A design method for an optoelectronic imaging system with laser protection and privacy protection functions

Abstract

This application discloses a design method for an optoelectronic imaging system with laser protection and privacy protection functions, relating to the field of optoelectronic imaging. The method includes: determining an optimized point spread function based on preset point spread distribution characteristics, target energy divergence, and point spread radius using an optical-algorithm joint optimization framework; determining an initial phase distribution based on imaging parameters using the Gerchberg-Saxton algorithm; adjusting the initial phase distribution using a stochastic gradient descent algorithm, guided by the difference in energy divergence of the point spread function, to determine the optimized phase distribution corresponding to the optimized point spread function; and using the optimized phase distribution to load the phase modulation component of the optoelectronic imaging system. This application can achieve high-quality optoelectronic imaging with laser protection and privacy protection capabilities.

Description

Technical Field

[0001] This application relates to the field of optoelectronic imaging, and in particular to a design method for an optoelectronic imaging system with laser protection and privacy protection functions. Background Technology

[0002] Currently, optoelectronic imaging systems are widely used in various fields, mainly consisting of imaging lenses, sensors, and image processing modules. To achieve clear imaging, optoelectronic imaging systems typically employ a point spread function with concentrated energy. While this design provides extremely high optical gain, it also introduces hardware and software security vulnerabilities. On the hardware side, when strong light, such as laser light, shines on an optoelectronic imaging system, the energy is highly concentrated due to the converging effect of the imaging lens, easily damaging the sensor and permanently degrading the imaging system's capabilities. On the software side, in the absence of strong light interference, imaging systems face the problem of privacy information loss. Because of the clear imaging, hackers can obtain the acquired images by attacking the data transmission link. These images generally contain private information such as faces and text; directly obtaining clear images may lead to privacy leaks, posing a significant security risk to the use of the imaging system.

[0003] In existing technologies, spectral filtering techniques can be used on photoelectric imaging devices to achieve protection against specific wavelengths. However, this requires prior prediction of the interfering laser's wavelength, making it ineffective when the interfering laser is unknown. Setting specific phase functions can improve the laser protection capability of the imaging system, but this method struggles to protect the privacy information of the captured images. Lens-less imaging can hide privacy information, but this method struggles to recover the scene from the blurred image with high quality, resulting in poor image quality. In short, while existing technologies offer hardware or software protection, they only consider one aspect and cannot simultaneously address both capabilities. Summary of the Invention

[0004] The purpose of this application is to provide a design method for an optoelectronic imaging system with laser protection and privacy protection functions, which can achieve high-quality optoelectronic imaging with laser protection and privacy protection capabilities.

[0005] To achieve the above objectives, this application provides the following solution:

[0006] This application provides a design method for an optoelectronic imaging system with laser protection and privacy protection functions, the method comprising:

[0007] Acquire the target energy divergence, point spread radius, and imaging parameters of the photoelectric imaging system;

[0008] Based on the preset point diffusion distribution characteristics, the target energy divergence, and the point diffusion radius, an optimized point diffusion function is determined using a joint optimization framework of optics and algorithms.

[0009] Based on the imaging parameters, the Gerchberg-Saxton algorithm is used to determine the initial phase distribution;

[0010] Guided by the difference in energy divergence of the point spread function, the initial phase distribution is adjusted based on the stochastic gradient descent algorithm to determine the optimized phase distribution corresponding to the optimized point spread function;

[0011] The optimized phase distribution is applied to the phase modulation component of the photoelectric imaging system to complete the optimized design of the photoelectric imaging system.

[0012] According to the specific embodiments provided in this application, this application achieves the following technical effects: This application establishes a joint optimization framework of optics and algorithms. Based on this, an optimized point spread function with an energy divergence approximating the target energy divergence is obtained, which ensures good photoelectric imaging performance. Then, guided by the difference in energy divergence of the point spread function, the initial phase distribution is adjusted according to the stochastic gradient descent algorithm to obtain the optimized phase distribution corresponding to the optimized point spread function, thus realizing hybrid phase retrieval. By loading the optimized phase distribution corresponding to the optimized point spread function into the phase modulation component of the photoelectric imaging system, an additional phase is introduced into the photoelectric imaging process, thereby achieving protection against specific wavelengths and providing laser protection capabilities. Furthermore, due to the introduction of the additional phase, the image obtained by the photoelectric imaging system is a blurred image, thus providing privacy protection capabilities. In summary, this application can achieve high-quality photoelectric imaging with laser protection and privacy protection capabilities. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is an application environment diagram of the design method of an optoelectronic imaging system with laser protection and privacy protection functions in one embodiment of this application.

[0015] Figure 2 This is a flowchart illustrating the design method of an optoelectronic imaging system with laser protection and privacy protection functions provided in an embodiment of this application.

[0016] Figure 3This is a schematic diagram illustrating the specific process steps of a method provided in an embodiment of this application.

[0017] Figure 4 This is a schematic diagram of the verification optical path provided in an embodiment of this application.

[0018] Figure 5 This is a comparative diagram showing the optimized point spread function obtained through the optical-algorithm joint design of this application and the point spread function after phase distribution optimization.

[0019] Figure 6 This is a schematic diagram of the conventional imaging system obtained from experimental measurements and the point spread function and energy distribution obtained in this application.

[0020] Figure 7 This is a simulation and experimental comparison of the privacy protection capabilities of this application for text-based information.

[0021] Figure 8 This is a schematic diagram of the operation of a photoelectric imaging device based on modulation phase provided in an embodiment of this application. Detailed Implementation

[0022] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] The technical solution provided in this application has a photoelectric imaging function with laser protection and privacy protection functions. By introducing phase modulation, the point spread function of the photoelectric imaging system is much larger than that of the conventional imaging system, thereby greatly reducing the peak energy of the interfering laser on the focal plane and greatly reducing the risk of the imaging system being damaged by the laser. At the same time, the introduction of large-area blur effectively protects privacy information.

[0024] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0025] The design method for an optoelectronic imaging system with laser protection and privacy protection functions provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be set up separately, integrated into server 104, or placed in the cloud or on another server. Terminal 102 can send the target energy divergence, point spread radius, and imaging parameters of the photoelectric imaging system to server 104. After receiving the data, server 104 determines the optimal point spread function based on preset point spread distribution characteristics, target energy divergence, and point spread radius using a joint optimization framework of optics and algorithms; it determines the initial phase distribution based on the imaging parameters using the Gerchberg-Saxton algorithm; and it adjusts the initial phase distribution based on the difference in energy divergence of the point spread function using a stochastic gradient descent algorithm to determine the optimal phase distribution corresponding to the optimal point spread function. Server 104 can feed back the obtained optimized phase distribution to terminal 102. Terminal 102 can be a phase modulation component of the photoelectric imaging system. Furthermore, in some embodiments, the design method of the photoelectric imaging system with laser protection and privacy protection functions can also be implemented separately by server 104 or terminal 102.

[0026] The terminal 102 can be, but is not limited to, various desktop computers, laptops, smartphones, tablets, and IoT devices. The server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers, or it can be a cloud server.

[0027] In one exemplary embodiment, such as Figure 2 As shown, a design method for an optoelectronic imaging system with laser protection and privacy protection functions is provided. This method is executed by a computer device, specifically by a terminal or server alone, or by both a terminal and server. In this embodiment, the method is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps 201 to 205.

[0028] Step 201: Obtain the target energy divergence, point spread radius, and imaging parameters of the photoelectric imaging system. The imaging parameters include the pupil diameter, wavelength, and lens focal length.

[0029] Step 202: Based on the preset point diffusion distribution characteristics, the target energy divergence, and the point diffusion radius, an optimized point diffusion function is determined using a joint optimization framework of optics and algorithms. For example... Figure 3The diagram shows the specific flowcharts for steps 202, 203, and 204, in which a joint optimization framework of optics and algorithm is constructed, and the retrieval of the optimized phase distribution corresponding to the obtained optimized point spread function is completed. Based on this, step 202 includes the following steps (21)-(27) to complete the design of the optimized point spread function.

[0030] In a practical application, before performing step (21), it is necessary to initialize the target energy divergence and point diffusion radius, and also to set the first preset number of times to provide basic data for subsequent steps.

[0031] (21) Determine the point diffusion constraint conditions based on the preset point diffusion distribution characteristics and the point diffusion radius. Specifically, in order to maximize information transmission, the large-volume point diffusion function should possess three characteristics, namely, the preset point diffusion distribution characteristics include: non-directionality, high contrast, and small spatial size. To satisfy the above three characteristics, the point diffusion constraint conditions are set as follows:

[0032] 1) The region corresponding to the point spread function is: within a circular area determined by the point spread radius; this constraint limits the size of the point spread function to ensure that the optimized point spread function satisfies the characteristics of being non-directional and having a small spatial size. 2) The point spread function p... The expression is: where b is a constant and h is an optimizable variable. The sigmoid function is used to make the energy distribution of the point spread function satisfy the high contrast characteristic.

[0033] (22) Based on the point diffusion constraint, an initial point diffusion function is randomly generated.

[0034] (23) Convolve the initial point spread function with the reference image to obtain a blurred image, thereby simulating the camera imaging process; specifically, the following formula is used: I E =S(I*p+η); where I E The image is measured by the sensor, i.e., the image detected by the imaging component; I represents the sharp scene, referring to the reference image; p is the learnable point spread function, where each pixel is a variable that can be independently optimized; η is Gaussian noise, and η~N(0,σ) 2 S(·) is a clipping operation that sets saturated pixels to 1.

[0035] (24) The blurred image is input into a preset image reconstruction model to obtain the decoded image, thereby restoring the blurred image. The preset image reconstruction model is trained based on one of the following: MIMO-Unet network, Uformer network, or DeblurGAN network. The training process of the above networks is the conventional neural network training process, which will not be described in detail here. In addition, traditional image deconvolution methods represented by Wiener filtering and the Lucy-Richardson algorithm can also be used.

[0036] (25) Based on the decoded image, the reference image, the energy divergence of the initial point spread function, and the target energy divergence, calculate the first loss function value and record the first iteration number plus one. Specifically, in order to enable the security camera to have both good security protection capabilities and imaging capabilities, the first loss function includes both the imaging component and the energy divergence. The formula for calculating the first loss function value is as follows:

[0037]

[0038] in, α1, α2, and α3 are the first loss function values, and α1, α2, and α3 are the weighting coefficients. Used to characterize the multi-scale L1 distance between the decoded image and the reference image, where K is the number of scale levels in the image, and t k The number of pixels at level k is given. Note that the default image reconstruction model has multi-scale input and output. Taking the input as an example, the input consists of three scales: the original blurred image, the blurred image downsampled by 2x, and the blurred image downsampled by 4x. Each scale is a level. The image is the decoded image, I is the reference image, and || ||1 is the L1 norm symbol; Used to characterize the multi-scale L1 distance between the decoded image and the reference image in the frequency domain. Fourier transform; The L1 distance is used to characterize the initial energy divergence ESR and the target energy divergence ESR0 of the point spread function. This loss gives the point spread function a specific energy divergence.

[0039] Energy divergence (ESR) is defined as the ratio of the peak energy at the focal plane of an imaging system without phase modulation to the peak energy of a security camera. A larger ESR implies better laser protection and better information hiding capabilities. Therefore, the formula for calculating ESR is: ESR = max(p0) / max(p... secure ); max() is the function to find the maximum value, p0 is the point spread function of the photoelectric imaging system without phase modulation, p secure Let be the initial point spread function.

[0040] (26) When the first iteration number does not reach the first preset number, the gradient is calculated using the first loss function to update the initial point spread function, and then the process returns to the step of convolving the initial point spread function with the reference image.

[0041] (27) When the first iteration number reaches the first preset number, the initial point spread function is marked as the optimized point spread function.

[0042] Step 203: Based on the imaging parameters, the Gerchberg-Saxton algorithm is used to determine the initial phase distribution. In a practical application, such as... Figure 3 As shown, before step 203, the pupil diameter, wavelength, and lens focal length of the photoelectric imaging system are used as initialization parameters.

[0043] Step 204: Guided by the difference in energy divergence of the point spread function, the initial phase distribution is adjusted based on the stochastic gradient descent algorithm to determine the optimized phase distribution corresponding to the optimized point spread function.

[0044] In one exemplary application, step 204 includes the following steps (41)-(44).

[0045] (41) Using the angular spectrum method, the corresponding simulated point spread function is calculated based on the initial phase distribution; the simulated point spread function p est The calculation process includes:

[0046] Because the distance between the scene and the imaging component is far enough, the light field before reaching the pupil (the light field before reaching the phase modulation component)... It can be considered as a plane wave, and the optical field after passing through the phase modulation component is:

[0047]

[0048] in, The light field after passing through the phase modulation component of the photoelectric imaging system; A(x,y) is a binary circular mask, specifically a binary circular mask with a diameter of D, simulating the finite aperture size of the pupil; φ p and The additional phase introduced by the phase modulation component is, since the phase modulation component includes a phase plate and a lens arranged in sequence, the two additional phases mentioned above are the additional phases introduced by the phase plate and the lens with a focal length of f, respectively; j is the imaginary unit; (x,y) are the coordinate axes at the pupil.

[0049]

[0050] in, The light field that has been diffracted and reaches the imaging component; This is the inverse Fourier transform. Fourier transform; For optical transmission nuclei, λ is the wavelength, and f is the wavelength. x f y (x′, y′) are the frequency domain coordinates; (x′, y′) are the coordinate axes at the imaging component.

[0051]

[0052] (42) Based on the simulated point spread function, the optimized point spread function, the energy divergence of the simulated point spread function, and the energy divergence of the optimized point spread function, calculate the second loss function value and record the second iteration number plus one. The formula for calculating the second loss function value is as follows:

[0053]

[0054] in, The second loss function value is represented by β1 and β2, which are the weighting coefficients; p est For the simulated point spread function, p target To optimize the point spread function, ESR target To optimize the energy divergence of the point spread function, ESR(p est ) represents the energy divergence of the simulated point spread function.

[0055] (43) When the second iteration number does not reach the second preset number, the gradient is calculated using the second loss function to update the initial phase distribution, and then the process returns to the step of using the angular spectrum method to calculate the corresponding simulated point spread function based on the initial phase distribution.

[0056] (44) When the second iteration number reaches the second preset number, the initial phase distribution is marked as the optimized phase distribution.

[0057] Step 205: The optimized phase distribution is applied to the phase modulation component of the photoelectric imaging system to complete the optimized design of the photoelectric imaging system.

[0058] In a practical application, after obtaining the optimized phase distribution from the above steps, it can be based on... Figure 4 A verification optical path was constructed with the following parameters: lens focal length of 50mm and working wavelength of 633nm; imaging component size of 400×400 and pixel size of 4.5μm; standard deviation of Gaussian noise uniformly distributed between 0.001 and 0.01; point spread radius r = 50 and target energy divergence ESR0 = 600.

[0059] like Figure 5The diagram shown is a comparison of the optimized point spread function and optimized phase distribution obtained through the design method of the photoelectric imaging system with laser protection and privacy protection functions in this application, with the results before optimization. Figure 5 (a) shows the designed point spread function and the point spread function obtained after phase retrieval (optimized point spread function) under the corresponding parameter settings, and the two are highly consistent; Figure 5 (b) shows the initial phase and the final optimized phase distribution.

[0060] The above Figure 5 The optimized phase distribution is loaded in Figure 4 The verification optical path and functional verification of the security camera (i.e., the imaging component) are performed. The specific working principle is as follows: A 633nm wavelength laser is emitted from a He-Ne laser and passes through a spatial filter SF, then through a beam-splitting cube BS1, and finally through an achromatic cemented doublet lens L1, becoming parallel light to simulate a light source from infinity. Next, the light passes through a polarizer P, becoming linearly polarized, and then through a beam-splitting cube BS2 to illuminate a phase-type liquid crystal spatial light modulator SLM (which has an optimized phase distribution) before reflection. After passing through a 4-f system composed of two achromatic cemented lenses L2 and L3, the phase-modulated wavefront is relayed to the pupil of the compound lens, and finally, the imaging component (sensor) receives the point spread function of the photoelectric imaging system. Similarly, after being illuminated by an LED, the reflected light from the object passes through a narrow-band filter LF and propagates in the same way, obtaining an encoded blurred scene on the imaging component. Finally, the acquired blurred image is processed by a deblurring network to obtain a clear image.

[0061] Figure 6 The diagram shows the point spread function and energy distribution obtained by the conventional imaging system and the present application, as measured by experiments. The experimental results show that the present application can reduce the peak energy by more than 99.73%, which greatly reduces the peak energy and effectively ensures the safety of the imaging system.

[0062] Figure 7 The simulation and experimental comparison diagrams show the ability of this application to protect privacy information, taking text as an example. The diagrams demonstrate that this application can effectively protect privacy information and, with the correct decoder parameters, can recover images with high quality.

[0063] In summary, this application proposes a distribution characteristic of a large-volume point spread function that maximizes information transmission. Guided by this, a joint optimization framework of optics and algorithms is constructed to design an ideal point spread function. Then, a hybrid phase retrieval algorithm based on stochastic gradient descent is proposed to obtain the camera's phase distribution. Finally, this phase distribution is loaded onto a spatial light modulator to realize the proposed optoelectronic imaging system. The above setup ensures high-quality imaging while providing laser protection and privacy protection capabilities. Specifically, this application not only maintains the high imaging quality of a large-area optoelectronic imaging system but also possesses laser protection capabilities far superior to conventional optoelectronic imaging systems. Simultaneously, it achieves privacy protection at the optical end, effectively overcoming the shortcomings of existing optoelectronic imaging systems in laser protection and privacy security, and demonstrating broad application adaptability.

[0064] Based on the same inventive concept, this application also provides a modulation phase-based photoelectric imaging device for implementing the above-described method. The solution provided by this device is similar to the implementation described in the above-described method; therefore, the specific limitations of one or more modulation phase-based photoelectric imaging device embodiments provided below can be found in the limitations of the method described above, and will not be repeated here.

[0065] In an exemplary embodiment, to design a high-performance phase distribution, the device of this application includes an optoelectronic imaging system and a processor. The optoelectronic imaging system is a wavefront-coded optoelectronic imaging system, including a phase modulation component, an imaging component, an image processing component, and an output display component. The processor is used to: execute a computer program to implement the design method of the optoelectronic imaging system with laser protection and privacy protection functions, to obtain an optimized phase distribution, and then load the optimized phase distribution onto the phase modulation component. The phase modulation component includes a phase plate (or phase mask) and a lens arranged sequentially. The lens can be a compound lens, providing information on the lens focal length and pupil aperture for the design method of the optoelectronic imaging system with laser protection and privacy protection functions.

[0066] At work, such as Figure 8 As shown, the incident light is modulated by the phase modulation component, reaches the imaging component, and captures a blurred image. The image processing component then performs a deblurring algorithm on the received blurred image to obtain a clear image, which is displayed on the output display component. By introducing phase modulation through the phase modulation component, the point spread function of the imaging system is changed, thereby enabling the imaging system to possess both laser protection and privacy protection functions.

[0067] In practical applications, the imaging component is a camera, and the imaging parameters of the camera can be changed, such as the size of the primary lens, focal length, wavelength, and energy divergence of the point spread function.

[0068] In practical applications, the phase modulation component can be a liquid crystal spatial light modulator, a diffractive optical element, or a metasurface to load the phase and achieve the control of the light field.

[0069] Compared with the prior art, this application also has the following advantages:

[0070] (1) This application utilizes the characteristics of a large-volume PSF with good imaging performance as a constraint, and uses a joint optimization framework of optics and algorithms to optimize and obtain the optimal diffusion function under a specific energy divergence. That is, this application is a technical solution to achieve laser protection and privacy protection functions by designing the point spread function of the photoelectric imaging system.

[0071] (2) The phase obtained by the Gerchberg-Saxton algorithm is used as the initial phase distribution. Guided by the loss function, the phase distribution is finely adjusted based on the gradient descent method to obtain the optimized phase distribution corresponding to the optimal point spread function.

[0072] (3) This application introduces a method for simultaneously performing laser protection and privacy protection using a phase modulation module. The phase modulation module in the lens is not limited to a spatial light modulator, but can also be implemented using diffractive optical elements and metasurface components.

[0073] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0074] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for designing an optoelectronic imaging system with laser protection and privacy protection functions, characterized in that, The method includes: Acquire the target energy divergence, point spread radius, and imaging parameters of the photoelectric imaging system; Based on the preset point diffusion distribution characteristics, the target energy divergence, and the point diffusion radius, an optimized point diffusion function is determined using a joint optimization framework of optics and algorithms. Based on preset point diffusion distribution characteristics, the target energy divergence, and the point diffusion radius, an optical-algorithm joint optimization framework is used to determine an optimized point diffusion function, including: determining point diffusion constraints based on preset point diffusion distribution characteristics and the point diffusion radius; randomly generating an initial point diffusion function based on the point diffusion constraints; convolving the initial point diffusion function with a reference image to obtain a blurred image; inputting the blurred image into a preset image reconstruction model to obtain a decoded image; calculating a first loss function value and recording the first iteration count incremented by one based on the decoded image, the reference image, the energy divergence of the initial point diffusion function, and the target energy divergence; when the first iteration count has not reached a first preset number, updating the initial point diffusion function by calculating the gradient using the first loss function, and then returning to the step of convolving the initial point diffusion function with the reference image; when the first iteration count reaches the first preset number, marking the initial point diffusion function as an optimized point diffusion function. The point diffusion constraint conditions include: the region corresponding to the point diffusion function is within a circular region determined based on the point diffusion radius; the point diffusion function p... This means that b is a constant and h is an optimizable variable. Based on the imaging parameters, the Gerchberg-Saxton algorithm is used to determine the initial phase distribution; Guided by the difference in energy divergence of the point spread function, the initial phase distribution is adjusted based on the stochastic gradient descent algorithm to determine the optimized phase distribution corresponding to the optimized point spread function; Guided by the difference in energy divergence of the point spread function, the initial phase distribution is adjusted using the stochastic gradient descent algorithm to determine the optimized phase distribution corresponding to the optimized point spread function. This includes: using the angular spectrum method to calculate the corresponding simulated point spread function based on the initial phase distribution; calculating a second loss function value based on the simulated point spread function, the optimized point spread function, the energy divergence of the simulated point spread function, and the energy divergence of the optimized point spread function, and recording the second iteration count incremented by one; when the second iteration count has not reached a second preset number, updating the initial phase distribution using the gradient calculated by the second loss function, and then returning to the step of using the angular spectrum method to calculate the corresponding simulated point spread function based on the initial phase distribution; when the second iteration count reaches the second preset number, marking the initial phase distribution as the optimized phase distribution. The simulated point spread function p est The calculation process, comprising: ； in, The light field after passing through the phase modulation component of the photoelectric imaging system. To reach the front optical field of the phase modulation component, It is a binary circular mask; and Additional phase introduced for the phase modulation component; j is the imaginary unit; ； in, The light field that has been diffracted and reaches the imaging component; This is the inverse Fourier transform. Fourier transform; For optical transmission core, For wavelength, , Frequency domain coordinates; ； The optimized phase distribution is applied to the phase modulation component of the photoelectric imaging system to complete the optimized design of the photoelectric imaging system.

2. The design method of the photoelectric imaging system with laser protection and privacy protection functions according to claim 1, characterized in that, The preset image reconstruction model is obtained by training one of the following: MIMO-Unet network, Uformer network, or DeblurGAN network.

3. The design method of the photoelectric imaging system with laser protection and privacy protection functions according to claim 1, characterized in that, The formula for calculating the first loss function value is: ； ； ； ； in, The first loss function value, These are the weighting coefficients; Used to characterize the multi-scale L1 distance between the decoded image and the reference image, where K is the number of scale levels in the image. Let be the number of pixels at level k. I represents the decoded image, and I represents the reference image. The L1 norm symbol; Used to characterize the multi-scale L1 distance between the decoded image and the reference image in the frequency domain. Fourier transform; The energy divergence (ESR) of the initial point spread function and the L1 distance of the target energy divergence (ESR0) are used to characterize the energy divergence. The formula for calculating the energy divergence (ESR) is as follows: max() is the function to find the maximum value. Let be the point spread function of the photoelectric imaging system without phase modulation. Let be the initial point spread function.

4. The design method of the photoelectric imaging system with laser protection and privacy protection functions according to claim 1, characterized in that, The formula for calculating the second loss function value is: ； in, This is the value of the second loss function. These are the weighting coefficients; p est To simulate the point spread function, To optimize the point spread function, To optimize the energy divergence of the point spread function, To simulate the energy divergence of the point spread function, This is the L1 norm symbol.

Images