Method and system for detecting subsurface defects of optical elements based on phase-contrast digital holography

Abstract

The application discloses an optical element subsurface defect detection method and system based on phase contrast digital holography, and belongs to the field of precision measurement technology; the system adopts annular Bessel beam Kohler illumination, utilizes polarization modulation technology and a specially-made polarization contrast ring to realize synchronous four-step phase shift high-speed quantitative phase measurement in a U-shaped light path; the method acquires complex amplitude through a polarization camera, compensates for resolution loss by using a self-supervision super-resolution reconstruction algorithm fused with a physical model; full-field and local inverse diffraction reconstruction is carried out based on angular spectrum diffraction theory, high-precision depth positioning is realized based on the diffraction light complex amplitude distribution characteristics as a focusing criterion; three-dimensional information representation of defects is realized in combination with an optical field scattering and diffraction model and a machine learning algorithm; the application can realize high-sensitivity, high-signal-to-noise ratio, real-time in-situ quantitative detection and three-dimensional reconstruction of submicron defects without significantly increasing the cost.

Description

Technical Field

[0001] This invention belongs to the field of precision measurement technology, and in particular relates to a method and system for detecting subsurface defects in optical components based on phase-contrast digital holography. Background Technology

[0002] High-power laser devices play an irreplaceable role in high-energy-density physics, inertial confinement fusion research, and other fields. Fused silica, as the primary material for large-aperture optical elements in high-power laser systems, directly impacts the stability and efficiency of the entire system. Researching defect detection techniques for fused silica optical elements can not only improve the quality and lifespan of these elements but also enhance the performance and stability of the laser system, thus better meeting the stringent quality requirements of high-energy-density physics and inertial confinement fusion research. By accurately detecting and evaluating defects in fused silica optical elements, appropriate repair and protection measures can be implemented, reducing the impact of defects on element performance, extending their lifespan, lowering replacement frequency and costs, and improving economic efficiency. Simultaneously, research on defect detection techniques will drive the development of optical materials science, providing reference and guidance for research on other optical materials, and promoting progress in optical materials science. High-power laser technology has wide applications in defense, materials processing, medicine, and communications, among other fields. Improving the performance of laser devices will meet the demands of various fields for high-power lasers and drive the development of high-power laser technology.

[0003] Research on defect detection technology for fused silica optical components is not only of great significance for promoting inertial confinement fusion research and improving high-power laser technology, but also for advancing optical materials science, reducing economic costs, ensuring safe production, and driving the development of high-end manufacturing. This research will have a profound impact and positive impetus on multiple scientific and engineering fields, enhance my country's international competitiveness in advanced optical manufacturing technology, break through foreign technological monopolies, and provide strong support for improving the performance of my country's major equipment and advanced instruments. Currently, the academic community's methods for detecting defects in fused silica all have limitations: existing non-optical defect detection methods all require specialized equipment and long sample preparation time, and do not have real-time in-situ measurement capabilities; among optical measurement methods, optical coherence tomography has low longitudinal resolution, confocal scanning microscopy has high requirements for sample and environmental stability and is difficult to measure phase-type defects, fluorescence confocal microscopy requires sample staining and labeling and cannot achieve in-situ measurement, and photothermal weak absorption signals are weak and require an external excitation source. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for detecting subsurface defects in optical elements based on phase-contrast digital holography. This addresses the problems of existing technologies, such as the lack of real-time in-situ measurement capabilities in non-optical methods for detecting subsurface defects in transparent optical elements like fused silica, the limitations of optical methods in various applications, susceptibility to interference in phase-shift holographic acquisition, low accuracy in defect depth localization, and the inability to achieve quantitative characterization and three-dimensional reconstruction due to the limitations of qualitative detection. This invention combines digital holographic diffraction reconstruction with phase-contrast microscopy to highlight object edge features. While utilizing phase imaging to improve the sensitivity of defect feature detection and reduce background interference, it uses a four-step phase-shift quantitative method to obtain the complex amplitude of the defect light field, thereby determining the defect depth. Ultimately, this achieves high-sensitivity, high-speed detection of defects in fused silica elements.

[0005] To achieve the above objectives, the present invention provides a method for detecting subsurface defects in optical elements based on phase-contrast digital holography, comprising the following steps: S1. Construct a measurement system that includes a ring-shaped illumination optical path, an imaging optical path, and a U-shaped phase contrast introduction optical path, and use polarization beam splitting modulation to realize the conversion measurement of dark field scattering and phase contrast microscopy imaging. S2. Using polarization coding and polarization phase contrast ring, synchronous four-step phase-shift images are acquired through a polarization camera, and the complex amplitude information of the object's light field is quantitatively calculated using a four-step phase-shift algorithm. S3. A digital holographic self-supervised super-resolution reconstruction algorithm based on a fusion physical model is used to compensate for the synchronous four-step phase-shifting image and complex amplitude acquired in S2. S4. Based on the angular spectrum diffraction theory, the complex amplitude after compensation in S3 is reconstructed by complex amplitude diffraction, and the energy Laplace method is used to calculate the clarity of the complex amplitude diffraction reconstruction image for depth positioning. S5. Combining the physical field models of diffraction and scattering of defects under bright and dark field conditions, a mapping relationship between the optical signal of the defect and its depth and physical properties is established based on machine learning methods to perform three-dimensional characterization of the defect.

[0006] Preferably, the digital holographic self-supervised super-resolution reconstruction algorithm based on the fusion physical model in S3 compensates for the decrease in resolution and lateral distortion of the synchronous four-step phase-shift image and complex amplitude acquired in S2 caused by the polarization camera.

[0007] Preferably, the specific content of S4 is as follows: S401. Perform numerical propagation calculation of complex amplitude inverse diffraction on the compensated complex amplitude in S3. S402. The energy Laplace method is used to perform depth localization by calculating the sharpness of the complex amplitude diffraction reconstruction pattern.

[0008] Preferably, the specific content of S401 is as follows: Based on the theory of angular spectral diffraction, the planar light field information of the polarization camera target surface is propagated backward to the plane of the optical element under test, and the defect light field distribution at different propagation distances is reconstructed. The expression is as follows: ; In the formula, For the propagation distance is Then, the complex amplitude distribution on the target depth plane; The complex amplitude of the original object's light field acquired by the camera target surface after compensation; The imaginary unit; The wavelength of the illumination light; This is the propagation distance of the inverse diffraction; In the frequency domain Spatial frequency of direction; In the frequency domain Spatial frequency of direction; Fourier transform; This is the inverse Fourier transform; The limited pixel size leads to a discretization sampling effect in the interference intensity, which is corrected in the actual diffraction reconstruction process, as shown in the following expression: ; ; In the formula, After discretization, propagation to depth The complex amplitude distribution in the plane, where for Orientation pixel index, for Orientation pixel index, for Orientation pixel size, for Orientation pixel size; To discretize the complex amplitude of the original optical field; This is the propagation distance of the inverse diffraction; This is a scaling factor for pixel size related to wavelength; for The fundamental frequency in the discrete frequency domain of the direction; for The fundamental frequency in the discrete frequency domain of the direction; For the image in Total number of pixels in the direction; For the image in Total number of pixels in the direction; The discretized interference intensity is denoted as The full-field optical field diffraction reconstruction calculation at different depths was completed by modifying the formula.

[0009] Preferably, the specific content of S402 is as follows: S4021. Determine the optimal resolution plane for the entire field; Based on the inverse diffraction formula in S401, the repaired hologram was reconstructed by full-field inverse diffraction to obtain full-field diffraction reconstruction images of different depth planes. The energy Laplace method was used as the focus evaluation function to calculate the sharpness values ​​of the full-field diffraction reconstruction images at each depth. The plane corresponding to the maximum sharpness value is selected as the plane with the highest full-field resolution. The hologram is then reconstructed by full-field diffraction to achieve global focusing. S4022, Achieve precise depth positioning of a single defect; Individual defects are identified using image recognition and connected component segmentation algorithms. Sub-regions containing individual defects are then used as computational objects. Inverse diffraction reconstruction is performed on the entire field, and the energy Laplace function value within each sub-region is calculated. The propagation distance corresponding to the maximum value of the energy Laplace function is taken as the depth position of the defect, thus completing the local focusing of all defects.

[0010] Preferably, S5 employs a method that integrates multimodal optical measurement and data analysis, as detailed below: S501. Based on the angular spectrum diffraction theory, establish a physical field model for the diffraction and scattering of micron-sized defects under bright and dark field conditions, and refine the sensitive index of depth change through numerical simulation. S502. Design a high-precision defect depth positioning algorithm using the depth change index of defect optical properties; determine the axial position and quantitative phase of the defect through digital holographic numerical reconstruction; and establish the mapping relationship between the defect optical signal and its depth and physical properties using machine learning. S503. Perform quantitative characterization of defect type, size, and depth; For surface defects located at the focal plane, dual-channel imaging results of phase-contrast microscopy and dark-field scattering are obtained by modulating polarized illumination light, and the results are characterized by quantitative phase information obtained by four-step phase shifting. To address subsurface defocusing defects, the spatial diffraction reconstruction characteristics of digital holography are utilized to perform depth localization and three-dimensional reconstruction by analyzing the interference fringes of diffracted light and transmitted background light.

[0011] Preferably, the optical element subsurface defect detection system based on phase-contrast digital holography includes: an optical path module, a core functional module, and a motion control module; The optical path module is used for the conversion between dark field scattering and phase-contrast microscopy imaging and for polarization-synchronized phase-shift image acquisition; The core functional modules are used to complete image conversion, data acquisition, and defect localization and characterization; The motion control module is used to precisely control the axial movement of the objective lens or the entire microscopic imaging system.

[0012] Preferably, the optical path module includes an imaging optical path unit, a U-shaped phase contrast introduction optical path unit, and an annular illumination optical path unit.

[0013] Preferably, the optical path unit includes a ring light source, a field stop, an amplitude mask, a first objective lens, a transparent stage, a second objective lens, and a tube lens; used for basic imaging of the optical element under test, providing the original light field for phase contrast introduction and data acquisition; The U-shaped phase-contrast optical path unit includes a first reflecting prism, a first Fourier transform lens, a phase mask, a second reflecting prism, a second Fourier transform lens, and a polarization camera; it introduces a phase-contrast effect for the modulation and acquisition of phase information, and completes synchronous four-step phase-shifting optical field processing. The annular illumination sub-optical path unit includes a first conical lens, a second conical lens, and a converging lens; the first and second conical lenses are placed symmetrically; they are used to generate a focused annular Bessel beam, forming a Kohler illumination pattern, and providing uniform and stable annular parallel illumination light for the entire system.

[0014] Preferably, the object plane, image plane, and polarization camera target surface are conjugate to ensure accurate image transmission; the amplitude mask, the back focal plane of the first objective lens, and the phase mask are conjugate to accurately modulate the phase contrast effect.

[0015] Therefore, the beneficial effects of the above-mentioned method and system for detecting subsurface defects in optical elements based on phase-contrast digital holography in this invention are as follows: (1) Relying on polarization modulation technology and a specially designed polarization phase-contrast ring, synchronous four-step phase shift high-speed quantitative phase measurement is completed in the U-shaped optical path. Four-step phase shift images can be obtained in a single exposure, avoiding the problem of traditional multiple exposure acquisition being easily affected by optical path vibration and sample displacement interference, greatly improving detection efficiency and adapting to the real-time detection needs of industrial sites. (2) Based on the angular spectrum diffraction theory, full-field and local inverse diffraction reconstruction is carried out. Combined with the dual-layer focusing strategy of global focusing and local focusing and the energy Laplace method, high-precision depth positioning of defects is achieved. Furthermore, the imaging deviation is compensated by the self-supervised super-resolution reconstruction algorithm that integrates physical models, ensuring the accuracy of the optical field complex amplitude data and laying a reliable foundation for defect characterization. (3) By integrating phase contrast microscopy and digital holography, and combining the physical field model of light field scattering and diffraction with machine learning algorithms, a mapping relationship between defect optical signals and physical properties can be established. This can accurately determine the type, size, depth and other parameters of defects, breaking through the limitation of traditional technology that can only perform qualitative detection, and providing accurate quantitative basis for quality assessment and defect repair of optical components. (4) A composite measurement system is used to achieve flexible conversion between dark field scattering and phase contrast microscopy imaging without the need for additional hardware replacement. The optical path adopts a U-shaped layout design, which is compact and has high space utilization. While significantly improving the detection performance, it does not significantly increase the equipment cost, and is both practical and economical. (5) Different detection and characterization strategies are designed for focal plane surface defects and sub-surface defocus defects, which can achieve full coverage detection of defects of different types and locations. No pretreatment such as staining and labeling of samples is required, and no external excitation source is required. The detection process is simple and has low requirements for the detection environment and samples. (6) Combining multi-dimensional physical field simulation, experimental verification and machine learning technology that integrates physical models not only ensures the physical interpretability of the detection method, but also improves the intelligence and accuracy of the detection.

[0016] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the subsurface defect detection system for optical elements based on phase-contrast digital holography according to the present invention; Figure 2 This is a diagram of the lighting path. Figure 3 This is a flowchart of the subsurface defect detection method for optical elements based on phase-contrast digital holography according to the present invention; Figure 4 This is a flowchart of the focusing and positioning process of the present invention. Figure 5 The symmetrical hologram obtained from the simulation; Figure 6 The curve showing how the focus evaluation function changes with depth; Reference numerals: 1. Ring light source; 2. Field stop; 3. Amplitude mask; 4. First objective lens; 5. Transparent stage; 6. Second objective lens; 7. Tube lens; 8. First reflecting prism; 9. First Fourier transform lens; 10. Phase mask; 11. Second reflecting prism; 12. Second Fourier transform lens; 13. Polarizing camera; 14. First cone lens; 15. Second cone lens; 16. Converging lens. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages disclosed in the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the embodiments of the present invention and are not intended to limit the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of this application. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout.

[0019] It should be noted that the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, such as a process, method, system, product, or server that includes a series of steps or units, not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such process, method, product, or device.

[0020] The following is combined Figures 1-6 The embodiments of the present invention will be described in detail below.

[0021] Example 1 The present invention provides a method for detecting subsurface defects in optical elements based on phase-contrast digital holography. This method is based on phase-contrast microscopy and polarization-synchronous phase-shifting techniques, achieving the conversion measurement of dark-field scattering and phase-contrast microscopic imaging through polarization-splitting modulation. On this basis, a system combining phase-contrast microscopy and digital holography for subsurface defect microscopic imaging and synchronous phase-shifting is designed. Real-time measurement of phase information is achieved using polarization encoding and a specially designed polarization phase-contrast ring. A digital holographic self-supervised super-resolution reconstruction algorithm with a fused physical model is used to compensate for the resolution reduction and lateral distortion caused by the polarization camera 13. A physical field model of diffraction and scattering of defects under bright and dark field conditions is constructed to study the diffraction and scattering characteristics of submicron-level micro-defects at different focusing distances. Machine learning is used to guide the design of a high-precision three-dimensional defect information representation model and a depth-determining algorithm based on simulation results. The specific steps are as follows: S1. Construct a measurement system that includes a ring-shaped illumination optical path, an imaging optical path, and a U-shaped phase contrast introduction optical path, and use polarization beam splitting modulation to realize the conversion measurement of dark field scattering and phase contrast microscopy imaging. S2. Using polarization coding and polarization phase contrast ring, synchronous four-step phase-shift images are acquired through polarization camera 13, and the complex amplitude information of the object's light field is quantitatively calculated using the four-step phase-shift algorithm. S3, a digital holographic self-supervised super-resolution reconstruction algorithm based on a fusion physical model, compensates for the decrease in resolution and lateral distortion of the synchronous four-step phase-shift image and complex amplitude acquired by S2 caused by polarization camera 13.

[0022] S4. Based on the angular spectrum diffraction theory, the complex amplitude after compensation in S3 is reconstructed by complex amplitude diffraction, and the energy Laplace method is used to calculate the clarity of the complex amplitude diffraction reconstruction image for depth positioning. S401. Perform numerical propagation calculation of complex amplitude inverse diffraction on the compensated complex amplitude in S3. Based on the angular spectral diffraction theory, the planar light field information of the target surface 14 of the polarization camera 13 is propagated backward to the object plane of the optical element under test, and the defect light field distribution at different propagation distances is reconstructed, as expressed below: ; In the formula, For the propagation distance is Then, the complex amplitude distribution on the target depth plane; The complex amplitude of the original object's light field acquired by the camera target surface after compensation; The imaginary unit; The wavelength of the illumination light; This is the propagation distance of the inverse diffraction; In the frequency domain Spatial frequency of direction; In the frequency domain Spatial frequency of direction; Fourier transform; This is the inverse Fourier transform; The limited pixel size leads to a discretization sampling effect in the interference intensity, which is corrected in the actual diffraction reconstruction process, as shown in the following expression: ; ; In the formula, After discretization, propagation to depth The complex amplitude distribution in the plane, where for Orientation pixel index, for Orientation pixel index, for Orientation pixel size, for Orientation pixel size; To discretize the complex amplitude of the original optical field; This is the propagation distance of the inverse diffraction; This is a scaling factor for pixel size related to wavelength; for The fundamental frequency in the discrete frequency domain of the direction; for The fundamental frequency in the discrete frequency domain of the direction; For the image in Total number of pixels in the direction; For the image in Total number of pixels in the direction; The discretized interference intensity is denoted as The full-field optical field diffraction reconstruction calculation at different depths was completed by modifying the formula.

[0023] S402. The energy Laplace method is used to perform depth localization by calculating the sharpness of the complex amplitude diffraction reconstruction pattern.

[0024] S4021. Determine the optimal resolution plane for the entire field; Based on the inverse diffraction formula in S401, the repaired hologram was reconstructed by full-field inverse diffraction to obtain full-field diffraction reconstruction images of different depth planes. The energy Laplace method was used as the focus evaluation function to calculate the sharpness values ​​of the full-field diffraction reconstruction images at each depth. The plane corresponding to the maximum sharpness value is selected as the plane with the highest full-field resolution. The hologram is then reconstructed by full-field diffraction to achieve global focusing. S4022, Achieve precise depth positioning of a single defect; Individual defects are identified using image recognition and connected component segmentation algorithms. Sub-regions containing individual defects are then used as computational objects. Inverse diffraction reconstruction is performed on the entire field, and the energy Laplace function value within each sub-region is calculated. The propagation distance corresponding to the maximum value of the energy Laplace function is taken as the depth position of the defect, thus completing the local focusing of all defects.

[0025] S5. Combining the physical field models of diffraction and scattering of defects under bright and dark field conditions, a mapping relationship between the optical signal of the defect and its depth and physical properties is established based on machine learning methods to perform three-dimensional characterization of the defect.

[0026] A method combining multimodal optical measurement and data analysis is employed. S501. Based on the angular spectrum diffraction theory, establish a physical field model for the diffraction and scattering of micron-sized defects under bright and dark field conditions, and refine the sensitive index of depth change through numerical simulation. S502. Design a high-precision defect depth positioning algorithm using the depth change index of defect optical properties; determine the axial position and quantitative phase of the defect through digital holographic numerical reconstruction; and establish the mapping relationship between the defect optical signal and its depth and physical properties using machine learning. S503. Perform quantitative characterization of defect type, size, and depth; For surface defects located at the focal plane, dual-channel imaging results of phase-contrast microscopy and dark-field scattering are obtained by modulating polarized illumination light, and the results are characterized by quantitative phase information obtained by four-step phase shifting. To address subsurface defocusing defects, the spatial diffraction reconstruction characteristics of digital holography are utilized to perform depth localization and three-dimensional reconstruction by analyzing the interference fringes of diffracted light and transmitted background light.

[0027] A subsurface defect detection system for optical components based on phase-contrast digital holography includes: an optical path module, a core functional module, and a motion control module; The optical path module is used for the conversion between dark field scattering and phase-contrast microscopy imaging and for polarization-synchronized phase-shift image acquisition; The core functional modules are used to complete image conversion, data acquisition, and defect localization and characterization; The motion control module is used to precisely control the axial movement of the objective lens or the entire microscopic imaging system.

[0028] The optical path module includes an imaging optical path unit, a U-shaped phase contrast introduction optical path unit, and a ring-shaped illumination optical path unit.

[0029] The optical path unit includes a ring light source 1, a field stop 2, an amplitude mask 3, a first objective lens 4, a transparent stage 5, a second objective lens 6, and a tube lens 7; it is used for basic imaging of the optical element under test and provides the original light field for phase contrast introduction and data acquisition. The U-shaped phase-contrast optical path unit includes a first reflecting prism 8, a first Fourier transform lens 9, a phase mask 10, a second reflecting prism 11, a second Fourier transform lens 12, and a polarization camera 13; it introduces a phase-contrast effect for the modulation and acquisition of phase information, and completes synchronous four-step phase-shifting optical field processing. The annular illumination sub-optical path unit includes a first conical lens 14, a second conical lens 15, and a converging lens 16; the first conical lens 14 and the second conical lens 15 are placed symmetrically; they are used to generate a focused annular Bessel beam, forming a Kohler illumination pattern, and providing uniform and stable annular parallel illumination light for the entire system.

[0030] The object plane, image plane, and target surface 14 of the polarization camera 13 are conjugate to ensure accurate image transmission; the amplitude mask 3, the back focal plane of the first objective lens 4, and the phase mask 10 are conjugate to accurately modulate the phase contrast effect.

[0031] Example 2 The subsurface defect detection system for optical components based on phase-contrast digital holography designed in this invention is described in [reference needed]. Figure 1As shown, the system includes an annular light source 1, a field stop 2, an amplitude mask 3, a first objective lens 4, a second objective lens 6, a transparent stage 5, a tube lens 7, a first reflecting prism 8, a second reflecting prism 11, a first Fourier transform lens 9, a second Fourier transform lens 12, a phase mask 10, and a polarization camera 13. The annular light source 1, field stop 2, amplitude mask 3, first objective lens 4, transparent stage 5, second objective lens 6, and tube lens 7 are arranged coaxially from top to bottom in the optical path, forming the imaging optical path. Below, the first reflecting prism 8 folds the optical path, followed by the phase contrast introduction optical path, which is composed of the first Fourier transform lens 9, amplitude mask 3, second reflecting prism 11, second Fourier transform lens 12, and polarization camera 13, making the entire phase contrast microscopy optical path U-shaped for integration in a smaller space. Figure 1 The solid line represents the direction of light propagation, and the dashed line represents the focal plane and the image plane. The focal length of the first Fourier transform lens 9 and the second Fourier transform lens 12 used in this invention is 50mm, the magnification of the first objective lens 4 and the second objective lens 6 is 50×, the numerical aperture NA is 0.6, and the working distance is 11mm.

[0032] In the entire optical path, the light source section consists of a first conical lens 14, a second conical lens 15, and a converging lens 16, as shown below. Figure 2 As shown, the converging lens 16 is placed between the first conical lens 14 and the second conical lens 15 to produce a focused annular Bessel beam. The object plane, image plane, and polarizing camera 13 are conjugates of each other, as are the amplitude mask 3, back focal plane, and phase mask. The illumination light transmitted through the object plane is annular parallel light, constituting Kohler illumination. The cone angle of the first conical lens 14 and the second conical lens 15 is 140°, the focal length of the converging lens 16 is 50mm, the distance between the first conical lens 14, the second conical lens 15, and the converging lens is 15mm, with the focal point at one focal length. At this point, the diameter of the annular illumination light is 6.5mm, and the width is 0.24mm.

[0033] The optical fields generated by five phase-type objects at the object plane were simulated, and the measurement results obtained after passing through the system designed in this invention are as follows: Figure 5 As shown, phase information that was previously difficult to observe under bright-field illumination is highlighted due to phase-contrast interference, demonstrating the system's sensitivity to fine phase structure detection. The measured complex amplitude of the light field is reconstructed at different distances, and then evaluated according to the focusing criterion function. The resulting focusing evaluation curve is shown below. Figure 6As shown, the recording depth at this point corresponds to the optimal focusing depth during scanning, which is also the maximum value of the focusing evaluation function curve. It can be seen that despite the fluctuation effect caused by the periodic change in phase, the focusing evaluation curve still reaches its maximum value near the focal point, with an axial deviation of 3 μm relative to the true focal value, far smaller than the 100 μm scale of the subsurface region. Therefore, the invented method can provide higher axial sensitivity for the detection of subsurface defects in optical components.

[0034] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for detecting subsurface defects in optical elements based on phase-contrast digital holography, characterized in that, Includes the following steps: S1. Construct a measurement system that includes a ring-shaped illumination optical path, an imaging optical path, and a U-shaped phase contrast introduction optical path, and use polarization beam splitting modulation to realize the conversion measurement of dark field scattering and phase contrast microscopy imaging. S2. Using polarization coding and polarization phase contrast ring, synchronous four-step phase-shift images are acquired through a polarization camera, and the complex amplitude information of the object's light field is quantitatively calculated using a four-step phase-shift algorithm. S3. A digital holographic self-supervised super-resolution reconstruction algorithm based on a fusion physical model is used to compensate for the synchronous four-step phase-shifting image and complex amplitude acquired in S2. S4. Based on the angular spectrum diffraction theory, the complex amplitude inverse diffraction numerical propagation calculation is performed on the compensated complex amplitude in S3, and the energy Laplace method is used to perform depth positioning by calculating the clarity of the complex amplitude diffraction reconstruction map. S5. Combining the physical field models of diffraction and scattering of defects under bright and dark field conditions, a mapping relationship between the optical signal of the defect and its depth and physical properties is established based on machine learning methods to perform three-dimensional characterization of the defect.

2. The method for detecting subsurface defects in optical elements based on phase-contrast digital holography according to claim 1, characterized in that: The digital holographic self-supervised super-resolution reconstruction algorithm based on the fusion physical model in S3 compensates for the resolution reduction and lateral distortion of the synchronous four-step phase-shift image and complex amplitude acquired by S2 caused by the polarization camera.

3. The method for detecting subsurface defects in optical elements based on phase-contrast digital holography according to claim 2, characterized in that, The specific details of S4 are as follows: S401. Perform numerical propagation calculation of complex amplitude inverse diffraction on the compensated complex amplitude in S3. S402. The energy Laplace method is used to perform depth localization by calculating the sharpness of the complex amplitude diffraction reconstruction pattern.

4. The method for detecting subsurface defects in optical elements based on phase-contrast digital holography according to claim 3, characterized in that, The specific details of S401 are as follows: Based on the theory of angular spectral diffraction, the planar light field information of the polarization camera target surface is propagated backward to the plane of the optical element under test, and the defect light field distribution at different propagation distances is reconstructed. The expression is as follows: ； In the formula, For the propagation distance is Then, the complex amplitude distribution on the target depth plane; The complex amplitude of the original object's light field acquired by the camera target surface after compensation; The imaginary unit; The wavelength of the illumination light; This is the propagation distance of the inverse diffraction; In the frequency domain Spatial frequency of direction; In the frequency domain Spatial frequency of direction; Fourier transform; This is the inverse Fourier transform; The limited pixel size leads to a discretization sampling effect in the interference intensity, which is corrected in the actual diffraction reconstruction process, as shown in the following expression: ； ； In the formula, After discretization, propagation to depth The complex amplitude distribution in the plane, where for Orientation pixel index, for Orientation pixel index, for Orientation pixel size, for Orientation pixel size; To discretize the complex amplitude of the original optical field; This is the propagation distance of the inverse diffraction; This is a scaling factor for pixel size related to wavelength; for The fundamental frequency in the discrete frequency domain of the direction; for The fundamental frequency in the discrete frequency domain of the direction; For the image in Total number of pixels in the direction; For the image in Total number of pixels in the direction; The discretized interference intensity is denoted as The full-field optical field diffraction reconstruction calculation at different depths was completed by modifying the formula.

5. The method for detecting subsurface defects in optical elements based on phase-contrast digital holography according to claim 4, characterized in that, The specific details of S402 are as follows: S4021. Determine the optimal resolution plane for the entire field; Based on the inverse diffraction formula in S401, the repaired hologram was reconstructed by full-field inverse diffraction to obtain full-field diffraction reconstruction images of different depth planes. The energy Laplace method was used as the focus evaluation function to calculate the sharpness values ​​of the full-field diffraction reconstruction images at each depth. The plane corresponding to the maximum sharpness value is selected as the plane with the highest full-field resolution. The hologram is then reconstructed by full-field diffraction to achieve global focusing. S4022, Achieve precise depth positioning of a single defect; Individual defects are identified using image recognition and connected component segmentation algorithms. Sub-regions containing individual defects are then used as computational objects. Inverse diffraction reconstruction is performed on the entire field, and the energy Laplace function value within each sub-region is calculated. The propagation distance corresponding to the maximum value of the energy Laplace function is taken as the depth position of the defect, thus completing the local focusing of all defects.

6. The method for detecting subsurface defects in optical elements based on phase-contrast digital holography according to claim 5, characterized in that, S5 employs a method that integrates multimodal optical measurement and data analysis, as detailed below: S501. Based on the angular spectrum diffraction theory, establish a physical field model for the diffraction and scattering of micron-sized defects under bright and dark field conditions, and refine the sensitive index of depth change through numerical simulation. S502. Design a high-precision defect depth positioning algorithm using the depth change index of defect optical properties; determine the axial position and quantitative phase of the defect through digital holographic numerical reconstruction; and establish the mapping relationship between the defect optical signal and its depth and physical properties using machine learning. S503. Perform quantitative characterization of defect type, size, and depth; For surface defects located at the focal plane, dual-channel imaging results of phase-contrast microscopy and dark-field scattering are obtained by modulating polarized illumination light, and the results are characterized by quantitative phase information obtained by four-step phase shifting. To address subsurface defocusing defects, the spatial diffraction reconstruction characteristics of digital holography are utilized to perform depth localization and three-dimensional reconstruction by analyzing the interference fringes of diffracted light and transmitted background light.

7. A subsurface defect detection system for optical elements based on phase-contrast digital holography, employing the subsurface defect detection method for optical elements based on phase-contrast digital holography as described in any one of claims 1-6, characterized in that, include: Optical path module, core functional module and motion control module; The optical path module is used for the conversion between dark field scattering and phase-contrast microscopy imaging and for polarization-synchronized phase-shift image acquisition; The core functional modules are used to complete image conversion, data acquisition, and defect localization and characterization; The motion control module is used to precisely control the axial movement of the objective lens or the entire microscopic imaging system.

8. The optical element subsurface defect detection system based on phase-contrast digital holography according to claim 7, characterized in that: The optical path module includes an imaging optical path unit, a U-shaped phase contrast introduction optical path unit, and a ring-shaped illumination optical path unit.

9. The optical element subsurface defect detection system based on phase-contrast digital holography according to claim 8, characterized in that: The optical path unit includes a ring light source, a field stop, an amplitude mask, a first objective lens, a transparent stage, a second objective lens, and a tube lens; it is used for basic imaging of the optical element under test and provides the original light field for phase contrast introduction and data acquisition. The U-shaped phase-contrast optical path unit includes a first reflecting prism, a first Fourier transform lens, a phase mask, a second reflecting prism, a second Fourier transform lens, and a polarization camera; it introduces a phase-contrast effect for the modulation and acquisition of phase information, and completes synchronous four-step phase-shifting optical field processing. The annular illumination sub-optical path unit includes a first conical lens, a second conical lens, and a converging lens; the first and second conical lenses are placed symmetrically; they are used to generate a focused annular Bessel beam, forming a Kohler illumination pattern, and providing uniform and stable annular parallel illumination light for the entire system.

10. The optical element subsurface defect detection system based on phase-contrast digital holography according to claim 9, characterized in that: The object plane, image plane, and polarization camera target surface are conjugates to ensure accurate image transmission; the amplitude mask, the back focal plane of the first objective lens, and the phase mask are conjugates to accurately modulate the phase contrast effect.

Images