Wavefront recovery method and device for large-gradient interference fringes

Abstract

The invention discloses a wavefront recovery method and device for large-gradient interference fringes, and relates to the technical field of optical measurement. The method is used for solving the problems that an additional compensator is used for reducing the interference fringe density in existing optical element surface shape detection, operation is complex, and different compensation devices need to be arranged for different pieces to be detected. Comprising the following steps: mapping each pixel point included in a target interferogram into a unit circle to obtain a Zernike polynomial including target interferogram information; through linearly combining the Zernike polynomials and randomly generating an estimation coefficient of the Zernike polynomials, obtaining iteration wavefront information of each pixel point and an iteration interference pattern containing the estimation coefficient, the Zernike polynomials and a wave number; and determining the difference between the target interferogram and the iterative interferogram according to a loss function, if the difference does not meet a threshold value, updating the estimation coefficient according to a particle swarm optimization algorithm, and when the number of iterations reaches a set number of times or the loss function meets the threshold value, determining an update coefficient corresponding to the current iteration as an optimal coefficient, and obtaining recovered wavefront information according to the optimal coefficient and the obtained iteration wavefront information, wherein the recovered wavefront information is the wavefront information to be measured.

Description

Technical Field

[0001] The present invention relates to the field of optical measurement technology, and more particularly to a wavefront recovery method and device for large-gradient interference fringes. Background Art

[0002] Optical components are essential components of optical systems, and their processing and inspection technologies have a profound impact on the development of optical technology. Common optical components include planar, spherical, aspherical, and free-form surfaces. Each of these components possesses distinct optical properties and is widely used in aerospace, telescopes, and advanced weaponry. In these applications, the surface parameters of optical components play a decisive role in optical performance. To ensure that optical components achieve the expected performance in practical applications, high-precision surface inspection is essential. Currently, two common surface inspection methods are non-interference and interferometry. Non-interference, based on the principles of geometric optics, offers greater versatility but lower surface inspection accuracy. Interferometry, based on the principle of interference, achieves higher accuracy by splitting a single beam of light into two or more coherent beams, which are then passed through the surface of the optical component under test and a reference surface. The coherent beams are then superimposed to produce interference fringes. By processing the interference fringes, phase information is extracted and wavefronts are restored, thereby obtaining the corresponding surface information of the optical component under test. The effectiveness of phase extraction directly impacts surface inspection accuracy.

[0003] Currently, the main phase extraction methods include phase shifting and single-interference pattern-based phase recovery. Phase shifting methods collect interferograms at different phase shift states and use specific algorithms to obtain the phase distribution. Common phase shifting methods, such as the four-step phase shifting method, collect four interferograms with fixed phase differences, perform inverse trigonometric operations to obtain the wrapped phase, and then unwrap the wrapped phase to obtain the actual phase. While this method offers high phase extraction accuracy, it still has certain limitations. The phase shifting method primarily relies on a piezoelectric ceramic (PZT)-driven reference mirror to achieve phase shifting. However, the nonlinear displacement of the PZT can cause the actual phase difference to deviate from the theoretical value, introducing phase shift errors. Furthermore, system stability must be maintained during image acquisition, and phase recovery is significantly affected by the environment. Finally, image acquisition must comply with the Nyquist sampling theorem, which requires at least two pixels to represent each interference fringe. When the gradient of the interference fringes varies significantly (i.e., when the fringes are densely packed), the unwrapping process can result in discontinuities in the unwrapped phase, making it impossible to recover the true phase, thus affecting wavefront recovery. Phase retrieval methods based on single-frame interferograms only require the acquisition of one interference image to achieve phase extraction and are less affected by environmental factors. Among the phase retrieval methods based on single-frame interferograms, the Fourier transform method is the most widely used. The Fourier transform method converts the spatial interference fringes into the frequency domain, extracts the phase information through an inverse Fourier transform, and obtains the true phase distribution through phase unwrapping. Compared with the phase shifting algorithm, the Fourier transform method has better phase extraction capabilities for large-gradient interference fringes. However, the premise of using the Fourier transform method to extract phase is that the interference fringes have a single-directional spatial carrier frequency. The spatial frequency distribution of closed fringes contains frequency components in multiple directions, and a carrier needs to be added to eliminate the closed fringes. Therefore, although the Fourier transform method can extract phase from large-gradient interference fringes, it is not suitable for phase extraction of closed interference fringes.

[0004] Regarding the phase extraction method for large-gradient interference fringes, compared to the 2009 paper "Partial Nulling Lens for Universal Aspheric Surface Detection" by Liu Dong et al., which designed a partial compensation lens to compensate for most of the normal aberration, thereby reducing the density of the interference fringes and achieving phase extraction of large-gradient interference fringes, the paper required the design of an additional lens to address the difficulty of phase extraction of large-gradient interference fringes, and different partial compensation lenses needed to be designed for different test surfaces. The comparative document (CN113008148A) achieves large-gradient phase recovery by compensating for large-gradient wavefront aberrations using a spatial light modulator.

[0005] In summary, the above methods for optical component surface shape detection all achieve the recovery of large-gradient interference fringe wavefronts by reducing the interference fringe density through additional compensators. However, they have the problem of complex operation and the need to set different compensation devices for different test pieces. Summary of the Invention

[0006] The embodiments of the present invention provide a wavefront recovery method and device for large-gradient interference fringes, which are used to solve the problem that the existing optical element surface shape detection uses additional compensators to reduce the interference fringe density, which is complicated to operate and requires different compensation devices to be set for different test pieces.

[0007] An embodiment of the present invention provides a wavefront recovery method for large-gradient interference fringes, comprising:

[0008] A target interferogram containing large-gradient interference fringes is collected by a detector, and the target interferogram is normalized to obtain an initial normalized interferogram containing phase information to be measured;

[0009] Based on the phase to be measured and the wavefront information to be measured corresponding to the phase to be measured, the wavefront information is characterized by using orthogonal Zernike polynomials as basis functions to obtain a target interferogram including target coefficients of the Zernike polynomials, the Zernike polynomials, and the wavenumbers;

[0010] Mapping each pixel point included in the target interferogram to a unit circle to obtain a Zernike polynomial containing the target interferogram information; obtaining iterative wavefront information of each pixel point and an iterative interferogram containing the estimated coefficient, the Zernike polynomial, and the wave number by linearly combining the Zernike polynomial and the estimated coefficients of the randomly generated Zernike polynomial;

[0011] The difference between the target interference graph and the iterative interference graph is determined according to the loss function. If the difference does not meet the threshold, the estimated coefficient is updated according to the particle swarm optimization algorithm. When the number of iterations reaches the set number or the loss function meets the threshold, the updated coefficient corresponding to the current iteration is determined as the optimal coefficient. The restored wavefront information is obtained according to the optimal coefficient and the iterative wavefront information. The restored wavefront information is the wavefront information to be measured.

[0012] An embodiment of the present invention provides a wavefront recovery device for large-gradient interference fringes, comprising:

[0013] A first obtaining unit is configured to collect a target interferogram containing large-gradient interference fringes through a detector, and perform normalization processing on the target interferogram to obtain an initial normalized interferogram containing phase information to be measured;

[0014] a second obtaining unit, configured to characterize the wavefront information using orthogonal Zernike polynomials as basis functions based on the phase to be measured and the wavefront information to be measured corresponding to the phase to be measured, and obtain a target interferogram including target coefficients of the Zernike polynomials, the Zernike polynomials, and the wavenumber;

[0015] A third obtaining unit is configured to map each pixel point included in the target interferogram to a unit circle to obtain a Zernike polynomial containing the target interferogram information; obtain iterative wavefront information of each pixel point and an iterative interferogram containing the estimated coefficient, the Zernike polynomial, and the wave number by linearly combining the Zernike polynomial and the estimated coefficients of the randomly generated Zernike polynomial;

[0016] The fourth obtaining unit is used to determine the difference between the target interference pattern and the iterative interference pattern according to the loss function. If the difference does not meet the threshold, the estimated coefficient is updated according to the particle swarm optimization algorithm. When the number of iterations reaches the set number or the loss function meets the threshold, the update coefficient corresponding to the current iteration is determined as the optimal coefficient, and the restored wavefront information is obtained according to the optimal coefficient and the iterative wavefront information.

[0017] An embodiment of the present invention provides a computer device, which includes a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the processor executes any one of the above-mentioned wavefront recovery methods for large-gradient interference fringes.

[0018] An embodiment of the present invention provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes any one of the above-mentioned methods for recovering the wavefront of large-gradient interference fringes.

[0019] The embodiment of the present invention provides a wavefront recovery method and device for large gradient interference fringes. The method realizes wavefront recovery by taking a target interference pattern containing large gradient interference fringes as the target. The accuracy of wavefront recovery is independent of the density and fringe morphology of the target interference fringes. When there is no dark area in the target interference pattern, wavefront recovery of large gradient interference fringes can be realized, and the method has good versatility. The method can realize accurate recovery of wavefront by using only one target interference pattern, reduces the interference of environmental factors, solves the problem of inaccurate phase shift when solving phase by multi-step phase shift, and further improves the accuracy of wavefront recovery. Furthermore, the method uses Zernike polynomials to fit the wavefront to further characterize the phase to be measured. When the iterative interference pattern is compared with the actual collected target interference pattern, the wavefront can be accurately recovered. Figure 1 When the phase is consistent, the recovered wavefront information can be directly obtained without the need for phase unwrapping processing. This method solves the problem that the existing optical element surface shape detection requires an additional compensator to reduce the interference fringe density. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0021] Figure 1 A schematic flow chart of a wavefront recovery method for large-gradient interference fringes provided by an embodiment of the present invention;

[0022] Figure 2 A schematic diagram of a target interference pattern provided by an embodiment of the present invention;

[0023] Figure 3 A schematic diagram of a wavefront to be measured provided by an embodiment of the present invention;

[0024] Figure 4 A schematic diagram of a restored wavefront provided by an embodiment of the present invention;

[0025] Figure 5 A schematic diagram of part of the iterative process in a wavefront recovery method for large-gradient interference fringes provided by an embodiment of the present invention;

[0026] Figure 6 A schematic diagram of the fitting residual between the restored wavefront and the wavefront to be measured provided by an embodiment of the present invention;

[0027] Figure 7 A schematic structural diagram of a wavefront recovery device for large-gradient interference fringes provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0029] Figure 1 A schematic flow chart of a wavefront recovery method for large gradient interference fringes provided by an embodiment of the present invention; Figure 1 A wavefront recovery method for large gradient interference fringes provided by an embodiment of the present invention is described in detail. Figure 1 As shown, the method includes the following steps:

[0030] Step 101: collecting a target interferogram containing large-gradient interference fringes through a detector, and performing normalization processing on the target interferogram to obtain an initial normalized interferogram containing phase information to be measured;

[0031] Step 102, based on the phase to be measured and the wavefront information to be measured corresponding to the phase to be measured, using orthogonal Zernike polynomials as basis functions to characterize the wavefront information, and obtaining a target interferogram including target coefficients of the Zernike polynomials, the Zernike polynomials, and the wavenumbers;

[0032] Step 103: Map each pixel included in the target interferogram to a unit circle to obtain a Zernike polynomial containing the target interferogram information; obtain iterative wavefront information of each pixel point and an iterative interferogram including the estimated coefficient, the Zernike polynomial, and the wavenumber by linearly combining the Zernike polynomial and the estimated coefficients of the randomly generated Zernike polynomial;

[0033] Step 104: determine the difference between the target interference pattern and the iterative interference pattern according to the loss function; if the difference does not meet the threshold, update the estimated coefficient according to the particle swarm optimization algorithm; when the number of iterations reaches the set number or the loss function meets the threshold, determine the updated coefficient corresponding to the current iteration as the optimal coefficient; and obtain the restored wavefront information based on the optimal coefficient and the iterative wavefront information; the restored wavefront information is the same as the wavefront information to be measured.

[0034] In step 101, a target interferogram containing large-gradient interference fringes is collected based on a detector, wherein the target interferogram contains background light, modulated light, and phase information to be measured. The target interferogram is expressed by the following formula (1):

[0035]

[0036] Where I(x,y) represents the target interference pattern, A(x,y) represents the background light, and B(x,y) represents the modulated light. Indicates the phase to be measured.

[0037] In practical applications, the target interferogram containing large-gradient interference fringes collected by the detector usually contains multiple interference factors. In order to purify the phase information to be measured, background light elimination and modulated light intensity normalization processing are required so that the target interferogram only contains the phase information to be measured. The initial normalized interferogram is obtained, as shown below:

[0038]

[0039] in, represents the initial normalized interferogram, Indicates the phase to be measured.

[0040] like Figure 2When the interferogram shown above is used as the target interferogram, although there are no dark areas in the interferogram, the detector can detect the changes in brightness of the fringes. However, the edge fringe gradients of the target interferogram are large (the number of fringe periods per unit pixel is large). According to the Nyquist sampling theorem, the image sensor's sampling frequency for the interference fringes must be at least twice the spatial frequency of the fringe (that is, each fringe is represented by at least two pixels). When the edge fringe gradients are large, using a four-step phase shift to extract the phase may result in a phase jump exceeding π per pixel. According to the wrapped phase unwrapping principle, the phase changes between adjacent points must satisfy the Nyquist sampling theorem to achieve a continuous distribution of the unwrapped phase. Otherwise, phase discontinuity will occur, and the true phase cannot be recovered.

[0041] In step 102, according to the corresponding relationship between the phase to be measured and the wavefront information to be measured as shown below, the expression of the initial normalized interferogram can be rewritten to obtain the expression of the first normalized interferogram:

[0042]

[0043] in, represents the phase to be measured, k represents the wave number, k=2π / λ, W(x,y) represents the wavefront information to be measured, represents the first normalized interferogram.

[0044] In an embodiment of the present invention, the wavefront information to be measured W(x, y) can be represented by a set of orthogonal Zernike polynomials as basis functions, which are:

[0045]

[0046] Among them, W(x,y) represents the wavefront information to be measured, A i represents the target coefficient of the i-th Zernike polynomial, Z i (x,y) represents the i-th Zernike polynomial, n represents the number of terms in the Zernike polynomial, and i represents the i-th term in the Zernike polynomial.

[0047] Furthermore, substituting formula (5) into formula (4) yields the expression of the target interference pattern:

[0048]

[0049] Among them, I norm (x,y) represents the target interference pattern, A i represents the target coefficient of the i-th Zernike polynomial, Z i (x,y) represents the i-th Zernike polynomial, n represents the number of terms in the Zernike polynomial, and i represents the i-th term in the Zernike polynomial.

[0050] According to the target interference pattern expression (6), it can be seen that the target interference pattern is only related to the Zernike coefficient. Figure 2 When the interferogram shown in the figure is used as the target interferogram, although the detector can detect the light and dark changes of the fringes, the edge gradient of the interferogram is large, and the four-step phase shift method cannot accurately extract the phase information. Figure 2 The target interference pattern shown is obtained by fitting a set of Zernike coefficients using Zernike polynomials. Figure 3 After the wavefront information to be measured is shown, when the target interferogram corresponding to the wavefront information to be measured is obtained by formula (6), the wavefront peak-to-valley value is 27.5542λ and the root mean square value is 7.2819λ.

[0051] In step 103, the coordinates of the target interferogram collected by the detector typically correspond to the circular pupil region of the optical system (such as the clear aperture of a lens or reflector). Mapping the pixel coordinates to the unit circle (a normalized circle with a radius of 1) is done to unify the spatial reference of the wavefront description and facilitate mathematical modeling using Zernike polynomials defined within the unit circle.

[0052] Specifically, given a set of random Zernike estimated coefficients, which include 36 items, each pixel point included in the target interferogram is mapped to the unit circle to obtain the Zernike polynomial containing the target interferogram information. By linearly combining the Zernike polynomial containing the target interferogram information and the Zernike estimated coefficients, the iterative wavefront information shown below is obtained:

[0053]

[0054] in, represents the estimated coefficients of the random Zernike polynomial, W * (x,y) represents the iterative wavefront information, Z i (x,y) represents the i-th Zernike polynomial, which is an orthogonal function defined within the unit circle.

[0055] Furthermore, according to formula (4), the iterative interferogram of the estimated coefficients, Zernike polynomials and wave numbers can be obtained, and the specific expressions are as follows:

[0056]

[0057] in, represents the iterative interference pattern, represents the estimated coefficients of the Zernike polynomial, k represents the wave number, k=2π / λ, Z i(x,y) represents the i-th Zernike polynomial.

[0058] In the embodiment of the present invention, it can be seen from formula (8) and formula (6) that the iterative interferogram is determined only by the Zernike coefficients, and the estimated coefficients of the Zernike coefficients are continuously changed. Iterative interferogram obtained by iteration Continuously approaching the target light intensity diagram I norm The process of iterative interference pattern continuously approaching is the process in which the estimated Zernike coefficients set as variables in the iteration continuously approach the Zernike target coefficients that characterize the actual wavefront, thereby recovering the wavefront information.

[0059] In step 104, in an embodiment of the present invention, a structural similarity index is used as an indicator to judge whether the iterative interferogram is consistent with the target interferogram, and the following loss function is established:

[0060]

[0061] Among them, J represents the loss function value, SSIM represents the structural similarity index, which takes into account the brightness, contrast and structural information of the image. The value range is [0,1]. The larger the value, the closer the iterative interferogram is to the target interferogram.

[0062] In an embodiment of the present invention, after the loss function between the iterative interferogram and the target interferogram is obtained according to formula (9), a set of Zernike update coefficients is obtained by particle swarm optimization algorithm so that the loss function value J is minimized, that is, the loss function is less than a set threshold or reaches the number of iterations; the obtained set of Zernike update coefficients maximizes the structural similarity index SSIM of the iterative interferogram and the target interferogram. The smaller the loss function value, the closer the two images are, which further indicates that the Zernike update coefficients obtained by iteration are closer to the Zernike target coefficients. By fitting the Zernike polynomials to the Zernike update coefficients obtained, the recovered wavefront information can be obtained.

[0063] Figure 4 The wavefront recovery process of the embodiment of the present invention is as follows: Figure 5As shown, (a) is the wavefront obtained after the randomly given Zernike coefficients are fitted by the Zernike polynomials, (e) is the interference fringe pattern corresponding to the wavefront, (b) is a wavefront pattern in the iterative process, (f) is the interference fringe pattern corresponding to the wavefront (b), (c) is the wavefront pattern restored by the method of the present invention, (g) is the interference fringe pattern corresponding to the wavefront (c), (d) is the target wavefront given in the embodiment, (h) is the interference pattern corresponding to the target wavefront (d), and is the target interference pattern given in the embodiment. Figure 5 It can be seen that from the interference pattern (e) to the interference pattern (g), it is constantly approaching the target interference pattern (h). Correspondingly, from the initial wavefront (a) to the wavefront (c), it is also gradually approaching the target wavefront (d) given in the embodiment, which more clearly illustrates that the continuous approach of the interference pattern is the continuous approach of the wavefront. Figure 5 The process of wavefront recovery is given. Figure 6 The fitting residual graph between the restored wavefront information and the measured wavefront information provided by the embodiment of the present invention has a peak-to-valley value of 4.7791E-07λ and a root mean square value of 6.3161E-08λ, indicating that the wavefront restoration accuracy is relatively high.

[0064] It should be noted that, in the embodiment of the present invention, the particle swarm optimization algorithm updates the particle position and particle velocity according to the following formula:

[0065]

[0066] in, represents the velocity of the oth particle at iteration t+1, ω t represents the inertia factor of the t-th iteration, represents the velocity of the oth particle at the tth iteration, c1 and c2 represent acceleration factors, r1 and r2 represent random numbers in the interval [0,1], represents the optimal position of the o-th particle after t iterations, represents the position of the o-th particle at the tth iteration, represents the optimal position of the particle swarm, o=1,2,…,n represents the size of the particle swarm, and t represents the number of iterations of the algorithm; represents the position of the oth particle at iteration t+1, ω max Indicates the maximum value of the inertia factor, ω min Indicates the minimum value of the inertia factor, t max Indicates the maximum number of algorithm iterations.

[0067] In summary, an embodiment of the present invention provides a wavefront recovery method for large-gradient interference fringes, comprising: collecting a target interference pattern containing large-gradient interference fringes through a detector, normalizing the target interference pattern to obtain an initial normalized interference pattern containing information of a phase to be measured; based on the phase to be measured and the wavefront information to be measured corresponding to the phase to be measured, characterizing the wavefront information with orthogonal Zernike polynomials as basis functions to obtain a target interference pattern containing target coefficients of Zernike polynomials, Zernike polynomials, and wavenumbers; mapping each pixel point included in the target interference pattern to a unit circle to obtain a Zernike polynomial containing information of the target interference pattern; term; by linearly combining the Zernike polynomials and the estimated coefficients of the randomly generated Zernike polynomials, the iterative wavefront information of each pixel point and the iterative interferogram including the estimated coefficients, Zernike polynomials and wavenumbers are obtained; the difference between the target interferogram and the iterative interferogram is determined according to the loss function, if the difference does not meet the threshold, the estimated coefficients are updated according to the particle swarm optimization algorithm, when the number of iterations reaches the set number or the loss function meets the threshold, the updated coefficient corresponding to the current iteration is determined as the optimal coefficient, and the restored wavefront information is obtained according to the optimal coefficient and the iterative wavefront information, and the restored wavefront information is equal to the wavefront information to be measured. This method realizes wavefront recovery by taking the target interference pattern containing large gradient interference fringes as the target. The accuracy of the recovered wavefront information is independent of the density and fringe shape of the target interference fringes. When there is no dark area in the target interference pattern, the wavefront recovery of large gradient interference fringes can be realized, which has good versatility. This method can realize accurate wavefront recovery by using only one target interference pattern, which reduces the interference of environmental factors, solves the problem of inaccurate phase shift when using multi-step phase shift to solve the phase, and further improves the accuracy of wavefront recovery. In addition, this method uses Zernike polynomials to fit the wavefront to further characterize the phase to be measured. When the iterative interference pattern is consistent with the actual collected target interference pattern, the wavefront can be accurately recovered. Figure 1 When the phase is consistent, the recovered wavefront information can be directly obtained without the need for phase unwrapping processing. This method solves the problem that the existing optical element surface shape detection requires an additional compensator to reduce the interference fringe density.

[0068] Based on the same inventive concept, an embodiment of the present invention provides a wavefront recovery device for large gradient interference fringes. Since the principle of the device in solving the technical problem is similar to a wavefront recovery method for large gradient interference fringes, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0069] like Figure 7 As shown, the apparatus includes a first obtaining unit 701 , a second obtaining unit 702 , a third obtaining unit 703 and a fourth obtaining unit 704 .

[0070] A first obtaining unit 701 is configured to collect a target interferogram containing large-gradient interference fringes through a detector, and perform normalization processing on the target interferogram to obtain an initial normalized interferogram containing phase information to be measured;

[0071] A second obtaining unit 702 is configured to characterize the wavefront information using orthogonal Zernike polynomials as basis functions based on the phase to be measured and the wavefront information to be measured corresponding to the phase to be measured, and obtain a target interferogram including target coefficients of the Zernike polynomials, the Zernike polynomials, and the wavenumber;

[0072] The third obtaining unit 703 is configured to map each pixel point included in the target interferogram to a unit circle to obtain a Zernike polynomial containing the target interferogram information; obtain iterative wavefront information of each pixel point and an iterative interferogram containing the estimated coefficient, the Zernike polynomial, and the wave number by linearly combining the Zernike polynomial and the estimated coefficients of the randomly generated Zernike polynomial;

[0073] The fourth obtaining unit 704 is used to determine the difference between the target interference pattern and the iterative interference pattern according to the loss function. If the difference does not meet the threshold, the estimated coefficient is updated according to the particle swarm optimization algorithm. When the number of iterations reaches the set number or the loss function meets the threshold, the updated coefficient corresponding to the current iteration is determined as the optimal coefficient, and the restored wavefront information is obtained based on the optimal coefficient and the iterative wavefront information.

[0074] It should be understood that the units included in the aforementioned device for recovering large-gradient interference fringes are merely logical divisions based on the functions implemented by the device. In actual applications, these units can be combined or separated. Furthermore, the functions implemented by the device for recovering large-gradient interference fringes provided in this embodiment correspond exactly to the wavefront recovery method for recovering large-gradient interference fringes provided in the aforementioned embodiment. A more detailed processing flow implemented by the device has been described in detail in the aforementioned method embodiment 1 and will not be described in detail here.

[0075] Another embodiment of the present invention also provides a computer device, which includes: a processor and a memory; the memory is used to store computer program code, and the computer program code includes computer instructions; when the processor executes the computer instructions, the electronic device executes the various steps of the wavefront recovery method for large gradient interference fringes shown in the above method embodiment.

[0076] Another embodiment of the present invention also provides a computer-readable storage medium, which stores computer instructions. When the computer instructions are executed on a computer device, the computer device executes each step of the wavefront recovery method for large gradient interference fringes shown in the above method embodiment.

[0077] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A wavefront recovery method for large gradient interference fringes, characterized in that: include: A target interferogram containing large-gradient interference fringes is collected by a detector, and the target interferogram is normalized to obtain an initial normalized interferogram containing phase information to be measured; Based on the phase to be measured and the wavefront information to be measured corresponding to the phase to be measured, the wavefront information is characterized by using orthogonal Zernike polynomials as basis functions to obtain a target interferogram including target coefficients of the Zernike polynomials, the Zernike polynomials, and the wavenumbers; Mapping each pixel point included in the target interferogram to a unit circle to obtain a Zernike polynomial containing the target interferogram information; obtaining iterative wavefront information of each pixel point and an iterative interferogram containing the estimated coefficient, the Zernike polynomial, and the wave number by linearly combining the Zernike polynomial and the estimated coefficients of the randomly generated Zernike polynomial; The difference between the target interference graph and the iterative interference graph is determined according to the loss function. If the difference does not meet the threshold, the estimated coefficient is updated according to the particle swarm optimization algorithm. When the number of iterations reaches the set number or the loss function meets the threshold, the updated coefficient corresponding to the current iteration is determined as the optimal coefficient. The restored wavefront information is obtained according to the optimal coefficient and the iterative wavefront information. The restored wavefront information is the wavefront information to be measured.

2. The method according to claim 1, wherein The expression of the initial normalized interferogram is as follows: The expression of the target interference pattern is as follows: in, represents the initial normalized interferogram, represents the phase to be measured, k represents the wave number, k=2π / λ, W(x,y) represents the wavefront information to be measured, I norm (x,y) represents the target interference pattern, A i represents the target coefficient of the i-th Zernike polynomial, Z i (x,y) represents the i-th Zernike polynomial, n represents the number of terms in the Zernike polynomial, and i represents the i-th term in the Zernike polynomial.

3. The method according to claim 1, wherein Mapping each pixel point included in the target interferogram to a unit circle to obtain a Zernike polynomial containing the target interferogram information; and obtaining iterative wavefront information of each pixel point by linearly combining the Zernike polynomial and estimated coefficients of a randomly generated Zernike polynomial, specifically comprising: Randomly generate estimated coefficients of 36 Zernike polynomials, map each pixel point included in the target interferogram to a unit circle, and determine the values ​​of the 36 Zernike polynomials at the pixel point; The Zernike polynomial and the estimated coefficients of 36 randomly generated Zernike polynomials are used to obtain the iterative wavefront information of each pixel point through the following formula: in, represents the estimated coefficients of the Zernike polynomials, represents the estimated phase, W * (x, y) represents the iterative wavefront information, k represents the wave number, k = 2π / λ, Z i (x,y) represents the i-th Zernike polynomial.

4. The method according to claim 1, wherein The iterative interferogram including estimated coefficients, Zernike polynomials and wave numbers is shown below: The loss function is as follows: in, represents the iterative interference pattern, represents the estimated coefficients of the Zernike polynomial, k represents the wave number, k=2π / λ, Z i (x,y) represents the i-th Zernike polynomial, J represents the loss function value, SSIM represents the structural similarity index, I norm (x,y) represents the normalized interferogram.

5. A wavefront recovery device for large gradient interference fringes, characterized in that: include: A first obtaining unit is configured to collect a target interferogram containing large-gradient interference fringes through a detector, and perform normalization processing on the target interferogram to obtain an initial normalized interferogram containing phase information to be measured; a second obtaining unit, configured to characterize the wavefront information using orthogonal Zernike polynomials as basis functions based on the phase to be measured and the wavefront information to be measured corresponding to the phase to be measured, and obtain a target interferogram including target coefficients of the Zernike polynomials, the Zernike polynomials, and the wavenumber; A third obtaining unit is configured to map each pixel point included in the target interference pattern to a unit circle to obtain a Zernike polynomial containing information of the target interference pattern; Obtaining iterative wavefront information for each pixel and an iterative interferogram including the estimated coefficients, the Zernike polynomial, and the wave number by linearly combining the Zernike polynomial and the estimated coefficients of the randomly generated Zernike polynomial; The fourth obtaining unit is used to determine the difference between the target interference pattern and the iterative interference pattern according to the loss function. If the difference does not meet the threshold, the estimated coefficient is updated according to the particle swarm optimization algorithm. When the number of iterations reaches the set number or the loss function meets the threshold, the update coefficient corresponding to the current iteration is determined as the optimal coefficient, and the restored wavefront information is obtained according to the optimal coefficient and the iterative wavefront information.

6. A computer device, characterized in that: The computer device includes a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the wavefront recovery method for large-gradient interference fringes as described in any one of claims 1-4.

7. A computer-readable storage medium, characterized in that A computer program is stored, and when the computer program is executed by a processor, the processor is caused to execute the wavefront recovery method for large-gradient interference fringes as claimed in any one of claims 1 to 4.

Images