A three-dimensional imaging system and method

Abstract

The three-dimensional imaging system and method provided by the present disclosure comprises a three-dimensional imaging light field to be measured, an imaging unit, a polarized photoelectric detection unit and a processing unit; a diffractive optical element or a superlens with the ability to independently control the imaging distance and the light transmission phase is selected to form the imaging unit, and the diffractive optical element and the superlens each contain at least one coding point spread function related to the imaging distance; the three-dimensional light field to be measured located at the object plane of the imaging unit is imaged to the polarized photoelectric detection unit located at the image plane of the imaging unit, and the images coded along the z direction are located at different positions of the polarized photoelectric detection unit; the polarized photoelectric detection unit is used to detect the light intensity of each linear polarization direction of the three-dimensional light field to be measured after the three-dimensional light field to be measured passes through the imaging unit; and the processing unit decodes the three-dimensional distribution information of the three-dimensional light field to be measured according to the captured polarized light intensity distribution. The present disclosure can realize a small-sized integrated and high-precision single-aperture three-dimensional imaging system.

Description

Technical Field

[0001] This disclosure can be applied to fields such as 3D imaging, consumer electronics, industrial inspection, and autonomous driving, and specifically relates to a method and system for 3D imaging of 3D light fields. Background Technology

[0002] Three-dimensional imaging technology is an important foundation for the development of technologies such as autonomous driving, manufacturing automation, industrial design, virtual reality, medical plastic surgery, cultural relic protection, and machine vision. How to quickly, accurately, and efficiently acquire the three-dimensional information of a target has become a current research hotspot.

[0003] Common 3D imaging techniques mainly include Time-of-Flight (TOF) and structured light methods with active illumination from modulated light sources, and passive imaging methods using binocular vision. Each of these methods has many advantages and is applicable to different application scenarios, but they also have certain limitations. Especially when processing targets with weak textures such as tiles, white walls, cars, and glass, the relatively simple surface texture information, and even their high reflectivity or high transmissivity, leads to poor performance from these methods. For example, the TOF method suffers from low resolution, missing information, and multipath interference; the structured light method is susceptible to environmental and glare interference; and the binocular vision method performs poorly on targets with weak textures, and because the accuracy and range of 3D imaging are related to the baseline length, it results in a large image size.

[0004] 3D reconstruction technology using polarization information holds promise for significantly improving the imaging capability, reconstruction resolution, and enriching detail information of targets with weak textures. Polarization-based 3D reconstruction technology solves for the polarization degree and polarization angle information of each pixel, establishes a mapping relationship between the polarization degree and polarization angle and the target shape normal vector using Fresnel's formula, and then achieves 3D reconstruction of the target through integration. It boasts advantages such as strong imaging capability for targets with weak textures, no need for active illumination, simple equipment, and rich reconstruction detail information. However, it suffers from drawbacks: polarization-based 3D reconstruction lacks absolute distance information, and the process of solving the target surface normal vector is affected by the multi-valued nature of trigonometric function solutions, leading to ambiguity in the 3D information. To address the multi-valued nature of normal vector solving and the lack of absolute distance in polarization-based 3D reconstruction technology, researchers have proposed polarization-based TOF, polarization-based structured light, and polarization-based binocular vision schemes, respectively. All of these schemes can eliminate the ambiguity of polarization-based 3D information and achieve high-precision 3D imaging of targets with weak textures. However, fusion schemes also have limitations, such as the need for active illumination for polarization-based TOF and polarization-based structured light, and the large size of polarization-based binocular systems, which restricts their further application. How to achieve a passive, compact, and highly capable 3D imaging system for targets with weak texture has become an urgent problem to be solved.

[0005] In recent years, metasurfaces have become a research hotspot in the field of optics due to their ability to flexibly control the phase, amplitude, and polarization of light, as well as their advantages such as small thickness and ease of fabrication. Based on metasurfaces, researchers have developed various lightweight integrated active and passive 3D imaging optical devices, such as metasurface TOF imaging, metasurface structured light imaging, metasurface light field imaging, and metasurface coded point spread function (PSF) imaging. These studies have verified that metasurfaces can effectively reduce system size and weight while improving performance. Among these, metasurface coded point spread function 3D imaging technology has broad application prospects in integration and miniaturization due to its advantages such as single aperture, high precision, and passive imaging. However, its drawbacks include difficulties in 3D calculation of weakly textured targets, limited polarization information, and challenges in polarization 3D reconstruction. Summary of the Invention

[0006] This disclosure aims to at least partially address one of the technical problems in the related art.

[0007] To this end, the first aspect of this disclosure proposes a miniaturized and integrated general-purpose three-dimensional imaging system for high-precision capture of three-dimensional information of light fields, including a three-dimensional light field to be measured, an imaging unit, a polarization photoelectric detection unit, and a processing unit arranged along the light propagation direction, and constructing a right-handed coordinate system composed of the x-axis, y-axis, and z-axis, wherein the light propagation direction is defined as the z-axis, and the plane in which the three-dimensional light field to be measured is located is defined as the xy plane;

[0008] The imaging unit is selected from diffractive optical elements or metalenses that have the ability to independently adjust the imaging distance and light transmission phase, and it is required that both the diffractive optical elements and the metalenses contain at least one coded point spread function related to the imaging distance.

[0009] The polarization photoelectric detection unit is used to detect the intensity of polarized light along different polarization channel directions on the image plane after the three-dimensional light field to be measured is imaged by the imaging unit.

[0010] The processing unit is used to decode the polarization intensity distribution detected by the polarization photoelectric detection unit to obtain the three-dimensional distribution information of the three-dimensional light field to be measured.

[0011] In some embodiments, the three-dimensional light field to be measured is set on the object plane of the imaging unit and is generated by a light beam illuminating the surface of the object.

[0012] In some embodiments, the diffractive optical element is an optical element made of a material with a single refractive index and containing at least two stepped surface engravings.

[0013] In some embodiments, the material having a single refractive index is selected from optical glass, optical plastic, or imprinting adhesive.

[0014] In some embodiments, the meta-lens consists of a substrate and optical antennas distributed in a two-dimensional array on the surface of the substrate. The cross-sectional shape of each optical antenna along the xy plane has C2 symmetry, the height of each optical antenna does not exceed one operating wavelength, and there is a refractive index difference between the materials used to make the optical antennas and the substrate.

[0015] In some embodiments, the refractive index difference between the materials used to make the optical antenna and the substrate should be greater than or equal to 0.5.

[0016] In some embodiments, the polarization photodetector unit is selected from a polarization camera or a photodetector with polarization beam splitting function. The polarization photodetector unit has at least three polarization channels with polarization states. When four linear polarization channels are used, they are linear polarization directions of 0°, 45°, 90° and 135° respectively.

[0017] In some embodiments, the processing unit decodes the polarization intensity distribution detected by the polarization photodetector unit to obtain the three-dimensional distribution information of the three-dimensional light field to be measured, specifically including:

[0018] The image with the highest contrast and the fewest overexposed areas is selected from the coded point spread function imaging patterns of each polarization channel. The rough absolute depth information of the three-dimensional light field to be measured is calculated based on the correspondence between the point spread function and the imaging distance z.

[0019] Polarization information is calculated using images from each polarization channel, and the zenith angle and azimuth angle of the normal to the surface of the object illuminated by the beam are calculated based on the polarization information.

[0020] The roughness absolute depth information of the three-dimensional light field to be measured is used to correct the π ambiguity in the zenith angle and the error in the azimuth angle of the object surface normal calculated based on polarization information, so as to obtain the corrected object surface normal information.

[0021] The corrected surface normal information of the object is integrated to obtain the three-dimensional shape of the object surface, and then fused with the roughness absolute depth information of the three-dimensional light field to be measured to obtain accurate absolute depth information, thereby completing the reconstruction of the three-dimensional distribution information of the three-dimensional light field to be measured.

[0022] In some embodiments, let θ and azimuth angle be the zenith angle and azimuth angle of the normal to the surface of the object illuminated by the beam, calculated based on the polarization information, respectively. The calculation formula is as follows:

[0023]

[0024] In the formula, P is the degree of polarization calculated from the images of the four polarization channels of the polarization photodetector unit, which is calculated based on the images of the four polarization channels of the polarization photodetector unit, namely the linear polarization directions of 0°, 45°, 90°, and 135°; n is the refractive index of the object illuminated by the beam; I 0° I 45° I 90° I 135° These are the linearly polarized light intensities measured by the polarization photoelectric detection unit in four directions.

[0025] A second aspect of this disclosure provides a three-dimensional imaging method based on a three-dimensional imaging system according to any embodiment of the first aspect of this disclosure, comprising:

[0026] Design an imaging unit consisting of a DOE or a metalens with the ability to independently adjust the imaging distance and light transmission phase, and require that both the DOE and the metalens contain at least one coded point spread function related to the imaging distance.

[0027] The three-dimensional light field to be measured and the polarization photoelectric detection unit are respectively set at the object plane and the image plane of the imaging unit;

[0028] The polarization photoelectric detection unit is used to detect the intensity of polarized light along different polarization channel directions on the image plane after the three-dimensional light field to be measured is imaged by the imaging unit;

[0029] The polarization intensity distribution detected by the polarization photoelectric detection unit is decoded to obtain the three-dimensional distribution information of the three-dimensional light field to be measured.

[0030] Compared with the prior art, this disclosure has the following characteristics and beneficial effects:

[0031] The core component of this disclosure is a single-layer metalens composed of two materials with different refractive indices or a diffractive optical element with at least two steps composed of a single refractive index material. Through a well-designed system, the target three-dimensional light field is imaged onto a polarization photodetector unit. By using spatial and polarization multiplexing, the three-dimensional light field can be directly reconstructed from the captured images of each polarization channel, thus supporting snapshot-style real-time three-dimensional light field measurement. Compared to traditional three-dimensional imaging technologies, this disclosure is extremely compact and miniaturized, requiring only one DOE / metalens and one photodetector unit (such as a polarization camera), making it suitable for space-constrained environments. Furthermore, due to the use of a three-dimensional solution method based on encoded point spread function and polarization imaging, this system exhibits higher robustness and maintains higher measurement accuracy compared to traditional TOF, binocular, and structured light methods. The metalens / DOE fabrication is compatible with complementary metal-oxide-semiconductor (CMOS) fabrication processes, enabling integration with various micro / nano optical platforms, resulting in high compatibility and scalability. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of the structure of a three-dimensional imaging system provided in the first aspect of this disclosure.

[0033] Figure 2 This is a side view of a metasurface for three-dimensional imaging provided in the first aspect of this disclosure.

[0034] Figure 3 This is a top view of a metasurface for three-dimensional imaging provided in the first aspect of this disclosure.

[0035] Figure 4 This is a schematic diagram of the metasurface phase used for three-dimensional imaging in the imaging method provided in the first aspect of this disclosure.

[0036] Figure 5 This is a schematic diagram of a point spread function for three-dimensional imaging in an imaging system provided by the first aspect of this disclosure.

[0037] Figure 6 This is a schematic diagram of the data processing process of the processing unit in the imaging system provided in the first aspect of the present disclosure. Detailed Implementation

[0038] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0039] Conversely, this application covers any substitutions, modifications, equivalent methods, and schemes made within the spirit and scope of this application as defined by the claims. Furthermore, to provide the public with a better understanding of this application, certain specific details are described in detail below. However, this application can be fully understood by those skilled in the art even without these detailed descriptions.

[0040] This disclosure discloses a compact imaging system composed of an imaging unit made of a single-layer diffractive optical element (DOE) or a superlens, combined with a polarization sensor. This system enables high-precision 3D reconstruction of 3D light field information. The coded point spread function imaging pattern formed by the DOE or superlens is recorded by the polarization photodetector unit. The imaging pattern corresponds to the imaging distance and is used to calculate the 3D information recorded by the coded point spread function. Each polarization channel of the polarization photodetector unit can calculate the degree of polarization and the polarization angle, which are used to calculate the polarization 3D information. By organically fusing these two types of 3D information, a universal 3D light field information reconstruction is achieved, which is expected to be applied in various 3D imaging fields.

[0041] See Figure 1 The first aspect of this disclosure provides a three-dimensional imaging system, including a three-dimensional light field to be measured 100, an imaging unit 200, a polarization photoelectric detection unit 300, and a processing unit (not shown in the figure) arranged along the light propagation direction, and constructing a right-handed coordinate system composed of the x-axis, y-axis, and z-axis, wherein the light propagation direction is defined as the z-axis, and the plane where the three-dimensional light field to be measured is located is defined as the xy plane;

[0042] The imaging unit 200 is selected from diffractive optical elements (DOE) or metalenses that have the ability to independently adjust the imaging distance and light transmission phase, and both the DOE and the metalens are required to contain at least one coded point spread function related to the imaging distance.

[0043] The polarization photoelectric detection unit 300 is used to detect the intensity of polarized light along different polarization channel directions on the image plane after the three-dimensional light field to be measured is imaged by the imaging unit 200.

[0044] The processing unit is used to decode the polarization intensity distribution detected by the polarization photoelectric detection unit 300 to obtain the three-dimensional distribution information of the three-dimensional light field 100 to be measured.

[0045] In some embodiments, the imaging unit 200 focuses the three-dimensional light field 100 to be measured onto the plane of the polarization photoelectric detection unit 300; the imaging model can be represented as:

[0046]

[0047] Where (x',y') are the orthogonal rectangular coordinates of the plane where the polarization photodetector unit 300 is located; g(x',y') are the polarization intensity distributions of one or more bands acquired by the polarization photodetector unit 300, i.e., the coded image; f z (x,y) represents the light intensity of the 3D light field to be measured at depth z in the 3D scene, PSF. z (x'-x, y'-y) is the encoded point spread function of the imaging system for a point source at depth z, and η(λ) is the noise of the polarization photodetector unit 300 in the λ-wavelength channel. Ultimately, the light intensity collected by each pixel on the polarization photodetector unit 300 can be derived from the superposition of images from point sources located at different depths.

[0048] In some embodiments, the three-dimensional light field 100 to be measured is disposed on the object plane of the imaging unit 200 and can be generated by irradiating the object with a light beam. The embodiments of this disclosure are applicable to various objects, such as transparent media, high reflectivity objects, objects with rough surfaces (such as metals with high surface roughness such as aluminum, magnesium, zinc, titanium, copper and iron), and objects with little texture (such as cardboard boxes with little texture).

[0049] In some embodiments, the imaging unit 200 is a monolithic metalens. Figure 2 and Figure 3 The diagram shows a front view and a top view of a metalens in one embodiment. The metalens consists of a substrate 202 and several optical antennas 201 arranged in a two-dimensional array on the surface of the substrate 202. The height of each optical antenna 201 is within the subwavelength range, and they are all made of a high-refractive-index (refractive-index > 2) dielectric material, including silicon, silicon nitride, titanium dioxide, gallium phosphide, gallium nitride, or gallium arsenide. In this embodiment, the metalens is a transmissive type. The substrate 202 can be made of transparent materials such as fused silica or sapphire (transmittance > 80%), and each optical antenna is made of silicon. The row and column directions of the optical antenna array on the surface of the substrate 202 are parallel to the x-axis and y-axis, respectively. The cross-sectional shape of each optical antenna along the xy-plane can be rectangular, elliptical, or other shapes with C2 symmetry (circular in this embodiment). With a fixed height, each optical antenna in an optical antenna array can be used to calculate the modulation of the phase and transmittance of the incident light in the working band at different diameters using methods such as finite-difference time-domain analysis or rigorous coupled-wave analysis. This allows for the acquisition of a set of nano-antenna structures for controlling the phase and transmittance of the incident light in the ranges of 0-2π and 0-1, respectively.

[0050] In one embodiment, the metalens is designed to have Figure 4 The phase distribution shown is obtained by arranging spiral phases with an increasing topological quantum number in the Fresnel zone, which can achieve, as shown in the figure. Figure 5 The point spread function shown is strongly correlated with depth. The point spread function consists of two foci where the angle between the lines varies with the depth z of the object point. 601 represents the rotation angle of the point spread function, which has a definite one-to-one correspondence with the depth.

[0051] Furthermore, the wavelength of the incident light can be designed to range from ultraviolet to infrared. The materials used to fabricate the substrate 202 and the optical antenna 201 can be different combinations of sapphire and titanium dioxide or sapphire and gallium nitride, which have different refractive index differences. Optionally, the refractive index difference between the optical antenna 201 and the substrate 202 should be greater than or equal to 0.5. In specific implementations, more materials may be included, but this embodiment does not limit this.

[0052] In other embodiments, the imaging unit 200 is a DOE, specifically an optical element with at least a two-step surface engraving morphology made of a single refractive index material such as optical glass, optical plastic, or imprinting adhesive. The phase design of the DOE is initialized using the Fresnel zone method, which has a ring region with a spiral phase distribution carrying an increase in topological quantum number toward the outer ring. The morphology of the point spread function is further improved by Fourier iterative optimization.

[0053] In some embodiments, the polarization photodetector unit 300 is disposed on the image plane of the imaging unit 200, and may include, but is not limited to, a polarization camera and a photodetector with polarization beam splitting function. Optionally, the polarization photodetector unit 300 has four linear polarization channels, namely 0°, 45°, 90°, and 135° linear polarization directions, which can capture the polarization intensity of the three-dimensional light field 100 under test along the 0°, 45°, 90°, and 135° linear polarization directions on the image plane of the imaging unit 200 after it is imaged by the imaging unit 200, i.e., the encoded point spread function imaging pattern.

[0054] In some embodiments, see Figure 6 The processing unit is used to process the polarized light intensity detected by the polarization photoelectric detection unit 300 to reconstruct the three-dimensional light field 100 to be measured, including the following steps:

[0055] Step S1: Select the image with the highest contrast and the fewest overexposed areas from the aforementioned encoded point spread function imaging patterns of the four polarization channels. Calculate the rough absolute depth information of the three-dimensional light field under test with low resolution and accuracy based on the correspondence between the point spread function and the imaging distance z. Its relative error Δz / z can reach more than 2%, and the resolution is generally below 100*100.

[0056] Step S2: Calculate the degree of polarization and polarization angle information through the images of the four polarization channels, and then calculate the zenith angle and azimuth angle information of the normal to the object surface; then recover the morphology of the three-dimensional light field 100 to be measured through the normal information.

[0057] Step S3: Use the rough absolute depth information obtained in step S1 to correct the π ambiguity and azimuth angle of the zenith angle of the object surface normal calculated based on polarization information in step S2. The existing errors are used to obtain the corrected surface normal information of the object;

[0058] Step S4: Integrate the corrected object surface normal information obtained in step S3 to obtain the three-dimensional shape of the object surface, and fuse it with the rough absolute depth information obtained in step S1 to obtain the high-resolution and high-precision accurate absolute depth information of the three-dimensional light field to be measured, thereby completing the reconstruction of the three-dimensional distribution information of the three-dimensional light field to be measured.

[0059] Furthermore, the zenith angle and azimuth angle of the object surface normal in step S2 are calculated based on the following formula:

[0060]

[0061] In the formula, P is the degree of polarization calculated from the image through four polarization channels, n is the refractive index of the target in the three-dimensional scene (i.e., the object illuminated by the beam), and θ is the zenith angle of the surface normal of the object illuminated by the beam. Let I be the azimuth angle of the normal to the surface of the object illuminated by the light beam. 0° I 45° I 90° I 135° These are the polarization photoelectric detection units 300 measuring the intensity of linearly polarized light in four directions. The azimuth angle is one of them. Due to the inverse trigonometric function relationship, there are two solutions that differ by π. If this ambiguity is not corrected, it will cause obvious errors in the 3D reconstruction of the object.

[0062] Furthermore, in step S3, the azimuth angle of the object surface normal corresponding to the rough absolute depth information obtained in step S1 is first determined. The azimuth angle of the object surface normal calculated based on polarization information in correction step S2 The existing π ambiguity, specifically, is in each pixel's... and Choose the one that is closer to the zenith angle θ0 of the object surface normal corresponding to the rough absolute depth information obtained in step S1; then correct the error in the zenith angle θ of the object surface normal calculated based on the polarization information in step S2 based on the zenith angle θ0 of the absolute depth information. Specifically, find a suitable n by iteratively to minimize the sum of the absolute errors of θ0 and θ, thereby obtaining the corrected object surface normal information.

[0063] Further, in step S4, the corrected object surface normal information obtained in step S3 is first integrated in two dimensions to obtain the three-dimensional shape of the object surface. At this time, only the relative three-dimensional shape is available, and the true distance and size information between the target object and the scene are lacking. Therefore, the object distance and size information obtained in the rough absolute depth information obtained in step S1 are used to assign a value to the depth of each pixel of the relative three-dimensional shape, thus obtaining high-resolution and high-precision absolute depth information. The reconstruction of the three-dimensional distribution information of the three-dimensional light field to be measured is completed. Its relative error Δz / z is generally within 0.5%, and the resolution can reach more than 1000*1000.

[0064] The three-dimensional imaging system provided in the first aspect of this disclosure combines an imaging unit based on DOE or metalens with a polarization photodetector unit, improving the system's integration. The three-dimensional light field of the incident light is captured by the imaging unit 200 as a blurred image composed of two double-helix point spread functions. The polarization intensity of the linearly polarized light along different polarization channel directions (e.g., 0°, 45°, 90°, 135°) is received and detected by the polarization photodetector unit 300. The better polarization image is used to calculate the three-dimensional information recorded by the encoded point spread function. The degree of polarization and polarization angle can be calculated using different polarization channels, which are then used to calculate polarization three-dimensional information. The organic fusion of these two three-dimensional information results in a universal and high-precision reconstruction of the light field's three-dimensional information, potentially applicable to various three-dimensional imaging fields.

[0065] A three-dimensional imaging method provided in the second aspect of this disclosure includes the following steps:

[0066] Design an imaging unit consisting of a DOE or a metalens with the ability to independently adjust the imaging distance and light transmission phase, and require that both the DOE and the metalens contain at least one coded point spread function related to the imaging distance.

[0067] The three-dimensional light field to be measured and the polarization photoelectric detection unit are respectively set at the object plane and the image plane of the above imaging unit;

[0068] The intensity of polarized light along different polarization channels on the image plane after the three-dimensional light field under test is imaged by the imaging unit is detected using a polarization photoelectric detection unit.

[0069] The three-dimensional distribution information of the three-dimensional light field to be measured is obtained by decoding the polarization intensity distribution detected by the polarization photoelectric detection unit.

[0070] It should be noted that the foregoing explanation of the embodiments of the three-dimensional imaging system provided in the first aspect of this disclosure also applies to the three-dimensional imaging method of this embodiment, and will not be repeated here.

[0071] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0072] Although embodiments of this application have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of this application, the scope of which is defined by the claims and their equivalents.

Claims

1. A three-dimensional imaging system, characterized by, The application relates to a three-dimensional light field detection device, which comprises a to-be-detected three-dimensional light field, an imaging unit, a polarized photoelectric detection unit and a processing unit arranged along the light propagation direction, and a right-handed coordinate system composed of an x axis, a y axis and a z axis, wherein the light propagation direction is defined as the z axis, and a plane where the to-be-detected three-dimensional light field is located is defined as the xy plane. The imaging unit is selected from a diffractive optical element or a super lens which has the ability to independently control the imaging distance and the light transmission phase, and the diffractive optical element and the super lens each comprises at least one encoding point spread function related to the imaging distance. The polarized photoelectric detection unit is used for detecting the polarized light intensity of the to-be-detected three-dimensional light field on the image plane after imaging by the imaging unit along different polarization channel directions. The processing unit is used for decoding the polarized light intensity distribution detected by the polarized photoelectric detection unit to obtain the three-dimensional distribution information of the to-be-detected three-dimensional light field, and the decoding comprises the following steps: calculating the rough absolute depth information of the to-be-detected three-dimensional light field based on the corresponding relationship between the point spread function and the imaging distance; correcting the existing pi ambiguity of the zenith angle and the error of the azimuth angle of the object surface normal calculated based on the polarization information by using the rough absolute depth information of the to-be-detected three-dimensional light field; obtaining the three-dimensional topography of the object surface based on the corrected object surface normal information, and fusing the three-dimensional topography with the rough absolute depth information of the to-be-detected three-dimensional light field to obtain the accurate absolute depth information, so as to obtain the three-dimensional distribution information of the to-be-detected three-dimensional light field.

2. The three-dimensional imaging system of claim 1, wherein, The to-be-detected three-dimensional light field is arranged on the object plane of the imaging unit and is generated by illuminating the object surface with a light beam.

3. The three-dimensional imaging system of claim 1, wherein, The diffractive optical element is an optical element composed of a material with a single refractive index and having at least two stepped surface engraved topographies.

4. The three-dimensional imaging system of claim 3, wherein, The material with a single refractive index is selected from optical glass, optical plastic or embossed glue.

5. The three-dimensional imaging system of claim 1, wherein, The super lens is composed of a substrate and optical antennas distributed in a two-dimensional array form on the surface of the substrate, each optical antenna has C2 symmetry in the cross-sectional shape along the xy plane, the height of each optical antenna is not more than one working wavelength, and there is a refractive index difference between the material for manufacturing the optical antenna and the substrate.

6. The three-dimensional imaging system of claim 5, wherein, The refractive index difference between the material for manufacturing the optical antenna and the substrate should be greater than or equal to 0.

5.

7. The three-dimensional imaging system of claim 1, wherein, The polarized photoelectric detection unit is selected from a polarization camera or a photoelectric detector with polarization light splitting function, and the polarized photoelectric detection unit has at least three polarization channels of polarization states, and when four linear polarization channels are adopted, the linear polarization directions are 0 DEG, 45 DEG, 90 DEG and 135 DEG respectively.

8. The three-dimensional imaging system of claim 1, wherein, The processing unit decodes the polarized light intensity distribution detected by the polarized photoelectric detection unit to obtain the three-dimensional distribution information of the to-be-detected three-dimensional light field, and the decoding specifically comprises the following steps: selecting an image with the highest contrast and the least overexposure area from the encoding point spread function imaging patterns of each polarization channel to calculate the rough absolute depth information of the to-be-detected three-dimensional light field based on the corresponding relationship between the point spread function and the imaging distance z; calculating the polarization information by using the images of each polarization channel, and calculating the zenith angle and azimuth angle information of the object surface normal irradiated by the light beam according to the polarization information; The coarse absolute depth information of the three-dimensional light field to be measured is used to correct the π ambiguity of the zenith angle and the error of the azimuth angle of the object surface normal calculated based on the polarization information, to obtain corrected object surface normal information; The corrected object surface normal information is integrated to obtain the three-dimensional topography of the object surface, and is fused with the coarse absolute depth information of the three-dimensional light field to be measured, to obtain accurate absolute depth information, thereby completing the reconstruction of the three-dimensional distribution information of the three-dimensional light field to be measured.

9. The three-dimensional imaging system of claim 8, wherein, Let the zenith angle and the azimuth angle of the surface normal of the object irradiated by the light beam calculated according to the polarization information be θ and φ, respectively The calculation formula is as follows: In the formula, P is a degree of polarization calculated by images of four polarization channels of the polarization photoelectric detection unit, and is calculated according to images of four polarization channels, i.e., 0°, 45°, 90°, and 135° linear polarization directions, of the polarization photoelectric detection unit; n is a refractive index of an object irradiated by a light beam; , , , are linearly polarized light intensities in four directions measured by the polarization photoelectric detection unit.

10. A three-dimensional imaging method based on the three-dimensional imaging system according to any one of claims 1 to 9, characterized by, Comprise: A DOE or superlens with independent regulation of imaging distance and light transmission phase is designed to form an imaging unit, and it is required that the DOE and the superlens each contain at least one coding point spread function related to the imaging distance; A three-dimensional light field to be measured and a polarization photoelectric detection unit are arranged at the object plane and the image plane of the imaging unit respectively; The polarization photoelectric detection unit is used to detect the polarization light intensity along different polarization channel directions on the image plane after the three-dimensional light field to be measured is imaged by the imaging unit; The polarization light intensity distribution detected by the polarization photoelectric detection unit is decoded to obtain the three-dimensional distribution information of the three-dimensional light field to be measured.

Images