Integrated method and system for analyzing a single-frame interferogram for surface shape measurement and defect positioning

Abstract

The integrated method and system for analyzing single-frame interferogram surface shape measurement and defect positioning can complete surface shape measurement and defect full-field positioning simultaneously by using a surface shape measurement interferometer, shortens the detection process, accelerates the detection speed, and can realize shape and property integrated detection of ultra-precise optical elements. The method comprises the following steps: collecting a single-frame interferogram by using a surface shape measurement interferometer; obtaining phase information carried by the interferogram; obtaining each Zernike coefficient of the phase information carried by the interferogram; optimizing by using the optimization function of an optical design software until the system wave aberration obtained by a virtual interferometer is consistent with the phase information carried by the interferogram obtained by actual measurement; extracting each Zernike coefficient of the measured surface, reconstructing the surface shape of the measured surface; removing low-frequency terms; reconstructing concentric ring features in the phase information to obtain the center pixel coordinates of the feature recognition frame; and obtaining three-dimensional coordinates of the defect on the surface of the measured surface.

Description

Technical Field

[0001] This invention relates to the technical field of optical interferometry instruments, and more particularly to an integrated method for surface shape measurement and defect localization in analyzing single-frame interferograms, as well as an integrated system for surface shape measurement and defect localization in analyzing single-frame interferograms. Background Technology

[0002] Ultra-precision optical components operating under extreme conditions such as high-energy lasers and space radiation not only require high surface accuracy but also low surface defects and high load capacity. Shape-property integrated measurement technology is the key to solving problems such as multi-dimensional shape-property constraint measurement and achieving high-efficiency, controllable and predictable manufacturing. It is of great value for improving the shape-property integrated manufacturing and inspection of ultra-precision optical components operating under extreme conditions.

[0003] Currently, interferometry is widely used for the final inspection of the surface shape of ultra-precision optical components due to its high precision and speed. However, interferometry is susceptible to environmental vibrations, especially under long optical path conditions where environmental control requirements are stringent. Furthermore, the output of surface shape measurement interferometry only shows the surface shape and cannot reflect defects. In fact, surface defects exceeding a certain size threshold will diffract and modulate reflected light waves, appearing as diffraction rings in the interferogram. This feature could potentially be used for preliminary defect localization, facilitating subsequent observation. However, current methods primarily use denoising algorithms to eliminate the influence of surface shape measurements on diffraction caused by defects, failing to fully explore its application potential.

[0004] Microscopic defects mainly refer to defect indicators on the surface, subsurface, and within the microstructure. Surface defects include visible defects such as scratches, pitting, chipping, contaminants, and bubbles, as well as defects that are not visible to the naked eye, such as absorptive defects. Currently, visible surface defects are mostly observed visually, which is inefficient and relies heavily on experience. Microscopes can be used for precise observation, but the scanning measurement efficiency is even lower due to the limited field of view. After observing a defect, its location on the surface needs to be measured for subsequent processing and repair; direct full-field localization is not possible, resulting in low efficiency.

[0005] In summary, existing methods for measuring the surface of ultra-precision optical components are all form-property separation, which not only affects the detection efficiency and increases the detection cost, but also makes it impossible to directly establish a correlation between form and property. This has become a bottleneck problem that restricts the joint constraints of form and property on the manufacturing, detection and evaluation of optical components. Summary of the Invention

[0006] To overcome the shortcomings of existing technologies, the technical problem to be solved by this invention is to provide an integrated method for surface shape measurement and defect localization in analyzing single-frame interferograms. This method can simultaneously complete surface shape measurement and full-field defect localization using only a surface shape measurement interferometer, shortening the inspection process, accelerating the inspection speed, and enabling integrated shape and property inspection of ultra-precision optical components.

[0007] The technical solution of this invention is: an integrated method for surface shape measurement and defect localization in analyzing single-frame interferograms, comprising the following steps:

[0008] (1) A single frame of interferogram was acquired using a surface shape measurement interferometer;

[0009] (2) The phase information carried by the interferogram is obtained by using the single-frame interferogram phase decomposition method;

[0010] (3) Perform Zernike fitting on the phase information carried by the interferogram to obtain the Zernike coefficients of the phase information carried by the interferogram.

[0011] (4) Using the parameters of the surface shape measurement interferometer, compensating mirror, and the surface under test, a virtual interferometer system is established in the optical design software. The Zernike coefficients of the phase information carried by the interferogram are used as the optimization targets of the system wave aberration, and the Zernike coefficients of the surface under test are used as the optimization variables. The optimization function of the optical design software is used to optimize until the system wave aberration obtained by the virtual interferometer is consistent with the phase information carried by the interferogram obtained by actual measurement.

[0012] (5) Extract the Zernike coefficients of the measured surface and reconstruct the shape of the measured surface;

[0013] (6) Remove the low-frequency terms from the Zernike coefficients of each term carrying phase information in the interferogram, and reconstruct the phase information using the remaining high-frequency terms;

[0014] (7) Use neural networks to identify high-frequency terms and reconstruct concentric ring features in phase information to obtain the center pixel coordinates of the feature recognition box;

[0015] (8) Bring the center pixel coordinates of the feature recognition box back to the virtual interferometer system, calculate the three-dimensional coordinates on the test surface corresponding to the center pixel coordinates of the feature recognition box through ray tracing, obtain the three-dimensional coordinates of the defect on the test surface, and complete the positioning.

[0016] This invention analyzes the phase of a single-frame interferogram, optimizes the surface shape under test using a virtual interferometer, and uses Zernike analysis and neural networks to identify the concentric ring-shaped characteristic phase caused by defects, thereby completing the full-field localization of defects. Compared with traditional interferometry, it makes full use of the high-frequency phase information introduced by defects, and can perform full-field localization of defects without the need for other additional devices. It has a fast measurement speed and can integrate the measurement of optical element surface shape and the localization of surface micro-defects.

[0017] It also provides an integrated system for surface shape measurement and defect location analysis of single-frame interferograms, which includes: surface shape measurement interferometer (1), compensation mirror (2), measured surface (3), phase decomposition module, fitting module, surface shape measurement module, and defect location module;

[0018] The phase decomposition module uses a single-frame interferogram phase decomposition algorithm to obtain the phase information carried by a single-frame interferogram.

[0019] The fitting module is configured to perform Zernike fitting on the phase information of the interferogram and obtain the Zernike coefficients of various system wave aberrations.

[0020] The surface shape measurement module utilizes the parameters of a surface shape measurement interferometer, a compensating mirror, and the surface under test to establish a virtual interferometer system in optical design software. The Zernike coefficients of the phase information carried by the interferogram are used as optimization targets for the system wave aberrations, and the Zernike coefficients of the surface under test are used as optimization variables. The optimization function of the optical design software is used to optimize until the system wave aberrations acquired by the virtual interferometer match the phase information carried by the interferogram acquired through actual measurement. The Zernike coefficients of the surface under test are then extracted, and the surface shape is reconstructed.

[0021] The defect localization module removes the low-frequency terms from the Zernike coefficients of the phase information carried by the interferogram and reconstructs the phase information using the remaining high-frequency terms; it then uses a neural network to identify the concentric ring features in the reconstructed phase information from the high-frequency terms, thus obtaining the center pixel coordinates of the feature recognition box.

[0022] The center pixel coordinates of the feature recognition box are brought back to the virtual interferometer system. The three-dimensional coordinates corresponding to the center pixel coordinates of the feature recognition box on the test surface are calculated by ray tracing. The three-dimensional coordinates of the defect on the test surface are obtained, and the localization is completed. Attached Figure Description

[0023] Figure 1 This is a flowchart of the integrated method for surface shape measurement and defect location based on the analysis of a single-frame interferogram according to the present invention.

[0024] Figure 2 The obtained single-frame interferogram.

[0025] Figure 3 It is the surface shape measurement result obtained from reconstruction.

[0026] Figure 4 It is the phase information reconstructed using the remaining high-frequency terms after removing the low-frequency terms.

[0027] Figure 5 This is the result of locating the pixel coordinates of the defect center.

[0028] Figure 6 This is a schematic diagram of the hardware structure of an integrated system for measuring the surface shape of optical elements and locating microscopic defects on a single frame interferogram according to the present invention.

[0029] Among them, 1 is the surface shape measurement interferometer, 2 is the compensation mirror, and 3 is the surface being measured. Detailed Implementation

[0030] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0031] It should be noted that the term "comprising" and any variations thereof in the specification, claims and accompanying drawings of this invention are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is 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.

[0032] like Figure 1 As shown, this integrated method for analyzing surface shape measurement and defect localization using a single-frame interferogram includes the following steps:

[0033] (1) A single frame of interferogram was acquired using a surface shape measurement interferometer;

[0034] (2) The phase information carried by the interferogram is obtained by using the single-frame interferogram phase decomposition method;

[0035] (3) Perform Zernike fitting on the phase information carried by the interferogram to obtain the Zernike coefficients of the phase information carried by the interferogram.

[0036] (4) Using the parameters of the surface shape measurement interferometer, compensating mirror, and the surface under test, a virtual interferometer system is established in the optical design software. The Zernike coefficients of the phase information carried by the interferogram are used as the optimization targets of the system wave aberration, and the Zernike coefficients of the surface under test are used as the optimization variables. The optimization function of the optical design software is used to optimize until the system wave aberration obtained by the virtual interferometer is consistent with the phase information carried by the interferogram obtained by actual measurement.

[0037] (5) Extract the Zernike coefficients of the measured surface and reconstruct the shape of the measured surface;

[0038] (6) Remove the low-frequency terms from the Zernike coefficients of each term carrying phase information in the interferogram, and reconstruct the phase information using the remaining high-frequency terms;

[0039] (7) Use neural networks to identify high-frequency terms and reconstruct concentric ring features in phase information to obtain the center pixel coordinates of the feature recognition box;

[0040] (8) Bring the center pixel coordinates of the feature recognition box back to the virtual interferometer system, calculate the three-dimensional coordinates on the test surface corresponding to the center pixel coordinates of the feature recognition box through ray tracing, obtain the three-dimensional coordinates of the defect on the test surface, and complete the positioning.

[0041] This invention analyzes the phase of a single-frame interferogram, optimizes the surface shape under test using a virtual interferometer, and uses Zernike analysis and neural networks to identify the concentric ring-shaped characteristic phase caused by defects, thereby completing the full-field localization of defects. Compared with traditional interferometry, it makes full use of the high-frequency phase information introduced by defects, and can perform full-field localization of defects without the need for other additional devices. It has a fast measurement speed and can integrate the measurement of optical element surface shape and the localization of surface micro-defects.

[0042] Preferably, the single-frame interferometric phase method in step (2) includes, but is not limited to, Fourier carrier wave method, digital moiré phase shift method, and neural network method.

[0043] Preferably, in step (3), the number of terms fitted by Zernike is more than 37.

[0044] Preferably, in step (4), when the surface being measured is a plane, there is no compensation mirror.

[0045] Preferably, the low-frequency terms among the Zernike coefficients removed in step (6) include at least the first 10 terms. If the concentric ring feature is still not obvious in the remaining high-frequency terms, the first 15 or the first 20 terms are removed.

[0046] Preferably, in step (7), the neural network is, but is not limited to, the YOLO target recognition network.

[0047] Preferably, in step (7), the neural network dataset is generated through simulation, wherein the high-frequency phase information is generated by superimposing Gaussian noise using the peaks function, the concentric ring features are generated using the sinc function, and the center pixels of the concentric ring features are randomly distributed.

[0048] The present invention also provides an integrated system for surface shape measurement and defect location analysis of single-frame interferograms. The system includes: a surface shape measurement interferometer (1), a compensation mirror (2), a surface to be measured (3), a phase decomposition module, a fitting module, a surface shape measurement module, and a defect location module.

[0049] The phase decomposition module uses a single-frame interferogram phase decomposition algorithm to obtain the phase information carried by a single-frame interferogram.

[0050] The fitting module is configured to perform Zernike fitting on the phase information of the interferogram and obtain the Zernike coefficients of various system wave aberrations.

[0051] The surface shape measurement module utilizes the parameters of a surface shape measurement interferometer, a compensating mirror, and the surface under test to establish a virtual interferometer system in optical design software. The Zernike coefficients of the phase information carried by the interferogram are used as optimization targets for the system wave aberrations, and the Zernike coefficients of the surface under test are used as optimization variables. The optimization function of the optical design software is used to optimize until the system wave aberrations acquired by the virtual interferometer match the phase information carried by the interferogram acquired through actual measurement. The Zernike coefficients of the surface under test are then extracted, and the surface shape is reconstructed.

[0052] The defect localization module removes the low-frequency terms from the Zernike coefficients of the phase information carried by the interferogram and reconstructs the phase information using the remaining high-frequency terms; it then uses a neural network to identify the concentric ring features in the reconstructed phase information from the high-frequency terms, thus obtaining the center pixel coordinates of the feature recognition box.

[0053] The center pixel coordinates of the feature recognition box are brought back to the virtual interferometer system. The three-dimensional coordinates corresponding to the center pixel coordinates of the feature recognition box on the test surface are calculated by ray tracing. The three-dimensional coordinates of the defect on the test surface are obtained, and the localization is completed.

[0054] The following details specific embodiments of the present invention.

[0055] Step 1: Acquire a single-frame interferogram using a surface shape measurement interferometer, such as... Figure 2 As shown;

[0056] Step 2: Use the single-frame interferogram phase decomposition method to obtain the phase information carried by the interferogram;

[0057] Step 3: Perform Zernike fitting on the phase information carried by the interferogram to obtain the Zernike coefficients of each phase information carried by the interferogram;

[0058] Step 4: Using the parameters of the surface measurement interferometer and the surface under test, establish a virtual interferometer system in the optical design software. Use the Zernike coefficients of the phase information carried by the interferogram as the optimization target of the system wave aberration, and use the Zernike coefficient of the surface under test as the optimization variable. Optimize using the optimization function of the optical design software until the system wave aberration obtained by the virtual interferometer is consistent with the phase information carried by the interferogram obtained by actual measurement.

[0059] Step 5: Extract the Zernike coefficients of the measured surface and reconstruct the surface shape, such as... Figure 3 As shown;

[0060] Step Six: Remove the low-frequency terms from the Zernike coefficients carrying phase information in the interferogram, and reconstruct the phase information using the remaining high-frequency terms, such as... Figure 4 As shown;

[0061] Step 7: Use a neural network to identify high-frequency terms and reconstruct the concentric ring features in the phase information to obtain the center pixel coordinates of the feature recognition box, such as... Figure 5 As shown;

[0062] Step 8: Bring the center pixel coordinates of the feature recognition box back to the virtual interferometer system, and calculate the three-dimensional coordinates on the test surface corresponding to the center pixel coordinates of the feature recognition box through ray tracing. This will obtain the three-dimensional coordinates of the defect on the test surface and complete the localization.

[0063] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention shall still fall within the protection scope of the present invention.

Claims

1. An integrated method for surface shape measurement and defect localization of single-frame interferograms, characterized by: It includes the following steps: (1) A single frame of interferogram was acquired using a surface shape measurement interferometer; (2) The phase information carried by the interferogram is obtained by using the single-frame interferogram phase decomposition method; (3) Perform Zernike fitting on the phase information carried by the interferogram to obtain the Zernike coefficients of the phase information carried by the interferogram. (4) Using the parameters of the surface shape measurement interferometer, compensating mirror, and the surface under test, a virtual interferometer system is established in the optical design software. The Zernike coefficients of the phase information carried by the interferogram are used as the optimization targets of the system wave aberration, and the Zernike coefficients of the surface under test are used as the optimization variables. The optimization function of the optical design software is used to optimize until the system wave aberration obtained by the virtual interferometer is consistent with the phase information carried by the interferogram obtained by actual measurement. (5) Extract the Zernike coefficients of the measured surface and reconstruct the shape of the measured surface; (6) Remove the low-frequency terms from the Zernike coefficients of each term carrying phase information in the interferogram, and reconstruct the phase information using the remaining high-frequency terms; (7) Use neural networks to identify high-frequency terms and reconstruct concentric ring features in phase information to obtain the center pixel coordinates of the feature recognition box; (8) Bring the center pixel coordinates of the feature recognition box back to the virtual interferometer system, calculate the three-dimensional coordinates on the test surface corresponding to the center pixel coordinates of the feature recognition box through ray tracing, obtain the three-dimensional coordinates of the defect on the test surface, and complete the positioning.

2. The integrated method for surface shape measurement and defect localization of single-frame interferograms according to claim 1, characterized in that: The single-frame interferometric phase method in step (2) includes: Fourier carrier method, digital moiré phase shift method, and neural network method.

3. The integrated method for surface shape measurement and defect localization of single-frame interferograms according to claim 2, characterized in that: In step (3), the number of terms fitted by Zernike is more than 37.

4. The integrated method for surface shape measurement and defect localization of single-frame interferograms according to claim 3, characterized in that: In step (4), when the surface being measured is a plane, there is no compensation mirror.

5. The integrated method for surface shape measurement and defect localization of single-frame interferograms according to claim 4, characterized in that: In step (6), the low-frequency terms among the Zernike coefficients removed include at least the first 10 terms. If the concentric ring feature is still not obvious in the remaining high-frequency terms, the first 15 or the first 20 terms are removed.

6. The integrated method for surface shape measurement and defect localization of single-frame interferograms according to claim 5, characterized in that: In step (7), the neural network includes the YOLO target recognition network.

7. The integrated method for surface shape measurement and defect localization of single-frame interferograms according to claim 6, characterized in that: In step (7), the neural network dataset is generated through simulation, wherein the high-frequency phase information is generated by superimposing Gaussian noise using the peaks function, the concentric ring features are generated using the sinc function, and the center pixels of the concentric ring features are randomly distributed.

8. An integrated system for surface shape measurement and defect localization based on single-frame interferogram analysis, characterized by: It includes: Surface shape measurement interferometer (1), compensation mirror (2), surface to be measured (3), phase resolution module, fitting module, surface shape measurement module, defect location module; The phase decomposition module uses a single-frame interferogram phase decomposition algorithm to obtain the phase information carried by a single-frame interferogram. The fitting module is configured to perform Zernike fitting on the phase information of the interferogram and obtain the Zernike coefficients of various system wave aberrations. The surface shape measurement module utilizes the parameters of a surface shape measurement interferometer, a compensating mirror, and the surface under test to establish a virtual interferometer system in optical design software. The Zernike coefficients of the phase information carried by the interferogram are used as optimization targets for the system wave aberrations, and the Zernike coefficients of the surface under test are used as optimization variables. The optimization function of the optical design software is used to optimize until the system wave aberrations acquired by the virtual interferometer match the phase information carried by the interferogram acquired through actual measurement. The Zernike coefficients of the surface under test are then extracted, and the surface shape is reconstructed. The defect localization module removes the low-frequency terms from the Zernike coefficients of the phase information carried by the interferogram and reconstructs the phase information using the remaining high-frequency terms. It then uses a neural network to identify the concentric ring features in the reconstructed phase information from the high-frequency terms, obtaining the center pixel coordinates of the feature recognition box. The center pixel coordinates of the feature recognition box are then brought back to the virtual interferometer system, and the three-dimensional coordinates corresponding to the center pixel coordinates of the feature recognition box on the tested surface are calculated by ray tracing. This yields the three-dimensional coordinates of the defect on the tested surface, completing the localization.

Images