A diffraction imaging system and FZP imaging model optimization and image restoration method

Abstract

The application discloses a kind of diffraction imaging system and FZP imaging model optimization and image restoration method, draw lessons from solid physics annealing principle, FZP imaging model parameter is optimized by simulated annealing algorithm, obtains FZP non-imaging order light equivalent point spread function and diffraction efficiency, realize FZP imaging model full parameter construction, based on model realizes the fast image restoration of FZP imaging result, improves FZP diffraction imaging image quality.The initial value of PSF and FZP diffraction efficiency measured by experiment is used as condition in the application, and the optimal parameters of FZP imaging model are iteratively optimized by simulated annealing algorithm, to realize the restoration of diffraction imaging image, so as to improve the image quality of imaging.The image restoration method disclosed in the application has high reliability, simple realization, high real-time performance and strong robustness.The search strategy of the method is not easy to fall into local optimum, and is more suitable for real-time image restoration application of large-aperture space-based diffraction imaging system.

Description

Technical Field

[0001] This invention belongs to the field of thin-film diffraction imaging technology, specifically relating to a diffraction imaging system and an FZP imaging model optimization and image restoration method. Background Technology

[0002] Simulated annealing is derived from the principle of solid-state annealing, which involves slowly cooling a high-temperature object until it reaches a stable state. At high temperatures, the internal energy of particles is high, allowing them to move freely within the solid. As the temperature gradually cools, the internal energy decreases, reducing their activity and shrinking their range of movement. When the solid has cooled completely or reached equilibrium, the particle positions are fixed, allowing the acquisition of the optimal solution. Its main idea is to perform a random walk within the search interval and then use the Metropolis sampling criterion to gradually converge the random walk to a local optimum. Temperature is a crucial parameter in the Metropolis algorithm, determining the speed of convergence. This algorithm has a wide range of applications, high reliability in finding the global optimum, simplicity, and ease of implementation. The search strategy helps avoid getting trapped in local optima, improving the reliability of finding the global optimum and exhibiting strong robustness.

[0003] Thin-film diffraction imaging technology, using FZP as the primary imaging mirror, boasts advantages such as ultra-large aperture, lightweight design, relaxed tolerances, small size, ease of deployment and folding, and low cost, making it an important development direction for large-aperture space-based telescopes. However, related research indicates that the imaging performance of thin-film diffraction imaging systems still lags behind space applications. Specifically, multi-order diffraction by the FZP diffraction element leads to poor image quality, and wavefront aberrations in the imaging beam cause a decrease in image resolution—a problem that urgently needs to be addressed. Non-imaging-order diffracted light becomes strong background noise, severely reducing the signal-to-noise ratio and hindering the realization of improved imaging system resolution through wavefront aberration correction.

[0004] The FZP consists of a series of stepped concentric ring micro / nano substructures. These stepped micro / nano substructures modulate the phase of the incident beam. While aiming for focused imaging, they inevitably introduce non-designed order light, which becomes strong background noise in the designed order imaging light, thus reducing the diffraction efficiency of the FZP. The relationship between the number of steps and the diffraction efficiency is as follows:

[0005]

[0006] Where N is the number of steps in the FZP annular structure, λ0 is the design wavelength, λ is the working wavelength, and m is the diffraction order. Theoretically, the diffraction efficiency of a two-step FZP is 40.5%, that of a four-step FZP is 81%, and that of an eight-step FZP can reach over 98%. However, due to processing errors, the actual diffraction efficiency further decreases.

[0007] On the other hand, the wavefront of the FZP imaging beam is distorted due to factors such as FZP components, system errors, and operating environment, which in turn causes the point spread function (PSF) of the FZP imaging system to diffuse, ultimately reducing the imaging resolution. The effects of wavefront distortion are as follows:

[0008]

[0009] Where (x,y) are the focal plane coordinates, (x',y') are the pupil plane coordinates, A(x',y') is the beam amplitude, Φ(x',y') is the beam wavefront distortion, z is the transmission distance, and k is the wavenumber. The model of the FZP imaging system is as follows:

[0010]

[0011] In the formula, I represents the actual image of the target, I0 represents the ideal image of the target, η1 represents the diffraction efficiency of the imaging order, and PSF1 and PSF2 represent the diffraction efficiency of the target. n are the point spread functions of the primary and secondary beams, respectively, and N is the imaging noise.

[0012] Therefore, FZP diffraction imaging is affected by multiple levels of diffraction and wavefront aberration, resulting in extremely low image resolution and a sharp degradation in image quality. To address this problem, this patent proposes a simulated annealing search strategy to iteratively optimize the PSF of FZP imaging and achieve rapid image restoration based on the imaging model. Summary of the Invention

[0013] To address the technical problems existing in the background art, this invention aims to provide a diffraction imaging system and an optimization and image restoration method for FZP imaging models, specifically a method for restoring Fresnel zone plate (FZP) imaging images. This invention is applicable to image post-processing and restoration of thin-film diffraction imaging systems such as Fresnel lenses, photon sieves, and mosaic diffraction telescopes, as well as the measurement and optimization of the point spread function (PSF) of diffraction imaging systems.

[0014] To solve the technical problem, the technical solution of the present invention is as follows:

[0015] A diffraction imaging system, the system comprising: a laser point source and monochromatic light extension sources arranged sequentially along the optical axis of the laser point source, an imaging target image, an equivalent far-field lens, a diffraction imaging element, a beam-shrinking lens, and an imaging camera;

[0016] The system further includes: a control computer; the imaging camera is signal-connected to the control computer.

[0017] Furthermore, the diffraction imaging element employs a Fresnel zone plate.

[0018] Furthermore, the beam emitted by the laser point source or monochromatic extended source passes through the imaging target image and the equivalent far-field lens to reach the Fresnel zone plate, and then passes through the beam-shrinking lens to be shrunken to the target surface of the camera before entering the imaging camera for imaging.

[0019] Furthermore, the control computer is pre-configured with an FZP imaging model optimization and image restoration calculation unit. This unit, based on experimentally measured diffraction efficiency η0 and corresponding point spread function PSF0 of the principal imaging order diffracted light, it iteratively optimizes using a simulated annealing algorithm to obtain the optimal diffraction efficiency and point spread function of FZP, as well as the equivalent point spread function PSF0 for non-diffraction orders. m Based on the FZP diffraction imaging model, a fast reconstruction of FZP images is achieved using two-dimensional Fourier forward and inverse transforms.

[0020] An FZP imaging model optimization and image restoration method, applied to any of the systems described above, the method comprising:

[0021] Step 1: A laser point source is used as the light source of the imaging system. After the beam passes through the equivalent lens and the diffraction imaging element, it is acquired by the imaging camera located at the focal plane. The image data acquired by the imaging camera is the focal plane image of the principal diffraction order of FZP diffraction imaging, denoted as PSF0, with a dimension of 16×16. The diffraction efficiency of the principal diffraction order of FZP is calculated based on this focal plane image, denoted as η0.

[0022] Step 2: Set the initial temperature T = 300, the number of iterations for each T value L = 200, and repeat steps 3 to 6 for l = 1, 2, 3, ..., L. The cooling ratio K = 0.9, and the temperature range during the iteration is [0, 200].

[0023] Step 3: Except for the principal order diffracted light, consider the equivalent PSF of other order diffracted light as a whole, and denote it as PSF. m Its initial value is set to zero, and its dimension is 16×16. PSF0 and diffraction efficiency η0 are substituted into the incoherent imaging model to establish the FZP diffraction imaging image restoration model.

[0024] Step 4: Based on the diffraction imaging image restoration model described above, calculate the increment ΔJ of the evaluation function. A set of temperature evaluation functions corresponds to the gradient of a restored image. Select the optimal temperature PSF based on the change in gradient. m ;

[0025] Step 5: Simulate annealing algorithm temperature T and PSF m The update methods are as follows:

[0026] T N+1=K×T N (6)

[0027] PSF N+1 =PSF N +sign(rand(16, 16) - 0.5) (7)

[0028] Step 6: Calculate the evaluation function increment and decide: If ΔJ < 0, then accept PSF. N+1 Update the equivalent PSF of other orders of diffracted light, that is, update this PSF. N+1 As a new PSF m Otherwise, accept the PSF with probability k. N+1 As a new PSF m ;

[0029] Step 7: If the termination condition is met, output the current PSF. N+1 As the optimal PSF m Otherwise, update T and repeat steps 2 through 7;

[0030] Step 8: Replace the blurred image I obtained in the experiment with one of different details and grayscale distributions, and repeat steps 1 to 7 to obtain a new PSF. m The optimal values ​​of η0 and η0 are then calculated, and finally the average of all optimal values ​​is obtained to obtain the equivalent PSF of the FZP diffraction imaging system:

[0031] PSF average =(PSF) m1 +PSF m2 +…+PSF mN ) / N (9)

[0032] Step 9: Based on the diffraction imaging image restoration model described in Step 3, obtain a method for rapid image restoration:

[0033] PSF=PSF0×η0+PSF average ×(1-η0) (10)

[0034]

[0035] The FZP diffraction imaging image can be rapidly restored by using a two-dimensional fast Fourier inverse transform.

[0036] Furthermore, the diffraction imaging image restoration model specifically includes:

[0037]

[0038] In the formula, I is the blurred image acquired by the FZP imaging system, I0 is the principal diffraction light imaging image, N is the imaging noise, and * represents convolution. Then, according to formula (2), the time domain problem is transformed into the frequency domain for calculation, and finally transformed back into the time domain, that is:

[0039] I fft2 =ffishift(fft2{I0*[PSF0×η0+(1-η0)×PSF m ]}) (2)

[0040]

[0041] In the formula, fft2 and ifft2 are the two-dimensional Fourier forward and inverse transforms, respectively. The function of fftshift is to move the zero-frequency component of the signal to the center of the spectrum. It is usually used in conjunction with the Fourier transform and is very useful for analyzing the frequency domain information of the signal. η0 is the initial diffraction efficiency corresponding to the principal diffraction order spot spread function PSF0. The PSF is iteratively optimized using the simulated annealing algorithm. m And the correction value for the diffraction efficiency η0.

[0042] Furthermore, step S4 specifically includes:

[0043] Based on the above diffraction imaging image restoration model, the increment ΔJ of the evaluation function is calculated according to equations (4) and (5). A set of temperature evaluation functions corresponds to the gradient of a restored image. The optimal temperature PSF is selected based on the change in gradient. m The evaluation function J is defined as:

[0044]

[0045] ΔJ=J N -J N+1 (5)

[0046] In the formula, I(x, y) is the pixel value of the restored image obtained by the first optimization, (x, y) is the image coordinate, and m and n are the resolutions of the restored image in the x and y directions.

[0047] Furthermore, step S6 specifically includes:

[0048] Based on the evaluation function increment calculated by equation (5): if ΔJ < 0, then PSF is accepted. N+1 Update the equivalent PSF of other orders of diffracted light, that is, update this PSF. N+1 As a new PSF m Otherwise, accept the PSF with probability k. N+1 As a new PSF m The probability k is calculated as follows:

[0049] k = exp(-ΔJ / T) (8).

[0050] Compared with the prior art, the advantages of the present invention are as follows:

[0051] This invention is applicable to PSF optimization and image restoration in diffraction imaging systems such as Fresnel zone plates, photon sieves, and mosaic thin-film telescopes. For different diffraction imaging systems, the diffraction efficiency η0 and the corresponding point spread function PSF0 of the main imaging order diffracted light can be measured experimentally. The optimal diffraction efficiency and the equivalent point spread function PSF of the non-imaging light can then be solved using an optimization algorithm. m Finally, the optimal value of the point spread function of the diffraction imaging system is obtained. Then, based on the diffraction imaging model, the image is rapidly restored through two-dimensional Fourier forward and inverse transforms. This image restoration method has the advantages of wide applicability, high reliability in obtaining the global optimal solution, simple algorithm, easy implementation, and high real-time performance. It provides conditions for wavefront aberration correction and image post-processing in online thin-film diffraction imaging systems, thereby improving the imaging quality of thin-film diffraction imaging systems. Attached Figure Description

[0052] Figure 1 This embodiment presents a system principle diagram of an FZP diffraction imaging model optimization and image restoration method based on simulated annealing algorithm;

[0053] Figure 2 This embodiment presents a system experimental diagram of an FZP diffraction imaging model optimization and image restoration method based on simulated annealing algorithm;

[0054] Figure 3 The point spread function (PSF0) corresponding to the principal diffracted light of the FZP imaging system;

[0055] Figure 4 The optimization convergence curves of the equivalent point spread function corresponding to the non-imaging diffracted light of the FZP imaging system, (a) to (d) correspond to the coast, forest, resolution plate and river respectively;

[0056] Figure 5 The equivalent point spread function (PSF) of the non-imaging diffracted light in the FZP imaging system m (a) to (d) represent the coastline, forest, resolution panel, and river, respectively.

[0057] Figure 6 The equivalent point spread function (PSF) of the FZP imaging system, (a) to (d) are the coast, forest, resolution plate and river, respectively;

[0058] Figure 7 The actual images (i.e., I) of the FZP imaging system, (a) to (d) are the coast, forest, resolution panel and river, respectively;

[0059] Figure 8 This embodiment presents a flowchart of an iterative optimization solution for the equivalent point spread function of a diffraction imaging model based on a simulated annealing algorithm.

[0060] Figure 9 The fast-reconstructed images (i.e., I0) based on the FZP imaging model, (a) to (d) are the coast, forest, resolution plate and river, respectively.

[0061] Figure label:

[0062] 1- Laser point source; 2- Monochromatic light extended source; 3- Imaging target image; 4- Equivalent far-field lens; 5- Diffraction imaging element; 6- Beam-contracting lens; 7- Imaging camera; 8- Control computer. Detailed Implementation

[0063] The specific implementation of the present invention is described below with reference to embodiments:

[0064] It should be noted that the structures, proportions, sizes, etc. shown in this specification are only used to complement the content disclosed in the specification for those skilled in the art to understand and read, and are not intended to limit the conditions under which the present invention can be implemented. Any modifications to the structure, changes in the proportions, or adjustments to the size, without affecting the effects and objectives that the present invention can produce, should still fall within the scope of the technical content disclosed in the present invention.

[0065] Furthermore, the terms such as "upper," "lower," "left," "right," "middle," and "one" used in this specification are merely for clarity of description and are not intended to limit the scope of the invention. Any changes or adjustments to their relative relationships, without substantially altering the technical content, should also be considered within the scope of the invention.

[0066] Example 1:

[0067] This invention primarily provides a method for optimizing Fresnel zone plate (FZP) imaging models and restoring images based on the Simulated Annealing (SA) algorithm. This method draws upon the principles of solid-state annealing, optimizing FZP imaging model parameters through a simulated annealing algorithm. It obtains the non-imaging-level secondary equivalent point spread function (PSF) and diffraction efficiency of the FZP, achieving full parameter construction of the FZP imaging model. Based on this model, rapid image restoration of FZP imaging results is realized, improving the image quality of FZP diffraction imaging. This invention uses experimentally measured initial values ​​of PSF and FZP diffraction efficiency as conditions, iteratively optimizing the optimal parameters of the FZP imaging model through a simulated annealing algorithm to achieve diffraction imaging image restoration, thereby improving the image quality. The image restoration method described in this invention is highly reliable, simple to implement, highly real-time, and robust. The search strategy of this method is less prone to getting trapped in local optima, making it more suitable for real-time image restoration applications in large-aperture space-based diffraction imaging systems.

[0068] The problem addressed by this invention is that multi-order diffraction in FZP imaging systems leads to a sharp degradation in image quality, thereby limiting the correction of wavefront aberrations and the improvement of imaging resolution. The objective of this invention is to equate the noise and wavefront aberrations caused by diffraction orders other than the principal diffraction order to a noise PSF (Pressure-Sensitive Filter). m The PSF of FZP is obtained by optimizing the simulated annealing algorithm. m Simultaneously, based on the FZP diffraction imaging model, rapid image reconstruction is achieved. The imaging system is implemented using a laser point source, monochromatic extended source, equivalent lens, FZP, beam-shrinking lens, imaging camera, and other imaging optical paths, as well as a computer control system. This provides the prerequisites for subsequent optimization of the PSF of the FZP imaging system and image enhancement.

[0069] Example 2:

[0070] Figure 1 This is a schematic diagram of an experimental system for an FZP diffraction imaging model optimization and image restoration method based on simulated annealing algorithm, provided in an embodiment of the present invention. Figure 2 This is an optical experimental platform built according to the imaging system schematic diagram. Figure 3 denoted as the point spread function (PSF0) of the principal diffracted light in the FZP imaging system.

[0071] In one implementation, the experimental imaging system uses a monochromatic extended light source to illuminate the imaging target. The imaging beam passes through an equivalent remote lens, making the imaging target appear as if it were at infinity. The imaging beam is focused by an FZP (Focused Focusing Point), and a beam-shrinking lens adjusts the beam passing through the FZP to fit the target surface of the imaging camera. The camera acquires the imaging image and uploads it to a control computer via Ethernet for image reconstruction.

[0072] The specific implementation steps are as follows:

[0073] Step 1: Using laser point source 1 as the light source of the imaging system, the beam passes through equivalent lens 4 and FZP5 and is acquired by imaging camera 7 located at the focal plane. The image data acquired by the camera is the focal plane image of the principal diffraction order of FZP diffraction imaging, denoted as PSF0, with a dimension of 16×16. Based on this focal plane image, the diffraction efficiency of the principal diffraction order of FZP is calculated and denoted as η0.

[0074] Step 2: Set the initial temperature T = 300, the number of iterations for each T value L = 200, and repeat steps 3 to 6 for l = 1, 2, 3, ..., L. The cooling ratio K = 0.9, and the temperature range during the iteration is [0, 200].

[0075] Step 3: Except for the principal order diffracted light, the equivalent PSF of other order diffracted light is considered as a whole and denoted as PSF. m Its initial value is set to zero, and its dimension is 16×16. Substituting PSF0 and diffraction efficiency η0 into the incoherent imaging model, the FZP diffraction imaging model is established, namely:

[0076]

[0077] In the formula, I represents the blurred image acquired by the FZP imaging system, I0 represents the principal diffraction light imaging image, N represents the imaging noise, and * denotes convolution. Then, according to formula (2), the time-domain problem is transformed into the frequency domain for calculation, and finally transformed back into the time domain, i.e.:

[0078] I fft2 =fftshift(fft2{I0*[PSF0×η0+(1-η0)×PSF m ]}) (2)

[0079]

[0080] In the formula, fft2 and ifft2 are the two-dimensional Fourier transforms and inverse transforms, respectively. fftshift moves the zero-frequency component of the signal to the center and is usually used in conjunction with the Fourier transform, which is very useful for analyzing the frequency domain information of the signal. η0 is the initial diffraction efficiency corresponding to the principal diffraction order spot spread function PSF0. The PSF is iteratively optimized using a simulated annealing algorithm. mAnd the correction value for diffraction efficiency η0;

[0081] Step 4: Based on the diffraction imaging image restoration model described above, calculate the increment ΔJ of the evaluation function. A set of temperature evaluation functions corresponds to the gradient of a restored image. Select the optimal temperature (PSF) based on the gradient change. m The evaluation function J is defined as follows:

[0082]

[0083] ΔJ=J N -J N+1 (5)

[0084] In the formula, I(x, y) is the pixel value of the restored image obtained by the first optimization, (x, y) is the image coordinate, and m and n are the resolutions of the restored image in the x and y directions.

[0085] Step 5: Simulate annealing algorithm temperature T and PSF m The update methods are as follows:

[0086] T N+1 =K×T N (6)

[0087] PSF N+1 =PSF N +sign(rand(16, 16) - 0.5) (7)

[0088] Step 6: Based on the evaluation function increment calculated by equation (5), determine: if ΔJ < 0, then accept PSF. N+1 Update the equivalent PSF of other orders of diffracted light, that is, update this PSF. N+1 As a new PSF m Otherwise, accept the PSF with probability k. N+1 As a new PSF m The probability k is calculated as follows:

[0089] k=exp(-ΔJ / T) (8)

[0090] Step 7: If the termination condition is met, output the current PSF. N+1 As the optimal PSF m Otherwise, update T and repeat steps 2 through 7.

[0091] Step 8: Replace the blurred image I obtained in the experiment with one of different details and grayscale distributions, and repeat steps 1 to 7 to obtain a new PSF. m The optimal values ​​of η0 and η0 are then calculated, and finally the average of all optimal values ​​is obtained to obtain the equivalent PSF of the FZP diffraction imaging system:

[0092] PSFaverage =(PSF) m1 +PSF m2 +…+PSF mN ) / N (9)

[0093] Step 9: Based on the FZP diffraction imaging model described in Step 3, obtain a method for rapid image restoration:

[0094] PSF=PSF0×η0+PSF average ×(1-η0) (10)

[0095]

[0096] The FZP diffraction imaging image can be rapidly restored by using a two-dimensional fast Fourier inverse transform.

[0097] In summary, this invention implements a diffraction imaging model optimization and image restoration method based on simulated annealing algorithm. The beam emitted from the laser point source 1 or monochromatic extended source 2 passes through the imaging target image 3 and the equivalent far-field lens 4 before reaching the Fresnel zone plate 5. It then passes through the beam-shrinking lens 6 to be shrunk to the target surface of the camera before entering the imaging camera 7 for imaging. Based on experimentally measured diffraction efficiency η0 and corresponding point spread function PSF0 of the principal imaging order diffracted light, the optimal diffraction efficiency and point spread function of FZP, as well as the equivalent point spread function PSF0 of non-diffraction orders, are obtained through iterative optimization using the simulated annealing algorithm. m Based on the FZP diffraction imaging model, a rapid reconstruction of FZP images is achieved using two-dimensional Fourier forward and inverse transforms. The image reconstruction method described in this invention has advantages such as wide applicability, high reliability in obtaining the global optimal solution, simple algorithm, ease of implementation, and high real-time performance. It provides conditions for wavefront aberration correction and image post-processing in online thin-film diffraction imaging systems, thereby improving the imaging quality of thin-film diffraction imaging systems.

[0098] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

[0099] Many other changes and modifications can be made without departing from the concept and scope of this invention. It should be understood that this invention is not limited to the specific embodiments, and the scope of this invention is defined by the appended claims.

Claims

1. A method for optimizing an FZP imaging model and restoring an image, characterized in that, The method includes: Step 1: A laser point source (1) is used as the light source for the imaging system. After the beam passes through the far-field lens (4) and the diffraction imaging element (5), it is acquired by the imaging camera (7) located at the focal plane. The image data acquired by the imaging camera (7) is the focal plane image of the principal order diffraction light of FZP diffraction imaging, denoted as Its dimensions are Based on this focal plane image, the diffraction efficiency of the principal order diffracted light of FZP is calculated and denoted as . ; Step 2: Set the initial temperature T=300, the number of iterations for each T value L=200, and repeat steps 3 to 6 for l=1,2,3,…,L. The cooling ratio K=0.9, and the temperature range during the iteration is [0, 200]. Step 3: Except for the principal order diffracted light, consider the equivalent PSF of other order diffracted light as a whole, denoted as Its initial value is set to zero, and its dimension is ,Will and diffraction efficiency By incorporating the incoherent imaging model, an FZP diffraction imaging image restoration model is established. Step 4: Based on the above diffraction imaging image restoration model, calculate the increment of the evaluation function. A set of temperature evaluation functions corresponds to the gradient of a restored image, and the optimal temperature is selected based on the change in gradient. ; Step 5: Simulate annealing algorithm temperature T and The update methods are as follows: (6) (7) Step 6: Calculate the increment of the evaluation function and determine: If Then accept Update the equivalent PSF of other orders of diffracted light, that is, this As a new Otherwise, accept with probability k. As a new ; Step 7: If the termination condition is met, output the current value. As the optimal Otherwise, update T and repeat steps 2 through 7; Step 8: Replace the blurred image I obtained in the experiment with one of different details and grayscale distributions, and repeat steps 1 to 7 to obtain a new image. and The optimal value is found, and finally, the average of all optimal values ​​is calculated to obtain the equivalent PSF of the FZP diffraction imaging system: (9) Step 9: Based on the diffraction imaging image restoration model described in Step 3, obtain a method for rapid image restoration: (10) （11） The FZP diffraction imaging image can be rapidly restored by using two-dimensional fast Fourier inverse transform. Step 4 specifically includes: Based on the above diffraction imaging image restoration model, the increment of the evaluation function is calculated according to equations (4) and (5). A set of temperature evaluation functions corresponds to the gradient of a restored image. The optimal temperature is selected based on the change in gradient. The evaluation function J is defined as: (4) (5) In the formula The pixel values ​​of the restored image obtained through one optimization are: Let m and n be the image coordinates, and m and n be the resolutions of the restored image in the x and y directions, respectively. Step 6 specifically includes: Based on the increment of the evaluation function calculated according to equation (5): if Then accept Update the equivalent PSF of other orders of diffracted light, that is, this As a new Otherwise, accept with probability k. As a new The probability k is calculated as follows: (8)。 2. The FZP imaging model optimization and image restoration method according to claim 1, characterized in that, The diffraction imaging image restoration model specifically includes: (1) In the formula, I represents the blurred image acquired by the FZP imaging system. The image is formed by the primary and secondary diffracted light, where N represents the imaging noise. Let convolution be represented, and then according to equation (2), the time-domain problem is transformed into the frequency domain for calculation, and finally transformed back into the time domain, that is: (2) (3) In the formula, fft2 and ifft2 are the two-dimensional Fourier forward and inverse transforms, respectively. The function of fftshift is to move the zero-frequency component of the signal to the center of the spectrum. It is usually used in conjunction with the Fourier transform and is very useful for analyzing the frequency domain information of the signal. Main order diffraction spot diffusion function The corresponding initial diffraction efficiency is iteratively optimized using a simulated annealing algorithm. and diffraction efficiency The correction value.

3. A diffraction imaging system, characterized in that, The system is used to perform the method described in any one of claims 1-2, and the system includes: a laser point source (1) or a monochromatic light extension source (2), and an imaging target image (3), an equivalent far-field lens (4), a diffraction imaging element (5), a beam-shrinking lens (6), and an imaging camera (7) arranged sequentially along the optical axis. The system further includes: a control computer (8); the imaging camera (7) is signal-connected to the control computer (8); The diffraction imaging element (5) is a Fresnel zone plate; The beam emitted by the laser point source (1) or monochromatic light extended source (2) passes through the imaging target image (3) and the equivalent far-field lens (4) to reach the Fresnel zone plate, and then passes through the beam-shrinking lens (6) to shrink the beam to the target surface of the camera and enter the imaging camera (7) for imaging. The control computer is pre-configured with an FZP imaging model optimization and image restoration calculation unit. This unit is based on experimentally measured diffraction efficiency of the principal order diffracted light. and the corresponding point spread function By iteratively optimizing using a simulated annealing algorithm, the optimal diffraction efficiency and point spread function of FZP, as well as the equivalent point spread functions of other orders of diffracted light, are obtained. Based on the FZP imaging model, a fast restoration of FZP images is achieved using two-dimensional Fourier forward and inverse transforms.

Images