured to receive one of the sub-beams from an associated collimator 108 of the collimator array 108. In one example, the second b-plane of the metasurfaces is a mid-plane in which the metasurfaces lie but the metasurfaces may be placed at an angle relative to the mid-plane b-plane.

[0047] As illustrated in [Fig.5a], each metasurface 110 comprises a substrate 304 and nano-elements 302.

[0048] The hyperspectral imaging system 100 also includes one or more detectors 112.

[0049] The metasurfaces 110 of the metasurface matrix 110 are arranged so as to focus and disperse the associated sub-beam in a plane c of the detector(s). In particular, the metasurfaces 110 make it possible to control the focusing of the different sub-fields at the different wavelengths on the detector 112.

[0050] Thus each portion of field is directed towards the matrix of metasurfaces 110, whose dispersive properties make it possible to simultaneously focus and disperse a wavelength continuum over a given spectral range. The scene imaged at the different wavelengths is then reconstituted by summing the optical response of each metasurface 110 in the plane of the detector 112.

[0051] The field of view is spatially decomposed into N ranges of sub-fields, each range being addressed by a subsystem independent of the neighboring subsystems, and the N ranges of sub-fields being in the x direction. This distribution of the optical functionality makes it possible to reduce the operating constraints of each component, and thus to widen the field of view while retaining a spectral function necessary for the separation of the wavelengths. The detector(s) 112 used depend on the spectral band being observed. For example, the detector(s) 112 may be one or more infrared or visible detectors. In one example, a single detector 112 is used, as illustrated in [Fig. 2]. The detector 112 is in a plane which comprises columns along the x axis and lines along the y axis (here also being an X axis representing the wavelengths).In this example, the detector 112 receives the sub-beams of the metasurface array 110 focused and scattered by the metasurfaces 110. Each metasurface 110 of the metasurface array 110 separately addresses portions of fields which are . then detected by corresponding portions of the detector.

[0052] The detection by the detector 112 is a geometric image of the slit 106. In particular, all the wavelengths are present at each point of the slit 106 and are then imaged on the detector 112 along the x axis. Each point of the slit 106 is therefore imaged on a point of the detector 112 along the x axis.

[0053] In one embodiment, the system 100 comprises a dispersive element 1004. The dispersive element 1004 (such as a prime or grating) is placed between the microlens array 108 and the metasurface array 110, along the z-axis. Thus, in this example, dispersion along the y-axis is also achieved. Detection by the detector 112 is therefore the width of the slit 106 (along the x-axis), with wavelength dispersion in the x-direction. For example, Figure 4 shows an example of the points of impact between the sub-beams and the detector 112. In particular, the impact between the sub-beams and the plane of the detector 112 forms a continuum of incident fields that is imaged along the columns of the sensor (i.e., along the x-axis), and the spectral components associated with each field are distributed along the lines of the detector (i.e., along the y-axis).In an ideal case, the separation of the sub-beam features (having a wavelength X and a field angle 0) is done in two orthogonal directions at the detector 112. Figure 4 represents a discrete number of field points and associated wavelengths for visibility, but there are an infinite number of sub-fields and wavelengths between each point. As shown in Figure 4, the rectangles 402 represent portions of field and wavelength addressed separately by a metasurface of the metasurface array 110. In an example illustrated in Figure 4, the N x M metasurfaces individually address the same range of sub-fields (x axis), but a reduced number of wavelengths as these tend towards the lower limit of the probed spectral band. In [Fig.4] the lower limit of the probed spectral band corresponds to the lowest value of y and the upper limit corresponds to the highest value of y.In this configuration, that is to say with a dispersive element 1004 placed between the slit and the metasurface matrix, it is possible to simplify the manufacture of the metasurfaces. Indeed, in this configuration, the metasurfaces allow focusing but part of the dispersion is ensured by the dispersive element, which simplifies the design of the metasurfaces and the parameters to be considered for their manufacture.

[0054] As shown in [Fig. 3], in another example, the hyperspectral imaging system 100 comprises several detectors which are placed in a plane c parallel to the planes a and b. Thus, for large area detections, it may be advantageous to use several detectors, limiting the costs of the system and allowing the use of detectors having different characteristics.

[0055] In another example, the system 100 may comprise a matrix of NXM detectors 112. Thus, each detector 112 of the matrix of N x M detectors 112 may detect the sub-beams of a metasurface 110 of the associated matrix of N x M meta-surfaces. The detectors 112 of the matrix of detectors 112 may be placed in the same plane. Alternatively, the detectors 112 may be placed in several planes c, d, e. The planes c, d, e in which the detectors 112 are placed may be parallel or at least two of the planes in which the detectors 112 are placed may not be parallel. Thus, by placing the detectors 112 on several planes, it is possible to minimize the size of the hyperspectral imaging system 100 and to optimize the detection of the sub-beams of the metasurface associated with each detector.

[0056] It is noted that the elements of the hyperspectral imaging system described above, i.e., the imager 104, the slit 106, the collimator array 108, the metasurface array 110 and the detectors are placed along the optical axis z.

[0057] As illustrated in Figures 5a to 5c, each metasurface 110 comprises a substrate 304 and nano-elements 302. In the examples of Figures 5a to 5c, the substrate is planar. In another example, the substrate of at least one of the metasurfaces 110 is curved. A curved substrate makes it possible to increase the number of variables describing the metasurface and therefore to take into account additional sizing constraints (performance, volume, etc.).

[0058] In one example, all the metasurfaces of the metasurface matrix are identical. In another example, at least two of the metasurfaces of the metasurface matrix are different. Indeed, at least two of the metasurfaces may have different designs, such as for example nano-elements 302 of different shape and / or size and / or materials, and / or a different number of nano-elements 302, and / or a different arrangement of the nano-elements 302, and / or a substrate of different size and / or material. The matrix comprising metasurfaces having a different design makes it possible to optimize the entire system 100 (for example by allowing the correction of the distortion along the x axis, optimizing the spectral dispersion along the y axis, etc.).

[0059] The desired optical function, i.e. the dispersion and focusing of the sub-beams in the plane of the detector(s) 112 can be achieved by modifying the geometric characteristics and the relative proximity of nano-elements 302. For an application in the visible spectrum, it is preferable to choose nano-elements 302 made of dielectric materials whose refractive index is between 1.45 and 3. The height h of the nano-elements 302 is chosen to be sufficiently large to modulate the phase of the incident light over an amplitude of at least 2ir. For example, for a material with an index n = 2.3 in the visible spectrum, the height h of the nano-elements 302 is of the order of 900 nm. The distance d between the nano-elements can be chosen to be less than the wavelength to locally avoid the diffraction of parasitic orders, typically between 200 and 400 nm. The multiplicity of these parameters (height h, thickness e, distance d, material index n...) increases the degrees of freedom available to design a multifunctional hyperspectral imaging system. If the spectral band probed is different, the geometry of the nano-elements 302 and the materials can be modified accordingly depending on the ratios of thickness e of the nano-elements 302 to wavelength and height h of the nano-elements 302 to wavelength, these ratios generally being between 0.5 and 1.

[0060] As shown in [Fig.5c], a predefined distance (d) separates each nanoelement 302 from the metasurfaces. The predefined distance d may be equal between each nanoelement 302. Alternatively, the distance d may be variable.

[0061] In the example of [Fig.5c], the nano-elements 302 have the same height h and the same thickness e. However, in another example, at least two of the nano-elements 302 of the metasurfaces 112 have a different thickness e, and / or a different height h.

[0062] As indicated in [Fig.5b], according to one embodiment the nano-elements 302 are cylindrical in shape. Indeed, the nano-elements can be solid cylinders, hollow cylinders, cylinders comprising another cylinder, or even cylinders with any base (for example a cross-shaped base in [Fig.5b]). In another example, the nano-elements 302 have a parallelepiped shape.

[0063] The architecture of the metasurfaces (i.e., the distance between the nano-elements, the thickness of the nano-elements, the height h of the nano-elements, the shape of the nano-elements, etc.) depends on the application of the spectral imager. For example, the architecture will depend on whether the object is observed in the visible spectrum or in the infrared and on the angle of incidence.

[0064] As illustrated in [Fig.5d], in one example, two nano-elements 302 may be stacked. In this example, two nano-elements 302 having different thicknesses eb e2 are stacked on top of each other on the substrate 304. The nano-elements 302 may be square-based or round-based.

[0065] As an example for detection between 600 and 700 nm (100 nm spectral band), a field between -8° and +8°, ​​the center-to-center distance between the two nano-elements 302 may be for example less than 600 nm. Each nano-element 302 may be characterized by a height h between 700 nm and 1 pm, and by a thickness e varying between 70 nm and 330 nm. The materials of the nano-elements 302 must be characterized by a refractive index high enough to transform the incident wave into a convergent spherical wavefront, typically greater than 2. In addition, the materials must be transparent, or weakly absorbent. over the spectral range of interest so as to maximize the flux at the detector 112.

[0066] As indicated above, the hyperspectral imaging system 100 comprises an array of nxm collimators 108 and an array of 11 xm metasurfaces 110. The number of collimators 108 and metasurfaces 110 of the nxm arrays depends on a set of parameters related to the optical architecture of the hyperspectral imaging system 100. Indeed, these parameters include an aperture of the imager 104, the focal length of the imager 104, the distance between the slit 106 and the collimators 108, the diameter of the collimators 108 and the surface area of ​​the metasurfaces 108.

[0067] In one example, the imager 104 is a telecentric imager. This makes it possible to compact the optical architecture of the hyperspectral imaging system 100. Indeed, it is preferable to have a telecentric imager upstream of the slit 106, as illustrated in [Fig. 6]. As indicated above, several sub-fields are received at the imager 104 and the slit 106. The telecentricity allows the light beams 114 of each sub-field to have a normal average incidence in the plane of the slit 106. At the output of the collimators 108, the sub-fields thus arrive with the same incidence on the matrix of metasurfaces 110, which reduces the number of different designs to be processed.

[0068] In addition, telecentricity makes it possible to juxtapose each metasurface 108 with the same orientation, preferably perpendicular to the optical axis z. If the hyperspectral imaging system 100 is not telecentric, it may be necessary to modify the orientation of each metasurface 110 relative to the optical axis z by a few degrees. Indeed, the plane b of the metasurfaces is substantially parallel to the plane a of the collimators but the metasurfaces can be oriented with an angle of a few degrees relative to the plane b (for example, 1 or 2).

[0069] It is noted that in the above examples, a portion of the object 102 is imaged. A scan of the object 102 is necessary to image the entirety of the object 102, as in the conventional system described in [Fig.l]. Thus, the system 100 is also in a "push-broom" configuration, i.e., one horizontal section of the object 102 is imaged at a time, as illustrated in [Fig.l], and a scan is performed, for example in the s direction (parallel to the y axis), a vertical direction, in order to image several sections of the object 102.

[0070] The metasurfaces described in this invention may be, for example, metasurfaces such as described in CHEN, Wei Ting, ZHU, Alexander Y., and CAPASSO, Federico. Fiat optics with dispersion-engineered metasurfaces. Nature Reviews Materials, 2020, vol. 5, no. 8, p. 604-620. It is possible to manufacture the metasurfaces 108 by etching. For example, the substrate 302 may be made of sapphire or SiO2 on which a layer of TiO2, GaN, Si3N4 or even monocrystalline silicon is deposited in order to etch the nano-elements 304 into the TiO2 layer. In one example, the meta-surfaces 112 can be manufactured by 3D printing. 3D printing allows for ease of manufacturing and the use of various materials that are not possible with conventional manufacturing methods.

[0071] It is noted that, in conventional imaging systems, the different imaged field points do not focus on the same line of the detector. This geometric distortion of the spread function of a point is called smile and is often observed on hyperspectral imagers. Indeed, in conventional imaging systems, the collimator 1002 and the imager 1006 generate distortion that is all the stronger the stronger the field. The hyperspectral imaging system 100 of the present invention makes it possible to overcome this effect by dividing the field range perceived by each metasurface 108. In particular, each metasurface receives a much smaller subfield (field divided by N), which makes it possible to reduce the distortion.

[0072] In the invention presented above, by dividing the field of view into multiple sub-fields individually addressed by a collimator 108 and a metasurface 110, the complexity of the optical response expected (i.e., wide-field and multi-wavelength response) by a single component is diluted over a set of sub-components. The use of metasurfaces 110 in a multifunctional optical architecture has several advantages. In particular, the metasurfaces are ultra-thin components, the thickness of which rarely exceeds one millimeter taking into account the thickness of the substrate 304. Thus, the hyperspectral imaging system 100 comprising metasurfaces 108 is lighter than the conventional system, often made of massive refractive materials whose thickness is greater.Furthermore, by using conventional dispersive components such as a diffraction grating, the wavelength separation is fixed by the grating period. By using the multiple degrees of freedom of the 108 metasurfaces, it is possible to control the dispersion arbitrarily, and thus generate a different optical function at each wavelength over the spectral range of interest. This control of the optical properties at each wavelength then offers a wider choice of architectures that can meet the operational need. Finally, the flux perceived at the detector can be kept constant despite the miniaturization of the components, and this by decreasing the focal length of the 108 metasurfaces by the same factor.

[0073] Although the invention has been illustrated and described in detail using a preferred embodiment, the invention is not limited to the disclosed examples. Other variations may be deduced by those skilled in the art without departing from the scope of protection of the claimed invention. For example, those skilled in the art will understand that the metasurfaces described above are only examples and that the materials and architecture described depend on the application of the hyperspectral imaging system. Likewise, the metasurface and collimator matrices are shown only as examples.

Claims

Claims

1. An imaging system (100) for detecting an image of an object (102), the imaging system comprising: - an imager (104) configured to receive a light beam (114) from the object; - a slit (106) configured to receive the light beam from the imager; - a collimator array (108) comprising N x M collimators placed in a first plane (a) and configured to collimate the light beam from the slit into a plurality of N x M sub-beams; - a metasurface array (110) comprising N x M metasurfaces placed in a second plane (b), the second plane being parallel to the first plane, one of the metasurfaces of the metasurface array being configured to receive one of the sub-beams from an associated collimator of the collimator array, each metasurface comprising a substrate (304) and nano-elements (302); - one or more detectors (112);and - a dispersive element (1004), the nano-elements of a metasurface of the metasurface matrix being arranged so as to focus and spectrally disperse the associated sub-beam in a plane of the detector(s).;

2. The imaging system of claim 1, wherein at least two of the metasurfaces of the metasurface array are different.

3. An imaging system according to one of the preceding claims, wherein the system comprises an array of N x M detectors.

4. An imaging system according to claim 3, wherein the detectors are placed in several planes (c, d, e).

5. An imaging system according to claim 4, wherein at least two of the planes in which the detectors are placed are not parallel.

6. An imaging system according to one of the preceding claims, wherein the substrate of at least one of the metasurfaces is curved.

7. Imaging system according to one of the preceding claims, wherein the nano-elements are cylindrical in shape.

8. An imaging system according to one of the preceding claims, wherein the metasurfaces are manufactured by 3D printing.

9. Imaging system according to one of the preceding claims, in which the imager is a telecentric imager.

Images