Optical system, configuration method, and identification method

Abstract

The present application discloses an optical system, a configuration method, and an identification method. The optical system comprises: an optical element; and a processing module, coupled to the optical element, and used for processing data information from the optical element by means of a first program. The configuration method comprises: acquiring sample data; inputting first data into the optical element, executing the first program for processing to obtain intermediate data, and determining an evaluation value on the basis of the intermediate data and second data; and on the basis of the evaluation value, updating the configuration of the optical element, or updating the configuration of the first program, or updating both the configuration of the optical element and the configuration of the first program, until the evaluation value satisfies a preset numerical condition, so as to obtain configuration information of the optical system. By means of the optical system configuration method provided in the present application, a relationship between an actuality and a target can be established, thereby implementing overall configuration of the optical system, improving the configuration efficiency, and preventing error accumulation.

Description

Optical system, configuration method and identification method

[0001] This application claims priority to Chinese patent application No. 202411555228.7, filed on November 4, 2024, entitled "Optical System, Configuration Method and Identification Method", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of optical imaging technology, and in particular to an optical system, configuration method and recognition method. Background Technology

[0003] Optical systems are widely used in medical, industrial, and military fields to form imaging systems, sensors, or detection systems, or to realize laser technology, display technology, or fiber optic communication.

[0004] When optical systems are used to perform tasks such as computation and information processing, the optical and electronic components of the system often exhibit poor coordination, hindering their ability to work together effectively to accomplish the intended tasks. Furthermore, due to the inherent errors in both the optical and electronic components, errors accumulate in the computational and information processing results when they work together, leading to poor optimization and training performance and low efficiency for the optical system.

[0005] Especially in applications such as recognizing and understanding the posture and movement of the human body and objects, or human-computer interaction, motion analysis, and security monitoring, optical systems need to use the cooperation of their optical and electrical components to sense the posture of the human body. As a result, the aforementioned problems become more prominent, reflecting the higher performance requirements that this field places on optical systems.

[0006] Application content

[0007] One of the purposes of this application is to provide a configuration method for an optical system to solve the technical problems of poor information processing effect, error accumulation and low training efficiency in existing optical systems.

[0008] One of the purposes of this application is to provide an identification method.

[0009] One of the purposes of this application is to provide an optical system.

[0010] To achieve one of the aforementioned objectives, one embodiment of this application provides a configuration method for an optical system. The optical system includes: an optical element; and a processing module coupled to the optical element, configured to process data information from the optical element according to a first program. The configuration method includes: acquiring sample data, including first data corresponding to input information of the optical element and second data corresponding to information output when the first program is executed; the second data being target data corresponding to the first data; inputting the first data into the optical element, executing the first program to obtain intermediate data, and determining an evaluation value based on the intermediate data and the second data; updating the configuration of the optical element, or updating the configuration of the first program, or updating both the configuration of the optical element and the configuration of the first program, based on the evaluation value, until the evaluation value meets a preset numerical condition, thereby obtaining configuration information for the optical system.

[0011] To achieve one of the above-mentioned objectives, one embodiment of this application provides an identification method applied to an optical system, the optical system comprising: an optical element; and a processing module coupled to the optical element for processing data information from the optical element according to a first program; the identification method comprising: configuring the optical system according to configuration information obtained by the configuration method described in any of the above technical solutions; controlling the optical element to obtain a detection image of a corresponding target object; controlling the processing module to execute the first program to process the detection image; and identifying the target object.

[0012] To achieve one of the above-mentioned objectives, one embodiment of this application provides an optical system configured according to configuration information obtained by the configuration method described in any of the above-mentioned schemes, or the optical system is used to implement the identification method described in any of the above-mentioned technical solutions.

[0013] Compared with the prior art, the configuration method of the optical system provided in this application determines the evaluation value for evaluating the performance of the optical system by comprehensively considering the second data, which is the processing target corresponding to the first data executing the first program, and the intermediate data, which is the current actual processing result corresponding to the first data executing the first program, thereby adjusting the configuration of the optical system. Therefore, it is possible to establish the relationship between the actual and the target based on the first data, so that the optical system can quickly update the configuration to achieve the desired target. Furthermore, since the first data is the input of the optical element in the optical system and the intermediate data is the output of the processing module in the optical system, the optical element and the processing module as a whole can be used as the configuration and update object, preventing the accumulation of errors in both. Attached Figure Description

[0014] Figure 1 is a schematic diagram of the structure of an optical system in one embodiment of this application.

[0015] Figure 2 is a schematic diagram of the optical system in the first embodiment of this application.

[0016] Figure 3 is a schematic diagram of the structure of the first specific embodiment of the optical element in this application.

[0017] Figure 4 is a structural schematic diagram of a second specific embodiment of the optical element in this application.

[0018] Figure 5 is a schematic diagram of the optical system in the second embodiment of this application.

[0019] Figure 6 is a schematic diagram of the optical system in the third embodiment of this application.

[0020] Figure 7 is a schematic diagram of the steps of an optical system configuration method in one embodiment of this application.

[0021] Figure 8 is a schematic diagram of the steps of a first specific embodiment of the configuration method of the optical system in this application.

[0022] Figure 9 is a schematic diagram of the steps of a second specific embodiment of the configuration method of the optical system in this application.

[0023] Figure 10 is a schematic diagram of the steps of the identification method in one embodiment of this application.

[0024] Figure 11(a) shows a first target object in one embodiment of this application.

[0025] Figure 11(b) is a detection image corresponding to the first target object in one embodiment of this application.

[0026] Figure 12 is a detection image corresponding to the second target object in one embodiment of this application.

[0027] Figure 13 is a detection image corresponding to a third target object in one embodiment of this application.

[0028] Figure 14 is a detection image corresponding to the fourth target object in one embodiment of this application.

[0029] Figure 15 is a detection image corresponding to the fifth target object in one embodiment of this application.

[0030] Figure 16 is a detection image corresponding to the sixth target object in one embodiment of this application.

[0031] Figure 17 is a detection image corresponding to the seventh target object in one embodiment of this application.

[0032] Figure 18 is a detection image corresponding to the eighth target object in one embodiment of this application.

[0033] Figure 19 is a detection image corresponding to the ninth target object in one embodiment of this application. Detailed Implementation

[0034] The present application will now be described in detail with reference to the specific embodiments shown in the accompanying drawings. However, these embodiments do not limit the present application, and any structural, methodological, or functional modifications made by those skilled in the art based on these embodiments are included within the scope of protection of this application.

[0035] It should be noted that the term "comprising" or any other variation thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0036] Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying a relative order or importance. For example, the use of terms like "second" does not necessarily mean that the embodiment or technical solution includes "first," etc.

[0037] As shown in Figure 1, one embodiment of this application provides an optical system.

[0038] The optical system can be a system for processing optical signals. The optical system can be a system containing optical components. These optical components include, but are not limited to, lenses, sensors, gratings, prisms, filters, and beam splitters.

[0039] An optical system includes optical elements. Referring to Figure 2, an optical element can be at least one of a metasurface element, a refractive optical element, a diffractive optical element (DOE), or a scattering medium element.

[0040] The optical system also includes a photosensitive element 13. The photosensitive element 13 can be a light sensor, specifically a CMOS (Complementary Metal Oxide Semiconductor) photosensitive element, a CCD (Charge Coupled Device) photosensitive element, or an array of photodetectors.

[0041] In one embodiment, the optical element includes at least one metasurface element 11.

[0042] In one embodiment, the optical system includes a photosensitive element 13, the optical element including a metasurface element 11 disposed on the light-incident side of the photosensitive element 13. In another embodiment, the optical system includes a photosensitive element 13, the photosensitive element 13 having a metasurface element 11 disposed on the side of the photosensitive element 13 closer to the target object 200.

[0043] In one specific embodiment, the metasurface element 11 is configured as a separate component. In another specific embodiment, the metasurface element 11 is configured independently of the photosensitive element 13.

[0044] In one specific embodiment, the metasurface element 11 is a layer structure formed on the photosensitive element 13. In another specific embodiment, the metasurface element 11 and the photosensitive element 13 are integrally disposed.

[0045] The metasurface element 11 is used to receive at least the optical signal from the target object and to perform optical modulation.

[0046] In one embodiment, the optical element may include multiple sets of metasurface elements. These multiple sets of metasurface elements may have the same or different configuration parameters. In a specific embodiment, the optical system may include a set of metasurface elements 11 and a photosensitive element 13. In a specific embodiment, the optical system may include multiple sets of metasurface elements 11 and a photosensitive element 13.

[0047] In one embodiment, the optical system further includes a focusing lens 14. In one embodiment, the focusing lens 14 is disposed on the light-incident side of the photosensitive element 13. In one embodiment, the focusing lens 14 is used at least to couple data information to the photosensitive element 13.

[0048] An optical system may include a focusing lens. An optical system may also include multiple focusing lenses, which together form a lens group. The focusing lens 14 is used to focus the light signal from the target object 200 to obtain a focused light signal corresponding to the target object 200, and then incident it onto one side of the optical element. For example, the focusing lens 14 incident the focused light signal onto the photosensitive element 13. Alternatively, the focusing lens 14 incident the focused light signal onto the metasurface element 11 and then onto the photosensitive element 13.

[0049] When the metasurface element 11 is disposed on the light-incident side of the photosensitive element 13, for example, when the metasurface element 11 is disposed on the light-incident side of the photosensitive element 13 or when the metasurface element 11 is formed on the surface of the light-incident side of the photosensitive element 13, the focusing lens 14 is used to couple data information to the metasurface element 11, and the metasurface element 11 couples the data information to the photosensitive element 13, or the metasurface element 11 processes the data information and then couples it to the photosensitive element 13.

[0050] In one embodiment, the optical system includes at least one set of metasurface elements, at least one focusing lens 14, and a photosensitive element 13. In another embodiment, the optical system includes multiple sets of metasurface elements and multiple focusing lenses 14, or multiple sets of metasurface elements and multiple focusing lenses 14. The multiple sets of metasurface elements may have the same or different configuration parameters; the multiple focusing lenses 14 may have the same or different configuration parameters. In a specific embodiment, the optical system includes one set of metasurface elements 11, one focusing lens 14, and one photosensitive element 13. In a specific embodiment, the optical system includes one set of metasurface elements 11, multiple focusing lenses 14, and one photosensitive element 13. In a specific embodiment, the optical system includes multiple sets of metasurface elements 11, one focusing lens 14, and one photosensitive element 13. In a specific embodiment, the optical system includes multiple sets of metasurface elements 11, multiple focusing lenses 14, and one photosensitive element 13.

[0051] As shown in Figure 3, the metasurface element 11 and the photosensitive element 13 are arranged independently and separately.

[0052] In one embodiment, the optical system may further include a dielectric layer disposed between the metasurface element 11 and the photosensitive element 13. The material of the dielectric layer may be any material having a low refractive index and absorption coefficient in the visible or near-infrared band, such as: silicon dioxide (SiO2), spin-on glass (SOG), or polymers such as polymethyl methacrylate (PMMA), polydimethylsiloxane (PDMS), polymethylpemtene (PMP), and combinations thereof.

[0053] As shown in Figure 4, the metasurface element 11 and the photosensitive element 13 are integrally disposed. The metasurface element 11 is formed on the light-incident surface of the photosensitive element 13. This facilitates the control of the wavelength and polarization of the incident light, and enables precise capture of complex three-dimensional motion postures.

[0054] In one embodiment, corresponding to the photosensitive element 13, the semiconductor substrate has a photoelectric conversion unit array region; the photoelectric conversion unit array region includes a plurality of photoelectric conversion units. In another embodiment, corresponding to the photosensitive element 13, the light-receiving surface of the semiconductor substrate is formed with metasurface elements arranged in the photoelectric conversion array region, the metasurface elements including a plurality of nanostructure units corresponding to the plurality of photoelectric conversion units, the nanostructure units being used to transmit incident light signals to the corresponding photoelectric conversion units.

[0055] In one embodiment, the nanostructure units for forming the metasurface are disposed on the light-incident side of the optical element. In another embodiment, the nanostructure units for forming the metasurface are disposed on the light-emitting side of the optical element. In yet another embodiment, nanostructure units for forming the metasurface are disposed on both the light-incident and light-emitting sides of the optical element.

[0056] In one embodiment, the metasurface element 11 and the photosensitive element 13 can be combined by means of thermal bonding, plasma bonding, or adhesive bonding; alternatively, the metasurface element 11 can be directly fabricated on the photosensitive element 13. For example, firstly, an optical film layer is formed on the light-receiving surface of a semiconductor substrate, and then the optical film layer is etched into the metasurface element 11 using a photolithography process. For example, the optical film layer can be etched into the metasurface element 11 corresponding to the photoelectric conversion unit array region using a photolithography process.

[0057] In one embodiment, the multiple sets of metasurface elements in the aforementioned embodiments can be independently and separately disposed, or they can be integrally disposed. When integrally disposed, two or more sets of metasurface elements can be combined by means of thermal bonding, plasma bonding, or adhesive bonding; or, another set of metasurface elements can be directly fabricated on the substrate side or the element side of one set of metasurface elements. For example, firstly, an optical film layer is formed on the substrate side or the nanostructure side of one set of metasurface elements, and then the optical film layer is etched into another set of metasurface elements by photolithography.

[0058] In one embodiment, the functional configuration of the optical system can be achieved by updating the configuration of the metasurface element 11.

[0059] The metasurface element described in any embodiment of this application refers to an artificial layered material with a size smaller than or approximately equal to the wavelength, which can be considered as a two-dimensional counterpart of a metamaterial. The metasurface element can achieve the manipulation of the polarization, phase, amplitude, frequency, propagation mode, and angular momentum of electromagnetic waves through subwavelength superstructure units on its surface, realizing properties such as beam shaping, beam deflection, superlensing, superholography, optical rotation, and anti-reflection and anti-reflection effects. The aforementioned polarization, phase, amplitude, and angular momentum properties can be flexibly controlled by adjusting at least one of the configuration parameters such as the geometry, size, or spatial arrangement of the nanostructure units.

[0060] The dimensions include, but are not limited to, radius, side length, and height. The spatial arrangement includes, but is not limited to, arrangement period and duty cycle.

[0061] Meanwhile, metasurface elements can be subwavelength optical elements, which are not only suitable for current micrometer-scale sensor architectures, but also compatible with mature semiconductor sensor technology in terms of fabrication process, thus having strong practicality and economy.

[0062] Specifically, the metasurface element includes a substrate and multiple superstructure units arranged in an array on the substrate. Each superstructure unit has a nanostructure unit at its center or vertex, or both its center and vertex positions. The superstructure unit is formed by dividing the metasurface element, resulting in structural units centered around each nanostructure unit. Each period of nanostructure units constitutes a superstructure unit. The superstructure unit is a close-packed pattern, such as a regular square, regular hexagon, or sector. Each period contains one nanostructure unit, and the vertex or center position of the superstructure unit can have a nanostructure unit, or both its vertex and center positions. In the case of a regular hexagonal superstructure unit, at least one nanostructure unit is located at each vertex and center position. The same applies to sector and square superstructure units.

[0063] The substrate for the metasurface element can be selected from materials with similar refractive indices, such as silicon dioxide, BF33, silicon, and polymethyl methacrylate. The nanostructure unit can be selected from materials such as monomeric silicon (c-Si), polycrystalline silicon (p-Si), amorphous silicon (a-Si), compound semiconductors (such as GaN, GaP, GaAs, SiC, etc.), TiO2, Si3N4, AlSb, AlAs, AlGaAs, AlGaInP, BP, ZnGeP2, and other suitable materials, as well as combinations of the above materials.

[0064] Specifically, depending on the application scenario, the nanostructure unit can be configured as a polarization-dependent structure or a polarization-independent structure. Polarization-independent structures include, for example, cylindrical, square prisms, cross-shaped prisms, and square prisms with circular holes. Polarization-dependent structures include, for example, elliptical cylinders, rectangular prisms, and hexagonal prisms. The nanostructure unit can be a positive or negative structure. For example, the shapes of the nanostructure unit include cylinders, hollow cylinders, square prisms, and hollow square prisms.

[0065] Nanostructures can also be combinations of various shapes, such as circles, crosses, regular polygons, rectangles, or any combination of these shapes.

[0066] Metasurface elements may also include a protective layer covering the nanostructure units. The material of the protective layer can be any material with a low refractive index and absorption coefficient in the visible or near-infrared band; for example: silicon dioxide (SiO2), spin-coated glass (SOG), or polymers such as polymethyl methacrylate (PMMA), polydimethylsiloxane (PDMS), polymethylpentene (PMP), and combinations of the above materials, or it can be air (i.e., no protective layer).

[0067] In one embodiment, the optical element may further include at least one of a refractive optical element, a diffractive optical element, and a scattering medium element; one or more refractive optical elements may be provided; one or more diffractive optical elements may be provided; one or more scattering medium elements may be provided. The refractive optical element may include a common spherical refractive optical element or an aspherical refractive optical element, specifically including but not limited to lenses or prisms made of materials such as optical glass, optical plastics, and optical crystals; the diffractive optical element includes but is not limited to two-step or multi-step diffractive optical elements, gratings, Dammann gratings, metasurfaces, holograms, diffusers, phase masks, intensity masks, spatial light modulators, etc.; the scattering medium element includes but is not limited to frosted glass, etc.

[0068] In one embodiment, the functional configuration of the optical system can be updated by updating the configuration of the refractive optical elements. In another embodiment, the functional configuration of the optical system can be updated by updating the configuration of the diffractive optical elements. In yet another embodiment, the functional configuration of the optical system can be updated by updating the configuration of the scattering medium elements.

[0069] The optical system includes a processing module 12.

[0070] Processing module 12 is coupled to an optical element. Processing module 12 is used to process data information from the optical element according to a first program. In one embodiment, the optical system includes a photosensitive element 13, and processing module 12 is coupled to the photosensitive element 13 to process data information from the photosensitive element 13 according to a first program.

[0071] The first program can be a collection of instruction information stored in a storage unit inside the processing module 12, or it can be a collection of instruction information stored in a separate storage unit outside the processing module 12. In one embodiment, the first program is used to implement a first neural network model algorithm. In one embodiment, the processing module 12 operates to implement the first neural network model algorithm. The first neural network model is used to process data information from optical elements (e.g., metasurface elements 11); in a specific embodiment, the first neural network is used to identify a target object 200 located on the light-incident side.

[0072] Processing module 12 can be used to execute only the first program to implement the first neural network model algorithm. In one embodiment, the optical system may include multiple processing modules, each executing a different program to implement a different neural network model algorithm.

[0073] The neural network model algorithm can have a one-to-one correspondence with the optical element. In this embodiment, the optical element is coupled to the corresponding processing module.

[0074] The processing module 12 can be used to execute various programs and implement various neural network model algorithms.

[0075] Neural network model algorithms can have a one-to-one correspondence with optical elements. In this embodiment, multiple optical elements are coupled to a corresponding processing module, which is used to execute various neural network model algorithms corresponding to the multiple optical elements.

[0076] The optical system does not necessarily only image or process data on the target object 200. In one embodiment, the optical system is used to image or process data on an imaging object, which includes the target object 200. The optical system performs optical processing on the target object 200 portion of the imaging object using the metasurface element 11; or, it uses the processing module 12 to process the data information corresponding to the target object 200 to achieve, for example, pose recognition of the target object 200; or, it performs both optical processing and pose recognition.

[0077] The processing module 12 can couple the optical element electrically, optically, or a combination of both. In one embodiment, the optical system includes a photosensitive element 13, which converts the received optical signal into data information in the form of an electrical signal and transmits it to the processing module 12, thus establishing an electrical connection between the processing module 12 and the optical element. In another embodiment, the optical element includes a metasurface element 11, which processes the received optical signal and transmits the data information in the form of an optical signal to the processing module 12, thus establishing an optical coupling between the processing module 12 and the optical element.

[0078] The coupling between the processing module 12 and the optical element can be achieved through direct or indirect connection.

[0079] In one embodiment, the optical system is configured according to configuration information obtained by a configuration method.

[0080] In one specific embodiment, the configuration method of the optical system includes: acquiring sample data; inputting first data into an optical element, executing a first program to process intermediate data, and determining an evaluation value based on the intermediate data and second data; updating the configuration of the optical element, or updating the configuration of the first program, or updating both the configuration of the optical element and the configuration of the first program, based on the evaluation value, until the evaluation value meets a preset numerical condition, thereby obtaining configuration information of the optical system. The sample data includes first data corresponding to the input information of the optical element, and second data corresponding to the information output when executing the first program; the second data is target data corresponding to the first data.

[0081] The first data corresponds to the input information of the optical element. The first data can be information used to input the optical element, or it can have a form, feature, format, standard, etc., that corresponds to the input information of the optical element.

[0082] In other embodiments, the optical system may also be configured using the configuration method of the optical system described in any of the embodiments below.

[0083] For example, in an embodiment where the optical element includes a metasurface element, the configuration method of the optical system includes: updating the configuration of the metasurface element, or updating the configuration of the first program, or updating both the configuration of the metasurface element and the configuration of the first program based on the evaluation value, until the evaluation value meets a preset numerical condition, thereby obtaining the configuration information of the optical system.

[0084] For example, in an embodiment where the optical element includes at least one of a refractive optical element, a diffractive optical element, and a scattering medium element, the configuration method of the optical system includes: updating the configuration of the optical element based on the evaluation value, or updating the configuration of a first program, or updating both the configuration of the optical element and the configuration of the first program, until the evaluation value meets a preset numerical condition, thereby obtaining the configuration information of the optical system.

[0085] In one embodiment, the optical system is used to implement an identification method.

[0086] In one specific embodiment, the identification method includes: configuring an optical system according to configuration information obtained by a configuration method; controlling optical elements to obtain a detection image of a corresponding target object; controlling a processing module to execute a first program to process the detection image and determine the identification information of the corresponding target object; for example, the identification information includes posture recognition information.

[0087] In other embodiments, the optical system may also be configured in any of the embodiments described below to implement the recognition method.

[0088] In one embodiment, the optical system may include a data acquisition module 10 for acquiring data information corresponding to a target object. This data information may be in the form of an optical signal. In a specific embodiment, the data acquisition module 10 is used to acquire an image of the target object 200, which is then processed by the processing module 12 to identify the target object based on the image. For example, the data acquisition module 10 is used to acquire a posture image of the target object 200, which is then processed by the processing module 12 to determine the posture recognition information of the target object based on the posture image.

[0089] The acquisition module 10 may include the aforementioned optical elements. In one embodiment, the optical system includes a photosensitive element 13, and the acquisition module 10 includes the photosensitive element 13. In one embodiment, the optical element includes a metasurface element 11, and the acquisition module 10 includes the metasurface element 11. In one embodiment, the acquisition module 10 includes a focusing lens 14.

[0090] As shown in Figure 5, in one embodiment, the optical system includes a first optical element.

[0091] In one embodiment, the first optical element includes a first metasurface element 11A. In another embodiment, the first optical element includes at least one of a refractive optical element, a diffractive optical element, and a scattering medium element; one or more refractive optical elements may be provided; one or more diffractive optical elements may be provided; one or more scattering medium elements may be provided.

[0092] In one embodiment, the optical system includes a first photosensitive element 13A. A first metasurface element 11A is disposed on the light-incident side of the first photosensitive element 13A.

[0093] In one specific embodiment, the first metasurface element 11A is formed on the surface of the first photosensitive element 13A on the light-incident side, and the two are integrally disposed. In another specific embodiment, the first metasurface element 11A is disposed on the light-incident side of the first photosensitive element 13A, and the two are separately disposed.

[0094] In one embodiment, the optical system includes a second optical element.

[0095] In one embodiment, the second optical element includes a second metasurface element 11B. In another embodiment, the second optical element includes at least one of a refractive optical element, a diffractive optical element, and a scattering medium element; one or more refractive optical elements may be provided; one or more diffractive optical elements may be provided; one or more scattering medium elements may be provided.

[0096] In one embodiment, the optical system includes a second photosensitive element 13B. A second metasurface element 11B is disposed on the light-incident side of the second photosensitive element 13B.

[0097] In one specific embodiment, the second metasurface element 11B is formed on the light-incident surface of the second photosensitive element 13B, and the two are integrally disposed. In another specific embodiment, the second metasurface element 11B is disposed on the light-incident side of the second photosensitive element 13B, and the two are separately disposed.

[0098] In one embodiment, the optical system includes a first processing module 12A.

[0099] In one embodiment, the first processing module 12A is coupled to a first optical element. In a specific embodiment, the first processing module 12A is coupled to a first photosensitive element 13A. In a specific embodiment, the first processing module 12A is coupled to a first metasurface element 11A.

[0100] The first processing module 12A is used to process data information from the first optical element according to a first program. In one embodiment, the first processing module 12A is used to process data information from the first metasurface element 11A according to the first program; the first processing module 12A is also used to process data information from the first photosensitive element 13A according to the first program.

[0101] In one embodiment, the optical system includes a second processing module 12B.

[0102] In one embodiment, the second processing module 12B is coupled to a second optical element. In a specific embodiment, the second processing module 12B is coupled to a second photosensitive element 13B. In a specific embodiment, the second processing module 12B is coupled to a second metasurface element 11B.

[0103] The second processing module 12B is used to process data information from the second optical element according to the first program. In one embodiment, the second processing module 12B is used to process data information from the second metasurface element 11B according to the first program; the second processing module 12B is used to process data information from the second photosensitive element 13B according to the first program.

[0104] In one embodiment, the first metasurface element 11A and the first photosensitive element 13A can be used to form a first acquisition module 10A, and the second metasurface element 11B and the second photosensitive element 13B can be used to form a second acquisition module 10B.

[0105] In one specific embodiment, the first acquisition module 10A may further include a focusing lens corresponding to the first metasurface element 11A, and the second acquisition module 10B may further include a focusing lens corresponding to the second metasurface element 11B.

[0106] In one specific embodiment, the optical system may include two or more of the aforementioned acquisition modules.

[0107] In one embodiment, a first optical element is used to generate a first intermediate image corresponding to a target object. The first intermediate image can be an image of the target object alone, or it can be an image of an object containing the target object. For the former, the first intermediate image contains only data information of the target object, and may specifically contain identification information of the target object; for example, the identification information includes posture recognition information. For the latter, the first intermediate image contains data information of the entire imaged object, and at least includes identification information of the target object; for example, the identification information includes posture recognition information.

[0108] In one embodiment, a second optical element is used to generate a second intermediate image corresponding to a target object. The second intermediate image may be an image of the target object alone, or it may be an image of an object containing the target object.

[0109] In one embodiment, the first intermediate image is one of a clear image, a blurred image, and a contour image, and the second intermediate image is another of a clear image, a blurred image, and a contour image. Thus, the constructed optical system can achieve different imaging effects and generate different types of images through the first optical element, the first processing module 12A, the second optical element, and the second processing module 12B.

[0110] The type of the first intermediate image can be set according to the user's needs. The type of the second image can also be set according to the user's needs.

[0111] When at least one set of optical elements and processing modules are used to generate blurred and outline images, since the specific content of the image cannot be intuitively distinguished from such intermediate images, the blurred and outline images can be directly output for rendering, which can also meet the user's privacy protection needs.

[0112] In this embodiment, the following specific implementations may exist:

[0113] The first intermediate image is a clear image, and the second intermediate image is a blurred image. In this way, the original information and the blurred encrypted information can be compared to determine the effect of the blur encryption.

[0114] The first intermediate image is a clear image, and the second intermediate image is a contour image. Thus, the original information and the contour information can be compared, or the contour information can be analyzed first to obtain recognition information, and then the recognition information can be compared with the original information to determine the encryption effect; for example, the recognition information may include pose recognition information.

[0115] The first intermediate image is a blurred image, and the second intermediate image is a contour image. In this way, user privacy can be fully protected, and on this basis, posture recognition information can be analyzed based on the contour image to understand the posture of the target object.

[0116] In one embodiment, when configuring or implementing the optical system, a step of determining the type of the intermediate image can be added. For example, determining whether the sharpness of the first intermediate image meets a preset sharpness condition; or determining whether the blurriness of the first intermediate image meets a preset blurriness condition; or determining whether the first intermediate image shows contour information. This determination step can be implemented by a corresponding processing module.

[0117] In this embodiment, since the expected first intermediate image and the second intermediate image are of different types, different judgment steps can be performed accordingly. For example, if the expected first intermediate image is a blurred image, the first processing module 12A can perform the step of judging whether the blur degree of the first intermediate image meets the preset blur degree condition; for example, if the expected second intermediate image is a contour image, the second processing module 12B can perform the step of judging whether the second intermediate image shows contour information.

[0118] The degree of blurring can be characterized by a blur value. The blur value can be determined by determining the first pixel channel value of the intermediate image and performing a convolution operation on that pixel channel value using a Laplacian mask; the first pixel channel value can be a grayscale value. The blur value can be determined using a digital image blurring algorithm based on grayscale values.

[0119] The contour information can be represented by edge information. If the change in pixel channel value in the intermediate image is greater than a preset change value, this position is determined to be an edge position, and the intermediate image is determined to include edge information. In one embodiment, an edge detection operator can be used to identify pixels in the intermediate image that meet the requirements, thereby determining whether the intermediate image includes edge information; the pixels that meet the requirements can be pixels whose pixel channel value changes are greater than a preset change value. In a specific embodiment, the edge detection operator includes at least one of the Sobel operator, the Prewitt operator, or the Canny operator. In one embodiment, the pixel channel value can be a brightness value.

[0120] In one embodiment, the first intermediate image has a first degree of blur, and the second intermediate image has a second degree of blur. The first degree of blur and the second degree of blur are different. Thus, the constructed optical system can achieve different imaging effects, such as generating images with different degrees of blur, through the first optical element, the first processing module 12A, the second optical element, and the second processing module 12B, respectively.

[0121] The blur level of the first intermediate image can be set according to the user's needs. The blur level of the second intermediate image can also be set according to the user's needs.

[0122] In one embodiment, when configuring or using the optical system, a step of judging the blur level of the intermediate image can be added. For example, it can be determined whether the blur level of the first intermediate image meets a preset first blur level condition; and whether the blur level of the second intermediate image meets a preset second blur level condition. Of course, the blur level of the intermediate image can also be evaluated by judging the sharpness of the intermediate image; the higher the sharpness, the lower the blur level and the worse the encryption effect.

[0123] The degree of ambiguity can be characterized by the point spread function value; specifically, the level of ambiguity can be quantified by the point spread function value.

[0124] In one embodiment, the standard function value can be calculated using the scalar diffraction formula, or the standard element can be obtained and the standard function value determined through simulation.

[0125] The standard function value can be used as a standard for quantitative evaluation of the degree of ambiguity; for example, the standard function value can be compared with the actual point spread function value of the optical element in response to the data, and the difference between the two can be used to determine whether the intermediate image generated by the optical element or optical system meets the corresponding degree of ambiguity conditions.

[0126] In one specific embodiment, the standard function value can be characterized by a point spread function standard value that conforms to the expected degree of ambiguity.

[0127] In this embodiment, the standard function value can be calculated and determined based on the scalar diffraction formula according to the expected degree of ambiguity. Alternatively, a standard element capable of achieving the expected ambiguity processing can be obtained, and the data can be input into the standard element. The standard function value corresponding to the expected degree of ambiguity can be determined based on the point spread function value of this process.

[0128] In this embodiment, the actual point spread function value can be directly compared with the standard function value, and when the difference between the two is within the allowable error range, it can be determined that the intermediate image corresponding to the actual point spread function value meets the expected blur level requirement.

[0129] In one specific embodiment, the standard function value can be characterized by a standard value of the dot spread function that conforms to the desired sharpness; correspondingly, the expected difference between the actual dot spread function value and the standard function value can be set according to the expected sharpness.

[0130] In this embodiment, the standard function value can be calculated and determined based on the scalar diffraction formula according to the sharpness requirements. Alternatively, a standard element capable of generating an intermediate image that meets the sharpness requirements can be obtained, the data can be input into the standard element, the point spread function value of this process can be determined, and the standard function value corresponding to the sharpness requirements can be determined.

[0131] In one embodiment, it is necessary to compare the difference between the actual point spread function value and the standard function value with a preset threshold condition. The threshold condition is used to represent the expected difference. For example, the expected point spread function value can be calculated and determined based on the scalar diffraction formula according to the expected degree of ambiguity, and the threshold condition can be set according to its difference from the standard function value. Alternatively, a standard element capable of achieving the expected ambiguity processing can be obtained, data can be input into the standard element, the point spread function value corresponding to the expected degree of ambiguity can be determined, and the threshold condition can be set according to its difference from the standard function value.

[0132] Although this embodiment provides two sets of optical elements and corresponding processing modules, this application is not limited to setting two sets in the optical system. In other embodiments, the optical elements and their corresponding processing modules can be set to three or more sets, and different sets of optical elements and processing modules can be used to generate the same or different intermediate images.

[0133] In the embodiment shown in Figure 5, the first acquisition module 10A and the second acquisition module 10B respectively include a first processing module 12A and a second processing module 12B. When configuring the optical system using the configuration method, the first acquisition module 10A and the first processing module 12A can be configured synchronously or asynchronously, as can the second acquisition module 10B and the second processing module 12B.

[0134] For example, the first data can be input into the first metasurface element 11A and the first photosensitive element 13A. The first processing module 12A executes the first program to process the intermediate data and determine the evaluation value. At this time, the configuration of the first metasurface element 11A, the configuration of the first program, or both can be updated based on the evaluation value, thereby determining the configuration information of the first acquisition module 10A and the first processing module 12A.

[0135] When the optical elements are of other types, the configuration of other types of optical elements can also be updated based on the evaluation values.

[0136] As shown in Figure 6, in one embodiment, the optical system includes a first optical element.

[0137] In one embodiment, the first optical element includes a first metasurface element 11A. In another embodiment, the optical system includes a first photosensitive element 13A. The first metasurface element 11A is disposed on the light-incident side of the first photosensitive element 13A.

[0138] In one specific embodiment, the first metasurface element 11A is formed on the surface of the first photosensitive element 13A on the light-incident side, and the two are integrally disposed. In another specific embodiment, the first metasurface element 11A is disposed on the light-incident side of the first photosensitive element 13A, and the two are separately disposed.

[0139] In one embodiment, the optical system includes a second optical element.

[0140] In one embodiment, the second optical element includes a second metasurface element 11B. In another embodiment, the optical system includes a second photosensitive element 13B. The second metasurface element 11B is disposed on the light-incident side of the second photosensitive element 13B.

[0141] In one specific embodiment, the second metasurface element 11B is formed on the light-incident surface of the second photosensitive element 13B, and the two are integrally disposed. In another specific embodiment, the second metasurface element 11B is disposed on the light-incident side of the second photosensitive element 13B, and the two are separately disposed.

[0142] In one embodiment, the first optical element is disposed at a first position relative to the target object 200, and the second optical element is disposed at a second position relative to the target object 200.

[0143] In one specific embodiment, the first orientation and the second orientation are two different orientations relative to the target object 200. Thus, the optical system can collect information about the target object 200 from different orientations, and use this information to comprehensively analyze the data from the corresponding first and second optical elements. In particular, when the optical system analyzes the identification information of the target object 200, applying multi-directional detection using optical elements can obtain more accurate identification information; for example, the identification information includes posture recognition information.

[0144] In one embodiment, the first orientation is one of the following: directly in front of the target object 200, to the left of the target object, to the right of the target object, to the bottom of the target object, or to the top of the target object.

[0145] In one embodiment, the second orientation is one of the following: directly in front of the target object 200, to the left of the target object, to the right of the target object, to the bottom of the target object, or to the top of the target object.

[0146] In one specific embodiment, the left front direction can be a 45-degree direction to the left of the target object 200; the right front direction can be a 45-degree direction to the right of the target object 200; the lower front direction can be a 45-degree direction to the lower front of the target object 200; and the upper front direction can be a 45-degree direction to the upper front of the target object 200.

[0147] In one embodiment, the first optical element includes a first metasurface element 11A, which is disposed at a first orientation relative to the target object 200. In a specific embodiment, the first metasurface element 11A is disposed on the light-incident side of the first photosensitive element 13A.

[0148] The first photosensitive element 13A and the first metasurface element 11A can be used to form the first acquisition module 10A. In one embodiment, the first acquisition module 10A can be positioned as a whole in a first orientation relative to the target object 200. In a specific embodiment, the first acquisition module 10A further includes a focusing lens.

[0149] The second photosensitive element 13B and the second metasurface element 11B can be used to form the second acquisition module 10B. In one embodiment, the second acquisition module 10B can be positioned entirely in a second orientation relative to the target object 200. In a specific embodiment, the second acquisition module 10B further includes a focusing lens.

[0150] In one embodiment, the optical system may be equipped with a processing module 12 corresponding to multiple optical elements located in different orientations.

[0151] The processing module 12 is coupled to several optical elements in the optical system and processes data from multiple optical elements according to a first program. Thus, the processing module 12 can perform a comprehensive analysis of the target object 200 based on data acquired from multiple perspectives.

[0152] When configuring an optical system, multiple optical components and a processing module can be configured as a whole.

[0153] For example, the first data can be input into the first optical element (e.g., the first metasurface element 11A) to obtain an intermediate image corresponding to one orientation, and the first data can be input into the second optical element (e.g., the second metasurface element 11B) to obtain an intermediate image corresponding to another orientation. The processing module 12 performs a first program processing based on the intermediate images corresponding to different orientations to obtain intermediate data and determine an evaluation value. At this time, the configuration of the first optical element (e.g., the first metasurface element 11A), the configuration of the second optical element (e.g., the second metasurface element 11B), or at least one of the configurations of the first program can be uniformly updated based on the evaluation value, thereby determining the configuration information of the first acquisition module 10A, the second acquisition module 10B, and the processing module 12.

[0154] When the optical system includes more optical elements, these optical elements can also be configured together with the processing module 12 to ensure that the optical system has multi-directional information processing capabilities and obtains more accurate data processing results (e.g., obtaining more accurate recognition information; for example, the recognition information includes attitude recognition information).

[0155] As shown in Figure 7, one embodiment of this application provides a method for configuring an optical system.

[0156] This method can be used to configure the structural features of components within an optical system, such as the structural features of the component itself, or the configuration of the component in relation to the function it is to perform (e.g., the distance between components).

[0157] This method can be used to configure parameters of programs within a fixed optical system, such as program instructions and model parameters of neural network models.

[0158] The optical system can be configured as described in any of the embodiments above.

[0159] For example, an optical system includes optical elements.

[0160] As shown in Figure 2, the optical element may include at least one of the following: metasurface element, refractive optical element, diffractive optical element, and scattering medium element; one or more refractive optical elements may be provided; one or more diffractive optical elements may be provided; one or more scattering medium elements may be provided.

[0161] The optical system also includes a photosensitive element 13.

[0162] In one embodiment, the optical element includes at least one metasurface element 11.

[0163] In one embodiment, the metasurface element 11 is disposed on the light-incident side of the photosensitive element 13.

[0164] In one embodiment, the optical element is used to construct a data acquisition module 10, which includes a photosensitive element 13 and a metasurface element 11; in a specific embodiment, the data acquisition module 10 may further include a focusing lens 14.

[0165] In one specific embodiment, the optical element or acquisition module 10 is used to generate an intermediate image based on the information input thereto. For example, the intermediate image may be a posture image; the posture image is an image carrying the posture status of the target object; the posture status may be the outline of the target object as shown in FIG11(b), or nodes indicating joints of the target object as shown in FIG12, or connecting lines indicating limbs of the target object as shown in FIG12.

[0166] The optical system includes a processing module 12. The processing module 12 is coupled to an optical element. The processing module 12 is used to process data information from the optical element according to a first program. In one embodiment, the optical element includes a metasurface element 11, and the processing module 12 is coupled to the metasurface element 11 to process data information from the metasurface element 11 according to the first program. In a specific embodiment, the optical system includes a photosensitive element 13, the metasurface element 11 is coupled to the photosensitive element 13, and the processing module 12 is coupled to the metasurface element 11 through the photosensitive element 13.

[0167] In one embodiment, when the first program is executed by the processing module 12, it implements a neural network model. In one specific embodiment, the neural network model is a convolutional neural network model. In other specific embodiments, the neural network model can be other deep learning models, such as feedforward neural network models, recurrent neural network models, generative adversarial network models, autoencoders, etc.

[0168] In one specific embodiment, when the first program is executed by the processing module, it is used to determine the identification information of the corresponding target object based on the information of the input optical element. For example, the identification information includes posture recognition information.

[0169] If the optical element is used to generate an attitude image as an intermediate image, then the first program, when executed by the processing module, is used to determine the attitude recognition information of the corresponding target object based on the attitude image.

[0170] The method for configuring an optical system specifically includes at least one of the following steps.

[0171] Step S1: Obtain sample data.

[0172] The sample data includes first data; the first data corresponds to the information input to the optical element.

[0173] In one embodiment, the first data is in the form of an optical signal. The first data can be directly input to an optical element, which modulates it and outputs it to the subsequent stage of the optical element in the optical system (i.e., other components located after the optical element in the light propagation path).

[0174] In one embodiment, the first data is in the form of an electrical signal. The first data may first be converted into an optical signal and then input to an optical element.

[0175] In other embodiments, the configuration process of the optical system is based on an optical path simulation platform, and the optical element is constructed as an electronic simulation model, so that first data in the form of electrical signals can be used as input data for the optical element.

[0176] The sample data includes second data; the second data corresponds to the information output by the processing module when executing the first program. The second data is the target data corresponding to the first data.

[0177] For a target optical system that has already been configured, first data is input to the optical elements within the target optical system. The processing module of the target optical system can then execute a first program to obtain second data corresponding to the first data. Therefore, the optical system to be configured can be functionally configured in a direction that approximates the target optical system using the first and second data.

[0178] In one embodiment, the first data is an optical signal corresponding to the sample object; in another embodiment, light rays from the sample object form an optical signal that can be used as data input to optical elements when configuring the optical system.

[0179] The optical components and processing module are jointly trained and configured based on sample data. In one embodiment, during the joint configuration process, a first piece of data from the sample data can be used as input data, and a second piece of data from the sample data can be used as training labels.

[0180] In one embodiment, the first data includes a sample image. In this embodiment, when configuring the optical system, the sample image can be directly used as the input to the optical element, and the optical element directly modulates the sample image, improving configuration efficiency. In some embodiments, when implementing the optical system, image information can be used as the input to the optical element; alternatively, light signals, electrical signals, etc., corresponding to the target object can also be used as the input to the optical element.

[0181] In one embodiment, the second data includes actual identification information. In this embodiment, the target optical system can analyze the characteristics of the object based on the input from one side of the optical element. Therefore, when configuring the optical system, the actual identification information can be used as the target data of the optical system. In one embodiment, the actual identification information corresponds to a sample image; in a specific embodiment, the actual identification information corresponds to the portion of the sample image that represents the target object.

[0182] In one specific embodiment, the actual identification information characterizes the actual features of the target object in the sample image. Thus, by using the actual identification information as the target data corresponding to the first data, and configuring it accordingly, the resulting optical system can accurately infer the identification information of the target object.

[0183] The identification information proposed in this application may include posture recognition information. This posture recognition information is used to help understand human posture and movements, enabling natural human-computer interaction and real-time monitoring of body movements. The content of the determined posture information may be the same or different depending on the application scenario.

[0184] When optical systems are applied to posture recognition scenarios, they can be used to identify whether a sitting posture is standard. The identified information can include: normal sitting posture, leaning forward, leaning backward, leaning to the side, hunching over, slouching, looking down, shoulder tilting to the left, shoulder tilting to the right, and single leg raised.

[0185] When optical systems are applied to fall or stumble detection scenarios, the recognition information determined by the optical system can include posture recognition information such as stumbles and stumbles.

[0186] When an optical system is applied to a lighting control scenario, the recognition information determined by the optical system may include gesture recognition information. This gesture recognition information can be used to control the lighting on and off.

[0187] When an optical system is applied to a DMS (Driver Monitor System), the recognition information determined by the optical system can include gesture recognition information. This gesture recognition information can be used to control vehicle functions.

[0188] When optical systems are applied to factory compliance inspection scenarios, the identification information determined by the optical systems can include posture recognition information such as wearing, hand washing, and smoking.

[0189] When optical systems are applied to household appliances such as washing machines, refrigerators, and water purifiers, the identification information determined by the optical system can be used to control the devices, or it can include control signals for controlling the devices.

[0190] The identification information proposed in this application is not limited to analyzing the posture of living bodies such as humans and animals, but can also be used to analyze the posture of objects such as motor vehicles and non-motor vehicles.

[0191] When optical systems are applied to smart street monitoring scenarios, they can be used to determine whether non-motorized vehicles are illegally parked in a designated street area. In this case, the identified information can include the posture of the parked non-motorized vehicles.

[0192] Optical systems can also be used to determine whether street vending is operating illegally in a designated street area. In this case, the identified information can include: attitude recognition information of all objects moving within the designated street area.

[0193] Optical systems can also be used to determine whether mobile vendors exist within a designated street area. In this case, the determined identification information may include: attitude recognition information of objects moving at fixed points within the designated street area and objects entering and leaving the area.

[0194] Optical systems can also be used to determine whether there are illegally parked motor vehicles within a designated street area. In this case, the identified information may include: the posture of the parked motor vehicle.

[0195] Optical systems can also be used to determine whether non-motorized vehicles are entering motorized vehicle lanes within a designated street area. In this case, the identified information can include: the posture recognition information of the non-motorized vehicles, and the recognition information of the positional relationship between the non-motorized vehicles and the motorized vehicle lanes.

[0196] Optical systems can also be used to determine whether there is road congestion within a specified street area. In this case, the identified information can include the attitude recognition information of all objects within the road.

[0197] Optical systems can also be used to determine whether there is dirt on the road surface within a specified street area. In this case, the identified information can include the posture recognition information of all objects on the road surface.

[0198] Optical systems can also be used to determine whether construction waste exists within a designated street area. In this case, the identified information can include the posture recognition information of the construction waste on the street.

[0199] Optical systems can also be used to determine whether litter exists in a designated street area. In this case, the identified information can include the posture of the litter on the street.

[0200] Optical systems can also be used to determine whether there is a problem of haphazardly piled materials within a designated street area. In this case, the identified information can include: information on the posture of the materials piled up on the street.

[0201] Step S2: Input the first data into the optical element, execute the first program to obtain intermediate data, and determine the evaluation value based on the intermediate data and the second data.

[0202] Regarding the relationship between existing optical simulation methods and the configuration method provided in this application:

[0203] On the one hand, in the optical system proposed in this application, the optical response of the optical element can be determined by implementing existing optical simulation methods (e.g., constructing an optical model corresponding to the optical element).

[0204] On the other hand, in the configuration method proposed in this application, after the first data is input into the optical element, the first program processes the output of the optical element as input to obtain intermediate data. That is, the intermediate data is obtained after the first data is processed by the optical element and the first program. Based on this, the evaluation value is determined according to the intermediate data and the second data (in one embodiment, the second data can be considered as a tag value), thereby guiding the configuration of the optical system. It can be seen that both the optical element and the first program participate in the construction of the evaluation value, and the construction of the evaluation value does not rely solely on the optical element or the first program.

[0205] Furthermore, the output of an optical element can be determined using the optical response or optical model of the optical element as determined by existing optical simulation methods. That is, the output of an optical element can be calculated and determined based on the optical response relationship simulated for that optical element, without requiring the optical element to be a physical component. Alternatively, the output of an optical element can also be determined experimentally by constructing a physical optical element.

[0206] In one embodiment, the intermediate data includes predicted identification information. In this embodiment, the optical system may generate predicted identification information that is the same as or different from the actual identification information during configuration. The process of configuring the optical system based on the second data and the intermediate data can be specifically a process of configuring the optical system based on the actual identification information and the predicted identification information.

[0207] In one specific embodiment, the optical system determines an evaluation value based on actual recognition information and predicted recognition information, and configures the optical system based on the evaluation value.

[0208] In one specific embodiment, the optical system is configured based on the difference between the actual recognition information and the predicted recognition information.

[0209] In one embodiment, the predicted recognition information corresponds to a sample image; in a specific embodiment, the predicted recognition information corresponds to the portion of the sample image that represents the target object.

[0210] In one specific embodiment, the predicted recognition information is characterized by features determined by a first program based on a sample image. Thus, by using the sample image as input to the optical element, determining the predicted recognition information through the first program, and configuring the actual recognition information corresponding to the sample image as the target, not only can a unified joint configuration of the optical element and the first program be achieved, but the resulting optical system also possesses a recognition function "from image to recognition information".

[0211] In other embodiments, an optical system with other recognition functions may also be configured. These other recognition functions can be constructed by adjusting the optical elements, the inputs and outputs of the first program.

[0212] In one embodiment, the first program, when executed, is used to implement a first neural network model algorithm. In a specific embodiment, the first neural network model algorithm, when implemented, is used to output recognition information; for example, the recognition information includes pose recognition information.

[0213] In one embodiment, the first program includes at least one set of program instructions, and the program instructions include multiple instruction parameters. The configuration method provided in this application can be specifically used to configure these instruction parameters.

[0214] In one embodiment, the first program is used to implement a first neural network model algorithm, which includes multiple model parameters. The configuration method provided in this application can be specifically used to configure these model parameters.

[0215] The model parameters may include at least one of the following: parameters for defining the model architecture, model weight parameters, model bias parameters, initialization parameters, hyperparameters (e.g., learning rate, optimizer, loss function, regularization parameter, batch size, training period, learning rate scheduling strategy), and process parameters (e.g., gradient, momentum, second moment estimation).

[0216] In one embodiment, the evaluation value is used to assess the performance of the current optical system, particularly the difference between the current performance and the performance matched to the target task. In another embodiment, the evaluation value is characterized by the difference between intermediate data and second data.

[0217] The evaluation value can be used to update the configuration of the optical system. Specifically, a standard value or standard range can be set for the evaluation value, and the determination of whether to update the configuration of the optical system or how to update the configuration can be made based on the numerical relationship between the evaluation value and the standard value, or the numerical relationship between the evaluation value and the standard range. For example, when the evaluation value is not equal to the standard value, or when the evaluation value does not fall within the standard range, it is determined that the configuration of the optical system should be adjusted.

[0218] Configuration updates can be based on evaluation values, or more specifically, configuration updates can be performed until the evaluation values ​​converge.

[0219] The configuration of the optical system includes configuration parameters of optical elements, configuration parameters of the first program, or both configuration parameters of optical elements and the first program.

[0220] In one embodiment, the evaluation value includes the loss value of intermediate data and second data.

[0221] The loss value is used to characterize the difference between the intermediate data and the second data. The loss value can be used to reflect the gap between the currently configured optical system and the target optical system.

[0222] The loss value can be determined based on the cross-entropy of the intermediate data and the second data, KL divergence, Hinge loss, etc. In one embodiment, the loss value is determined based on the mean square error of the intermediate data and the second data.

[0223] In one embodiment, the optical system includes multiple optical elements coupled to the same processing module, which executes a first program to obtain unified intermediate data.

[0224] In this embodiment, the first data can be input into multiple optical elements respectively. The processing module receives the data and executes a first program to obtain intermediate data based on the outputs from the multiple optical elements. The multiple optical elements can be positioned at different locations on the target object; the multiple optical elements can be used to generate different types of intermediate images; the multiple optical elements can be configured as in any of the embodiments provided in this application.

[0225] For example, an optical system includes a first optical element.

[0226] In one embodiment, the first optical element includes a first metasurface element. In another embodiment, the first optical element includes at least one of a refractive optical element, a diffractive optical element, and a scattering medium element; one or more refractive optical elements may be provided; one or more diffractive optical elements may be provided; one or more scattering medium elements may be provided.

[0227] In one embodiment, the optical system includes a first photosensitive element. A first metasurface element is disposed on the light-incident side of the first photosensitive element.

[0228] In one specific embodiment, the first metasurface element is formed on the surface of the light-incident side of the first photosensitive element, and the two are integrally disposed. In another specific embodiment, the first metasurface element is disposed on the light-incident side of the first photosensitive element, and the two are separately disposed.

[0229] For example, an optical system includes a second optical element.

[0230] In one embodiment, the second optical element includes a second metasurface element. In another embodiment, the second optical element includes at least one of a refractive optical element, a diffractive optical element, and a scattering medium element; one or more refractive optical elements may be provided; one or more diffractive optical elements may be provided; one or more scattering medium elements may be provided.

[0231] In one embodiment, the optical system includes a second photosensitive element. A second metasurface element is disposed on the light-incident side of the second photosensitive element.

[0232] In one specific embodiment, the second metasurface element is formed on the surface of the light-incident side of the second photosensitive element, and the two are integrally disposed. In another specific embodiment, the second metasurface element is disposed on the light-incident side of the second photosensitive element, and the two are separately disposed.

[0233] In one embodiment, the processing module is coupled to the first optical element and the second optical element respectively, and is used to process data information from the first optical element and the second optical element according to a first program.

[0234] In one embodiment, the optical system includes multiple optical elements coupled to different processing modules, which execute a first program to obtain the same or different intermediate data.

[0235] In this embodiment, the first data can be input into multiple optical elements, and different processing modules receive and execute a first program based on the output of the corresponding optical elements to obtain multiple sets of intermediate data. The multiple optical elements can be positioned at different locations on the target object; the multiple optical elements can be used to generate different types of intermediate images; the multiple optical elements can be configured as described in any embodiment provided in this application.

[0236] If the first program is used to determine the identification information, then multiple processing modules can be used to output multiple sets of identification information corresponding to the target object based on different optical elements; the multiple sets of identification information can represent the same or different features of the target object.

[0237] In one embodiment, the optical system includes multiple optical elements, which are coupled to different processing modules. The different processing modules execute different programs to obtain different intermediate data.

[0238] In this embodiment, the first data can be input into multiple optical elements, and different processing modules can receive the output of the corresponding optical elements and execute different programs to obtain multiple sets of intermediate data based on the received output of the optical elements. The multiple optical elements can be positioned at different locations on the target object; the multiple optical elements can be used to generate different types of intermediate images; the multiple optical elements can be configured as in any embodiment provided in this application.

[0239] Different programs can be used to determine different information. Different programs can be for the same target object or for different target objects; different programs can determine various types of information results such as identification information and encryption information for a target object, or they can execute different programs for a target object to determine the same type of information results, such as determining identification information; for example, the identification information includes posture recognition information.

[0240] Step S3: Based on the evaluation value, update the configuration of the optical components, or update the configuration of the first program, or update both the configuration of the optical components and the configuration of the first program, until the evaluation value meets the preset numerical conditions, and obtain the configuration information of the optical system.

[0241] In one embodiment, the optical element includes at least one metasurface element.

[0242] In one embodiment, the optical element includes nanostructure units for forming a metasurface. The nanostructure units are disposed on the light-incident side, the light-exit side, or simultaneously on both sides of the optical element body.

[0243] The configuration method for the optical system includes: updating the configuration of the metasurface element, or updating the configuration of the first program, or updating both the configuration of the metasurface element and the configuration of the first program, based on the evaluation value, until the evaluation value meets a preset numerical condition, thereby obtaining the configuration information of the optical system. This step can replace step S3.

[0244] In one embodiment, the optical element includes a metasurface element; updating the configuration of the optical element includes updating the configuration of the metasurface element; specifically, it includes updating the configuration of the nanostructure unit.

[0245] The configuration of metasurface elements updated based on evaluation values ​​may specifically include at least one of the geometry, size, or spatial arrangement of the metasurface nanostructure units. In some embodiments, the configuration of metasurface elements may also include at least one of the arrangement period, material, shape, and positional coordinates of the nanostructure units.

[0246] In one embodiment, the optical element includes at least one of a refractive optical element, a diffractive optical element, and a scattering medium element; one or more refractive optical elements may be provided; one or more diffractive optical elements may be provided; one or more scattering medium elements may be provided.

[0247] The configuration method of the optical system includes: updating the configuration of the optical element based on the evaluation value, or updating the configuration of the first program, or updating both the configuration of the optical element and the configuration of the first program, until the evaluation value meets a preset numerical condition, thereby obtaining the configuration information of the optical system.

[0248] In one embodiment, the optical element includes a refractive optical element. In this embodiment, updating the configuration of the optical element includes updating at least one of the following: the material, surface curvature, thickness, and surface shape of the refractive optical element.

[0249] For spherical refractive optical elements, at least one of the following can be updated: the material, surface curvature, and glass thickness.

[0250] For aspherical refractive optical elements, in addition to the aforementioned materials, surface curvature, and glass thickness, the surface shape can also be updated.

[0251] In one specific embodiment, the configuration of optical elements is updated based on evaluation values, and the updated configuration of refractive optical elements may specifically include refractive index, radius of curvature, etc.

[0252] In one embodiment, the optical element includes a diffractive optical element. In this embodiment, updating the configuration of the optical element includes updating at least one of the following: the diffraction order of the diffractive optical element, the period of the microstructure, the height of the microstructure, and the refractive index.

[0253] In one specific embodiment, the configuration of the optical element is updated based on the evaluation value. The updated configuration of the diffractive optical element may specifically include: focal length characteristics, phase function of the diffraction surface, radial radius at the abrupt change of each annulus of the diffraction surface, annulus depth of the diffraction surface, and diffraction efficiency, etc.

[0254] In one embodiment, the optical element includes a scattering medium element. In this embodiment, updating the configuration of the optical element includes updating at least one of the following: surface roughness, particle distribution uniformity, depth of abrasion or acid etching, and thickness of the diffractive optical element.

[0255] In one specific embodiment, the configuration of the optical element is updated based on the evaluation value. The updated configuration of the scattering medium element may specifically include: scattering coefficient, absorption coefficient, phase function, scattering length, opacity or transmittance, anisotropy factor, optical thickness, scattering efficiency, and particle size distribution.

[0256] In one embodiment, a metasurface element, a photosensitive element having a metasurface element, or both a metasurface element and a photosensitive element are used to constitute a data acquisition module. The configuration of the data acquisition module can be updated based on the evaluation value. The updated configuration of the data acquisition module may specifically include the combination method of the elements, the spacing between each element, etc. For example, updating the spacing between metasurface elements, or updating the spacing between focusing lenses, or updating the spacing between focusing lenses and metasurface elements.

[0257] In one embodiment, the configuration of the first program updated based on the evaluation value may specifically include program instructions or parameters. In another embodiment, when executed, the first program is used to implement a first neural network model algorithm. The configuration of the first program includes model parameters of the first neural network model. The configuration of the first program updated based on the evaluation value may specifically include model parameters of the first neural network model.

[0258] When the optical elements include multiple components, the configuration method provided in this application can be implemented to perform unified configuration of the optical system.

[0259] In one embodiment, the optical system includes a first optical element.

[0260] In one embodiment, the first optical element includes a first metasurface element. In another embodiment, the first optical element includes at least one of a refractive optical element, a diffractive optical element, and a scattering medium element; one or more refractive optical elements may be provided; one or more diffractive optical elements may be provided; one or more scattering medium elements may be provided.

[0261] In one embodiment, the optical system includes a second optical element.

[0262] In one embodiment, the second optical element includes a second metasurface element. In another embodiment, the second optical element includes at least one of a refractive optical element, a diffractive optical element, and a scattering medium element; one or more refractive optical elements may be provided; one or more diffractive optical elements may be provided; one or more scattering medium elements may be provided.

[0263] In one embodiment, the processing module is coupled to the first optical element and the second optical element respectively. In another embodiment, the processing module is used to process data information from the first optical element and the second optical element according to a first program.

[0264] In one embodiment, the first optical element and the second optical element are positioned at different locations. This allows the optical system to collect information about the target object from different angles, jointly determining the data processing result and improving accuracy.

[0265] In one embodiment, the optical system is used to determine the pose recognition information of a target object. When the optical system includes multiple optical elements positioned at different locations, the processing module can determine the pose recognition information by combining the poses from different locations based on intermediate images acquired at different locations.

[0266] In one embodiment, the first optical element is positioned relative to the target object in a first orientation, which is one of the following: directly in front of the target object, to the left of the target object, to the right of the target object, to the bottom of the target object, or to the top of the target object. In another embodiment, the second optical element is positioned relative to the target object in a second orientation, which is another of the following: directly in front of the target object, to the left of the target object, to the right of the target object, to the bottom of the target object, or to the top of the target object.

[0267] In some embodiments, the optical system may also include additional optical elements located in other orientations and coupled to the processing module.

[0268] In one embodiment, the first optical element and the second optical element are used to generate different intermediate images.

[0269] The optical system can be configured with a first processing module and a second processing module corresponding to the first and second optical elements, constructing a multi-parallel architecture. The first and second optical elements can also be coupled to the same processing module for integrated processing.

[0270] In one specific embodiment, a first optical element is used to generate a first intermediate image; the first intermediate image is one of a sharp image, a blurred image, and a contour image; a second optical element is used to generate a second intermediate image; the second intermediate image is another of a sharp image, a blurred image, and a contour image. In this specific embodiment, different optical elements are used to generate different types of intermediate images. In some other embodiments, the optical system may further include additional optical elements for generating other types of intermediate images.

[0271] In one specific embodiment, a first optical element is used to generate a first intermediate image; the first intermediate image has a first degree of blur; a second optical element is used to generate a second intermediate image; the second intermediate image has a second degree of blur. In this specific embodiment, different optical elements are used to generate intermediate images with different degrees of blur. In some other embodiments, the optical system may further include additional optical elements for generating intermediate images with other degrees of blur.

[0272] In any of the above embodiments that include multiple optical elements, the configuration method of the optical system may include the steps of: updating the configuration of the first optical element, or updating the configuration of the second optical element, or updating the configuration of the first program, or updating both the configuration of the first optical element and the configuration of the first program, or updating both the configuration of the second optical element and the configuration of the first program, or updating all three of the configurations of the first optical element, the second optical element, and the first program, until the evaluation value meets a preset numerical condition, thereby obtaining the configuration information of the optical system.

[0273] This step can be replaced by the aforementioned step S3. This step can be performed after the aforementioned step S2.

[0274] Thus, when dealing with an optical system containing multiple optical components, training can be performed on any part of the optical system, or the entire optical system can be trained uniformly.

[0275] Figure 8 illustrates some steps of a configuration method according to a specific embodiment of this application. In this specific embodiment, the configuration method includes the following steps.

[0276] Step S211: Input the first data into the optical element to generate an intermediate image.

[0277] Step S212: Determine whether the intermediate image meets the first preset condition.

[0278] If so, proceed to step S213A to execute the first program to process the intermediate image and obtain intermediate data.

[0279] If not, proceed to step S213B, update the configuration of the optical elements until the regenerated intermediate image meets the first preset condition, execute the first program to process the regenerated intermediate image, and obtain intermediate data.

[0280] This allows for the pre-judgment of intermediate images and the individual updating of optical component configurations when necessary, facilitating rapid optical system configuration and reducing the number of iterations in the configuration process. Because the intermediate images are pre-judged, it ensures that the optical components generate the desired intermediate images, resulting in a high degree of alignment between the optical system and the configuration target.

[0281] In one embodiment, the optical element includes at least one metasurface element; the configuration of the updated optical element may be the configuration of the metasurface element, specifically including at least one of the geometry, size, or spatial arrangement of the nanostructure units constituting the metasurface.

[0282] In some embodiments, the optical element includes at least one of a refractive optical element, a diffractive optical element, and a scattering medium element; one or more refractive optical elements may be provided; one or more diffractive optical elements may be provided; one or more scattering medium elements may be provided; the updated configuration of the optical element may be a configuration of at least one of a refractive optical element, a diffractive optical element, and a scattering medium element.

[0283] The intermediate image can be an image obtained by acquiring the light information actually generated by the optical element, or it can be an image obtained by adjusting the light corresponding to the first data (e.g., sample image) through the optical element based on a light simulation platform to generate new light, and simulating the new light.

[0284] In one embodiment, the first preset condition is a condition related to the intermediate image; the first preset condition may be related to the nature, content, size, and display of the intermediate image.

[0285] In one specific embodiment, the first preset condition is related to the type of the intermediate image. The intermediate image can be at least one of a sharp image, a blurry image, and a contour image. When the intermediate image is a sharp image, the intermediate image satisfies the first preset condition; when the intermediate image includes a sharp image, the intermediate image satisfies the first preset condition. When the intermediate image is a blurry image, the intermediate image satisfies the first preset condition; when the intermediate image includes a blurry image, the intermediate image satisfies the first preset condition. When the intermediate image is a contour image, the intermediate image satisfies the first preset condition; when the intermediate image includes a contour image, the intermediate image satisfies the first preset condition.

[0286] The type of intermediate image set by the preset conditions can be different for different optical elements or processing modules. For example, for the first optical element, it is required to generate a contour image, so the first preset condition corresponding to the first optical element can be that the intermediate image is a contour image, or that the intermediate image includes a contour image. Based on this, the first preset condition can also be: the type of the intermediate image conforms to the expected goal corresponding to the optical element or processing module; the expected goal includes the expected type goal.

[0287] In one specific embodiment, the first preset condition is related to the display effect of the intermediate image. The display effect of the intermediate image can be at least one of sharpness, blur level, and number of color channels. The first preset condition can be defined as the intermediate image being an image with a first degree of blur, or it can be defined as the intermediate image including an image with a first degree of blur; the first degree of blur can be quantized according to the point spread function.

[0288] The intermediate image display effect set by the preset conditions can be different for different optical elements or processing modules. For example, for a first optical element, it is required to generate an image with a first degree of blur. Therefore, the first preset condition corresponding to the first optical element can be that the intermediate image is an image with the first degree of blur, or that the intermediate image includes an image with the first degree of blur. Based on this, the first preset condition can also be: the type of the intermediate image conforms to the expected goal corresponding to the optical element or processing module; the expected goal includes the expected display goal.

[0289] In some embodiments, the expected target for the intermediate image can also be an expected size target, an expected content target, etc.

[0290] Steps S211 to S213B described above can be implemented as a separate configuration scheme to achieve separate verification and configuration of optical components.

[0291] Steps S211 to S213B described above can be implemented as part of step S2. Specifically, after determining intermediate data through steps S211 to S213A or S211 to S213B, this intermediate data can be used as the intermediate data in step S2 to continue executing the step of "determining the evaluation value based on the intermediate data and the second data".

[0292] In one embodiment, the first preset condition may be a condition related to the contour. When the intermediate image meets the first preset condition, the intermediate image is considered to be a contour image. When the intermediate image meets the first preset condition, the intermediate image is considered to include edge information.

[0293] In this embodiment, the configuration method of the optical system includes the following steps.

[0294] Step S2121: Determine whether the intermediate image includes edge information based on the changes in pixel channel values.

[0295] If so, proceed to step S2131A to determine that the intermediate image satisfies the first preset condition regarding the contour.

[0296] The change in pixel channel values ​​at the edge positions in the intermediate image is greater than a preset change value. In the intermediate image, the change in pixel channel values ​​at the edge positions is greater than a preset change value.

[0297] The pixel channel value can be the value of a specific pixel channel of a pixel in the intermediate image. For example, when the intermediate image is a grayscale image, the pixels in the intermediate image have a single pixel channel, and the pixel channel value refers to the grayscale value. As another example, when the intermediate image is another type of image, the pixel channel value of the pixels in the image can refer to the brightness value.

[0298] In one specific embodiment, the pixel channel value refers to the brightness value. When it is determined that the intermediate image includes an area with obvious brightness changes, the change value of the brightness value is greater than a preset change value. It can be determined that the area contains edge information and that the area is an edge position. Therefore, it is determined that the intermediate image meets the preset conditions about the contour and that the intermediate image is a contour image.

[0299] Steps S2121 to S2131A described above can be implemented as a separate configuration scheme to achieve separate verification and configuration of optical components.

[0300] Step S2121 can be included in step S212; step S2131A can be included in step S213A. Specifically, edge information can be included as the first predetermined condition; specifically, when it is determined that the intermediate image meets the first preset condition regarding the contour, it can be considered that the intermediate image meets the first preset condition, thereby implementing the step of "executing the first program to process the intermediate image and obtain intermediate data".

[0301] When performing the above-mentioned further configurations or combinations of technical features, steps not mentioned can still be implemented in the original order. For example, step S213B can still be implemented after determining that the intermediate image does not meet the first preset condition.

[0302] In one embodiment, determining whether an intermediate image includes edge information can be achieved using an edge detection operator.

[0303] In this embodiment, the configuration method of the optical system includes the steps of: identifying points in the intermediate image where the change in pixel channel value is greater than a preset change value by using an edge detection operator, and determining whether the intermediate image includes edge information.

[0304] In one specific embodiment, the edge detection operator includes at least one of the Sobel operator, the Prewitt operator, or the Canny operator.

[0305] In one embodiment, the first preset condition may be a condition related to encryption. When an intermediate image meets the first preset condition, the intermediate image is considered to be an encrypted image.

[0306] In one embodiment, the first preset condition may be a condition concerning blurred display. When the intermediate image meets the first preset condition, the intermediate image is considered to be a blurred image.

[0307] In one embodiment, the first preset condition may be a condition regarding clear display. When an intermediate image meets the first preset condition, the intermediate image is considered a clear image.

[0308] In one embodiment, the method for configuring the optical system includes the following steps.

[0309] Step S2122: Determine whether the blur value of the intermediate image is greater than the preset blur value.

[0310] If so, proceed to step S2132A to determine that the intermediate image meets the first preset condition for encryption.

[0311] Thus, this embodiment achieves image encryption by adjusting the degree of image blur.

[0312] The ambiguity value can be determined based on the point spread function value.

[0313] In one embodiment, the blur value of the intermediate image can be defined by calculating the difference between the dot spread function value of the optical element corresponding to the intermediate image and the dot spread function value of the optical element capable of generating a clear image. A preset blur value related to this difference can be set for the current optical element, and specifically, by comparing the numerical relationship between the blur value of the intermediate image and the preset blur value, it can be determined whether the intermediate image meets a first preset condition regarding encryption.

[0314] In one embodiment, the blur value of the intermediate image can be defined by calculating the dot spread function of the optical element corresponding to the intermediate image. For the current optical element, a preset function value related to the dot spread function can be set as the preset blur value, and specifically, by comparing the dot spread function value and the preset function value (that is, comparing the blur value of the current intermediate image with the preset blur value), it is determined whether the intermediate image meets the first preset condition for encryption.

[0315] The blur value can be determined according to a digital image blur algorithm based on grayscale values.

[0316] In one embodiment, the configuration method of the optical system further includes the steps of: determining a first pixel channel value of an intermediate image, performing a convolution operation on the first pixel channel value using a Laplacian mask, and determining the blur value of the intermediate image.

[0317] This step is set before the step in the configuration method that utilizes the ambiguity value. For example, it can be set before step S2122.

[0318] In one specific embodiment, the first pixel channel value can be a grayscale value.

[0319] Figure 9 illustrates some steps of a configuration method according to a specific embodiment of this application. In this specific embodiment, the configuration method includes the following steps.

[0320] Step S221: Input the first data into the optical element and determine the point spread function value of the optical element in response to the first data.

[0321] Step S222: Determine whether the difference between the point spread function value and the standard function value meets the preset threshold condition.

[0322] If so, proceed to step S223A to execute the first program to process the intermediate image generated by the optical element corresponding to the first data, and obtain the intermediate data.

[0323] If not, proceed to step S223B, update the configuration of the optical element until the newly determined point spread function value meets the preset threshold condition, execute the first program to process the intermediate image regenerated by the optical element, and obtain intermediate data.

[0324] In this way, the properties of the optical signal transmission at the optical element can be judged and verified during the optical signal processing stage, determining whether the optical signal matches the expected light field distribution effect. When combining with the aforementioned technical solution to perform unified configuration of optical elements and the first program, the optical elements are first configured based on the point spread function to meet the requirements of the light field distribution, and then the evaluation value is determined and the optical elements and the first program are configured in a unified manner. This helps to improve the convergence efficiency and system configuration speed, and can also ensure that the intermediate information (e.g., intermediate images) generated by the optical elements also meet expectations.

[0325] In one embodiment, the optical element includes at least one metasurface element; the configuration of the updated optical element may be the configuration of the metasurface element, specifically including at least one of the geometry, size, or spatial arrangement of the nanostructure units constituting the metasurface.

[0326] In some embodiments, the optical element includes at least one of a refractive optical element, a diffractive optical element, and a scattering medium element; one or more refractive optical elements may be provided; one or more diffractive optical elements may be provided; one or more scattering medium elements may be provided; the updated configuration of the optical element may be a configuration of at least one of a refractive optical element, a diffractive optical element, and a scattering medium element.

[0327] In one embodiment, the standard function value and the preset threshold condition are determined based on the expected light field distribution. In a specific embodiment, the standard function value and the preset threshold condition are determined based on a preset blur level requirement for the intermediate image.

[0328] In one embodiment, a standard function value can be used to characterize a preset ambiguity requirement. In another embodiment, a preset threshold condition is used to characterize a preset allowable error range.

[0329] In this embodiment, the difference between the point spread function value corresponding to the current optical element and the standard function value can be directly compared. When the difference meets a preset threshold condition, it indicates that the point spread function of the current optical element is within the allowable error range, and the steps of generating intermediate images and obtaining intermediate data can continue. When the difference does not meet the preset threshold condition, it indicates that the point spread function of the current optical element does not meet the expected requirements, and the configuration of the optical element can be updated. That is, the optical element can be configured first based on the point spread function, and then the optical element can be configured based on the intermediate data and the second data, and the first program can be configured, or both the optical element and the first program can be configured.

[0330] In one embodiment, the standard function value can be used to represent the point spread function value corresponding to the optical element achieving clear imaging. In another embodiment, a preset threshold condition is used to characterize a preset blur level requirement.

[0331] In this embodiment, the difference between the point spread function value and the standard function value corresponding to the current optical element can be determined first. This difference is used to characterize the actual blurring degree corresponding to the current optical element.

[0332] When the difference meets the preset threshold condition, it means that the actual blur level of the corresponding current optical element meets the preset blur level requirement, and the steps of generating intermediate images and obtaining intermediate data can continue.

[0333] When the difference does not meet the preset threshold condition, it indicates that the actual blur degree of the corresponding current optical element does not meet the preset blur degree requirement. The configuration of the optical element can be updated. That is, the optical element is configured first based on the difference between the point spread function and the standard function value, and then the optical element is configured based on the intermediate data and the second data, the first program is configured, or both the optical element and the first program are configured.

[0334] In one embodiment, the difference between the point spread function value and the standard function value can be determined by the L2 norm, thereby determining whether the light field distribution meets expectations or whether the degree of blurring meets preset requirements.

[0335] In this embodiment, the configuration method of the optical system includes the following steps.

[0336] Step S2221: Determine the L2 norm of the difference between the point spread function value and the standard function value.

[0337] Step S2222: Determine whether the L2 norm falls within the fuzzy threshold range.

[0338] If so, proceed to step S2231A to determine if the point spread function value meets the preset threshold condition.

[0339] If not, proceed to step S2231B to determine if the point diffusion value does not meet the preset threshold condition.

[0340] The blur threshold range is determined based on the blur level requirements set for the optical element.

[0341] In this way, by taking advantage of the positive correlation between the degree of ambiguity and the L2 norm, the actual degree of ambiguity of the current optical element can be determined by judging the numerical relationship between the L2 norm and the ambiguity threshold range.

[0342] In one embodiment, the point spread function value corresponding to the optical element can be calculated and determined based on the scalar diffraction formula. In another embodiment, the standard function value can be calculated and determined based on the scalar diffraction formula.

[0343] In other embodiments, the point spread function value or standard function value can also be determined by at least one of the following calculations:

[0344] (1) The wavefront aberration function based on geometric aberration is determined by wavefront expansion calculation;

[0345] (2) Treat the optical system as a linear system and use Fourier transform to calculate and determine it;

[0346] (3) The propagation of electromagnetic waves is simulated and calculated by solving Maxwell's equations and using numerical methods in the time domain;

[0347] (4) When the boundary conditions are clear, they are determined by calculating the boundary propagation of the electromagnetic field.

[0348] (5) Use the Zanick polynomial to fit the wavefront distortion, and then calculate and determine it based on this.

[0349] (6) The path of light is calculated and determined by simulating the propagation path of light through a large number of random samples of light rays or wavefronts using the Monte Carlo method.

[0350] In one embodiment, the standard function value can be used to characterize a preset fuzziness level requirement.

[0351] In one specific embodiment, the standard function value can be determined by performing the following steps: calculating the standard function value based on the scalar diffraction formula according to the blurring requirements corresponding to the optical element.

[0352] In one specific embodiment, the standard function value can be determined by performing the following steps: obtaining a standard element that meets the blurring requirements corresponding to the optical element, using first data as input to the standard element, determining the point spread function value of the standard element in response to the first data, and determining the standard function value based on the point spread function value.

[0353] In one embodiment, standard function values ​​can be used to characterize the performance of optical elements when performing sharp imaging.

[0354] In one specific embodiment, the standard function value is calculated and determined based on the scalar diffraction formula according to the sharpness requirements corresponding to the optical element.

[0355] In one specific embodiment, a standard element that meets the sharpness requirements corresponding to an optical element is obtained. First data is used as the input of the standard element to determine the point spread function value of the standard element in response to the first data, and a standard function value is determined based on the point spread function value.

[0356] As shown in Figure 10, one embodiment of this application provides an identification method.

[0357] This method can be implemented by an optical system to achieve the identification of the target object.

[0358] This method can be implemented in the optical system in the form of computer instructions, control steps, or operation steps. For example, configuring the optical system according to configuration information can be considered an operation step, controlling the optical elements to obtain the detection image can be considered an operation step, controlling the processing module to process the detection image can be considered a control step or computer instruction, and determining the detection information can be considered a control step or computer instruction.

[0359] The optical system can be configured as described in any of the embodiments above. For example, the optical system includes optical elements.

[0360] As shown in Figure 2, the optical element may include at least one of the following: metasurface element, refractive optical element, diffractive optical element, and scattering medium element; one or more refractive optical elements may be provided; one or more diffractive optical elements may be provided; one or more scattering medium elements may be provided.

[0361] The optical system also includes a photosensitive element 13.

[0362] In one embodiment, the optical element includes at least one metasurface element 11.

[0363] In one embodiment, the metasurface element 11 is disposed on the light-incident side of the photosensitive element 13.

[0364] In one embodiment, the optical element is used to construct a data acquisition module 10, which includes a photosensitive element 13 and a metasurface element 11; in a specific embodiment, the data acquisition module 10 may further include a focusing lens 14.

[0365] In one specific embodiment, the optical element or acquisition module 10 is used to generate an intermediate image based on the information input thereto. For example, the intermediate image may be a posture image; the posture image is an image carrying the posture features of the target object; the posture features may be the contour features of the target object, or the node features indicating the joints of the target object, or the connecting line features indicating the limbs of the target object.

[0366] The optical system includes a processing module 12. The processing module 12 is coupled to an optical element. The processing module 12 is used to process data information from the optical element according to a first program. In one embodiment, the optical element includes a metasurface element 11, and the processing module 12 is coupled to the metasurface element 11 to process data information from the metasurface element 11 according to the first program. In a specific embodiment, the optical system includes a photosensitive element 13, the metasurface element 11 is coupled to the photosensitive element 13, and the processing module 12 is coupled to the metasurface element 11 through the photosensitive element 13.

[0367] In one embodiment, the first program implements a neural network model when executed by the processing module. In a specific embodiment, the neural network model is a convolutional neural network model.

[0368] In one specific embodiment, when the first program is executed by the processing module, it is used to identify the target object based on the information of the input optical element; for example, to determine the posture recognition information of the corresponding target object.

[0369] If the optical element is used to generate an attitude image as an intermediate image, then the first program, when executed by the processing module, is used to determine the attitude recognition information of the corresponding target object based on the attitude image.

[0370] The identification method specifically includes at least one of the following steps.

[0371] Step P1: Configure the optical system based on the configuration information obtained by a configuration method.

[0372] In one embodiment, the configuration method is implemented according to the optical system configuration method provided by any of the above technical solutions. The configured optical system has more accurate and unified signal processing capabilities.

[0373] Step P2: Control the optical elements to obtain the detection image of the corresponding target object, and control the processing module to execute the first program to process the detection image and identify the target object.

[0374] The recognition method can be used to identify at least one of the following: the target object's limb posture, gesture, object position, object posture, and gender. Related technical solutions can be found in the preceding description of the application scenarios of the optical system.

[0375] In one embodiment, the recognition method is used to determine the posture recognition information of a target object, thereby achieving the recognition of the target object. In this embodiment, the recognition method can be used to recognize limb postures, gestures, object positions, object postures, etc.

[0376] Figure 11(a) shows a target object in one embodiment, and Figure 11(b) shows a detection image corresponding to the target object. In this embodiment, the detection image can be a contour image; the detection image includes edge information. When the change value of the pixel channel value of a pixel in the detection image is greater than a preset change value, the location of the pixel is determined to be an edge location; the detection image includes edge information; the detection image is a contour image used to show the contour of the target object.

[0377] The pixel channel value can be a brightness value.

[0378] The changes in the pixel channel values ​​can be obtained using an edge detection operator. The edge detection operator includes at least one of the Sobel operator, the Prewitt operator, or the Canny operator.

[0379] Figure 12 shows a detection image in one embodiment. In this embodiment, the detection image may be a pose image including nodes indicating joints of the target object; the detection image may be a pose image including connecting lines indicating limbs of the target object.

[0380] In one embodiment, the head and torso of the target object can be marked with a first color, the left limb of the target object can be marked with a second color, and the right limb of the target object can be marked with a third color.

[0381] In one embodiment, the nodes and connecting lines in FIG12 may be the posture recognition information, and the blurred image underneath FIG12 may be the detection image. In this embodiment, optical elements are used to generate the blurred image, and a first program is used to generate information such as nodes and connecting lines indicating limb movements and limb distribution based on the blurred image.

[0382] In one embodiment, the recognition method provided by this application further includes: outputting and displaying the posture recognition information. For example, in the embodiment provided in FIG12, when the posture recognition information is the node or connecting line, the node or connecting line can be output and displayed on the detection image.

[0383] Figure 13 shows a detection image in one embodiment. In this embodiment, the detection image can be a blurred image. The third target object in this embodiment has a normal sitting posture; the first program can determine that the target object's sitting posture is a normal sitting posture based on the blurred image; the information corresponding to the normal sitting posture can be the posture recognition information.

[0384] Figure 14 shows a detection image in one embodiment. In this embodiment, the detection image may be a blurred image. Corresponding to the fourth target object in this embodiment, it has an abnormal sitting posture with tilted shoulders; the first program can determine that the target object's sitting posture is abnormal based on the blurred image; the information corresponding to the abnormal sitting posture may be the posture recognition information.

[0385] Figure 15 illustrates a detection image in one embodiment. In this embodiment, the detection image can be a blurred image. Corresponding to the target object in Figure 15, it has a human upper body and an outstretched hand; the first program can determine the position of the human body and its hand based on the blurred image; for example, the movement of the human arm can be analyzed accordingly.

[0386] Figure 16 illustrates a detection image in one embodiment. In this embodiment, the detection image may be a blurred image. Corresponding to the target object in Figure 16, there is a hand showing a first gesture; a first program can determine the posture of the hand based on the blurred image; for example, information corresponding to the first gesture can be analyzed accordingly, where the first gesture is used to represent the number six.

[0387] Figure 17 illustrates a detection image in one embodiment. In this embodiment, the detection image may be a blurred image. Corresponding to the target object in Figure 17, there is a hand showing a second gesture; a first program can determine the posture of the hand based on the blurred image; for example, information corresponding to the second gesture can be analyzed accordingly, where the second gesture is used to indicate agreement (ok).

[0388] Figure 18 shows a detection image in one embodiment. In this embodiment, the detection image may be a blurred image. Corresponding to the target object in Figure 18, there is a hand showing a third gesture; a first program can determine the posture of the hand based on the blurred image; for example, information corresponding to the third gesture can be analyzed accordingly, where the third gesture is used to represent the palm shape.

[0389] Figure 19 illustrates a detection image in one embodiment. In this embodiment, the detection image may be a blurred image. Corresponding to the target object in Figure 19, there is a hand showing a fourth gesture; a first program can determine the posture of the hand based on the blurred image; for example, information corresponding to the fourth gesture can be analyzed accordingly, where the fourth gesture is used to represent a fist shape.

[0390] The output display step can be set after obtaining the posture recognition information in step P2, or it can be set after step P2 as step P3.

[0391] In one embodiment, the identification method provided in this application further includes: determining the difference between posture recognition information and standard posture, and outputting a comparison result or posture correction instruction. For example, in an embodiment corresponding to a third target object and a fourth target object, the target object with an abnormal sitting posture can be instructed to adjust its shoulder posture by comparing the difference between normal sitting posture and abnormal sitting posture.

[0392] The step of outputting the comparison result or attitude correction indication can be set in step P2 or after step P2.

[0393] In summary, the present application provides a method to determine an evaluation value for assessing the performance of an optical system by comprehensively considering the second data, which is the processing target corresponding to the first data in executing the first program, and the intermediate data, which is the current actual processing result corresponding to the first data in executing the first program. This allows for adjustments to the configuration of the optical system. Therefore, it is possible to establish a relationship between the actual situation and the target based on the first data, enabling the optical system to quickly update its configuration to achieve the desired target. Furthermore, since the first data serves as the input to the optical components in the optical system and the intermediate data serves as the output of the processing module in the optical system, the optical components and the processing module as a whole can be used as the object of configuration updates, preventing the accumulation of errors in both.

[0394] It should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This way of describing the specification is only for clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

[0395] The detailed descriptions listed above are merely specific descriptions of feasible implementation methods of this application and are not intended to limit the scope of protection of this application. All equivalent implementation methods or modifications made without departing from the spirit of the art of this application should be included within the scope of protection of this application.

Claims

1. A method of configuring an optical system, characterized by, The optical system comprises: an optical element; a processing module coupled to the optical element, configured to process data information from the optical element according to a first program; The configuration method comprises: obtaining sample data, including first data corresponding to input information of the optical element, and second data corresponding to information output when the first program is executed; the second data is target data corresponding to the first data; inputting the first data into the optical element, executing the first program to obtain intermediate data, and determining an evaluation value according to the intermediate data and the second data; based on the evaluation value, updating the configuration of the optical element, or updating the configuration of the first program, or updating both the configuration of the optical element and the configuration of the first program, until the evaluation value meets a preset numerical condition, to obtain the configuration information of the optical system.

2. The configuration method of claim 1, wherein, The configuration method comprises at least one of the following: The first data comprises a sample image, and the second data comprises actual recognition information corresponding to the sample image, the actual recognition information being represented by actual features of a target object in the sample image; The evaluation value is represented by the difference between the intermediate data and the second data; The intermediate data comprises predicted recognition information corresponding to the sample image; the predicted recognition information is represented by features determined by the first program according to the sample image.

3. The configuration method of claim 1, wherein, The configuration method comprises: inputting the first data into the optical element to generate an intermediate image; determining whether the intermediate image meets a first preset condition; if yes, executing the first program to process the intermediate image to obtain intermediate data; if no, updating the configuration of the optical element until the newly generated intermediate image meets the first preset condition, and executing the first program to process the newly generated intermediate image to obtain intermediate data.

4. The configuration method of claim 3, wherein, The configuration method comprises: determining whether the intermediate image comprises edge information according to the change of pixel channel values; the change value of the pixel channel values at the edge position in the intermediate image is greater than a preset change value; if yes, it is determined that the intermediate image meets the first preset condition about the contour.

5. The configuration method of claim 4, wherein, The configuration method comprises: identifying points in the intermediate image where the change of pixel channel values is greater than a preset change value by using an edge detection operator, and determining whether the intermediate image comprises edge information; the edge detection operator comprises at least one of a Sobel operator, a Prewitt operator or a Canny operator.

6. The configuration method of claim 3, wherein, The configuration method comprises: determining whether the blur value of the intermediate image is greater than a preset blur value, or determining whether the difference between the blur value of the intermediate image and the preset blur value meets a preset error condition; if yes, it is determined that the intermediate image meets the first preset condition about encryption.

7. The configuration method of claim 6, wherein, The configuration method comprises: determining a first pixel channel value of the intermediate image, and performing convolution operation on the first pixel channel value by using a Laplacian mask to determine the blur value of the intermediate image.

8. The configuration method of claim 1, wherein, The configuration method comprises: inputting the first data into the optical element to determine a point spread function value of the optical element in response to the first data; determining whether the difference between the point spread function value and a standard function value meets a preset threshold condition; if yes, executing the first program to process an intermediate image generated by the optical element corresponding to the first data to obtain intermediate data; If not, the configuration of the optical element is updated until the re-determined point spread function value meets the preset threshold condition, and the first program is executed to process the intermediate image re-generated by the optical element to obtain intermediate data.

9. The configuration method of claim 8, wherein, The configuration method comprises: determining an L2 norm of a difference between the point spread function value and a standard function value; judging whether the L2 norm falls within a blur threshold interval; the blur threshold interval is determined according to a blur degree requirement set for the optical element; if yes, determining that the point spread function value meets the preset threshold condition; if not, determining that the point spread function value does not meet the preset threshold condition.

10. The configuration method of claim 8, wherein, The standard function value is determined according to at least one of the following: determining the standard function value based on a scalar diffraction formula calculation according to a blur degree requirement corresponding to the optical element; determining the standard function value based on a scalar diffraction formula calculation according to a sharpness requirement corresponding to the optical element; obtaining a standard element meeting the blur degree requirement corresponding to the optical element, taking the first data as the input of the standard element, determining a point spread function value of the standard element in response to the first data, and determining the standard function value according to the point spread function value; obtaining a standard element meeting the sharpness requirement corresponding to the optical element, taking the first data as the input of the standard element, determining a point spread function value of the standard element in response to the first data, and determining the standard function value according to the point spread function value.

11. The configuration method of claim 1, wherein, The optical system comprises: a first optical element; a second optical element; The processing module is coupled to the first optical element and the second optical element respectively, and is configured to process data information from the first optical element and the second optical element according to the first program; The configuration method comprises: updating the configuration of the first optical element, or updating the configuration of the second optical element, or updating the configuration of the first program, or updating both the configuration of the first optical element and the configuration of the first program, or updating both the configuration of the second optical element and the configuration of the first program, or updating the configuration of the first optical element, the configuration of the second optical element and the configuration of the first program based on the evaluation value, until the evaluation value meets a preset numerical condition, to obtain configuration information of the optical system; The first optical element and the second optical element are arranged at different positions, or the first optical element and the second optical element are used to generate different intermediate images.

12. The configuration method of claim 1, wherein, The optical element comprises at least one super surface element; The configuration method comprises: updating the configuration of the super surface element, or updating the configuration of the first program, or updating both the configuration of the super surface element and the configuration of the first program based on the evaluation value, until the evaluation value meets a preset numerical condition, to obtain configuration information of the optical system; The configuration of the super surface element comprises at least one of the geometric structure, size or spatial arrangement of the nano structure unit used to constitute the super surface; The first program is used to implement a first neural network model algorithm when executed, and the configuration of the first program comprises model parameters of the first neural network model; The evaluation value comprises a loss value of the intermediate data and the second data.

13. The configuration method of claim 1, wherein, The configuration method comprises: updating the configuration of the optical element, or updating the configuration of the first program, or updating both the configuration of the optical element and the configuration of the first program, based on the evaluation value, until the evaluation value meets a preset numerical condition, to obtain configuration information of the optical system; The optical element and the configuration method are configured according to at least one of the following: The optical element comprises at least one refractive optical element, and the configuration of the optical element comprises at least one of a material, a surface curvature, a thickness, and a surface shape of the refractive optical element, The optical element comprises at least one scattering medium element, and the configuration of the optical element comprises at least one of a surface roughness, a particle distribution uniformity, a depth of sanding or acid etching, and a thickness of the scattering medium element, The optical element comprises at least one diffractive optical element, and the configuration of the diffractive optical element comprises at least one of a diffraction order, a period of a microstructure, a height of the microstructure, and a refractive index of the diffractive optical element; The first program, when executed, is used to implement a first neural network model algorithm, and the configuration of the first program comprises model parameters of the first neural network model; The evaluation value comprises intermediate data and a loss value of second data.

14. A method of identification, characterized by The optical system comprises: An optical element; A processing module coupled to the optical element and configured to process data information from the optical element according to a first program; The identification method comprises: The optical system is configured according to the configuration information obtained by the configuration method of any one of claims 1-13; The optical element is controlled to obtain a detection image of a corresponding target object, and the processing module is controlled to execute the first program to process the detection image, so as to identify the target object.

15. The identification method according to claim 14, characterized in that, The identification method is used to determine pose recognition information of a target object; and the identification method further comprises at least one of the following: Outputting the pose recognition information; Determining a difference between the pose recognition information and a standard pose, and outputting a comparison result or a pose correction instruction; The identification method is used to identify at least one of a limb pose, a gesture, an object position, and an object pose.

16. An optical system characterized by, The optical system is configured according to the configuration information obtained by the configuration method of any one of claims 1-13, or the optical system is used to implement the identification method of any one of claims 14-15.

17. The optical system of claim 16, wherein, The optical system comprises a photosensitive element; The optical element comprises a metasurface element arranged on an incident light side of the photosensitive element.

18. The optical system of claim 17, wherein: The incident light side of the photosensitive element is provided with a focusing lens, which is used at least to couple data information to the photosensitive element.

19. The optical system of claim 16, wherein, The optical system comprises: A first optical element; A first processing module coupled to the first optical element and configured to process data information from the first optical element according to a first program; A second optical element; A second processing module coupled to the second optical element and configured to process data information from the second optical element according to the first program; The first optical element is used to generate a first intermediate image corresponding to a target object, and the second optical element is used to generate a second intermediate image corresponding to the target object; The first intermediate image is one of a clear image, a blurred image, and a contour image, and the second intermediate image is another of a clear image, a blurred image, and a contour image; or, the first intermediate image has a first degree of blur, and the second intermediate image has a second degree of blur that is different from the first degree of blur.

20. The optical system of claim 16, wherein, The optical system includes: First optical element; Second optical element; The first optical element is disposed at a first position relative to the target object, and the second optical element is disposed at a second position relative to the target object; The first orientation is one of the following: directly in front of the target object, to the left of the target object, to the right of the target object, to the bottom of the target object, or to the top of the target object. The second orientation is the other of the following: directly in front of the target object, to the left of the target object, to the right of the target object, to the bottom of the target object, or to the top of the target object.

Images