High-precision imaging information reconstruction method and system based on nonlinear deconvolution

Abstract

The invention provides a high-precision imaging information reconstruction method and system based on nonlinear deconvolution, and the method comprises the steps: carrying out the Fourier decomposition of a target diffusion diagram I and a point spread function (PSF), and extracting an amplitude term and a phase term of the target diffusion diagram I and the PSF; and linearly recombining the phase item into a new phase, carrying out nonlinear recombination on the amplitude item to obtain a new amplitude, and carrying out inverse Fourier transform on the new phase and the new amplitude to obtain a reconstructed target. According to the method, gains and attenuation of high-frequency and low-frequency components in an image frequency spectrum are more accurately controlled, so that good balance between the target resolution and background noise is realized, high-quality target information reconstruction is finally realized, and the nonlinear reconstruction algorithm provided by the invention can be applied to the imaging field of various complex scattering environments and has a wide application prospect. For example, underwater detection, through-cloud imaging, astronomical imaging, remote sensing imaging and the like are realized. Different parameters are selected for imaging conditions in different fields to achieve the optimal reconstruction effect, and the method has a wide application prospect.

Description

Technical Field

[0001] The present invention belongs to the technical field of image processing, and relates to a high-precision imaging information reconstruction method and system based on nonlinear deconvolution. Background Art

[0002] Over the past decade, the rapid development of optical imaging technology has driven advancements in target reconstruction methods, making them a crucial tool for improving imaging quality. By enhancing image contrast and combining restoration models with optimization constraints, degraded information can be precisely constrained and compensated, restoring image information close to the real scene. During the imaging process, speckle images are typically viewed as the result of the convolution of the original image with the imaging system's point spread function (PSF) and the superposition of noise. Image reconstruction is the inverse of this degradation process, aiming to restore the degraded image using known or estimated PSF information. As a key parameter describing the spatial resolution and optical transfer characteristics of an imaging system, the PSF has been widely used in image reconstruction, particularly in the vignetting region, where the high correlation between PSFs provides a theoretical basis for restoration algorithms. However, practical applications of speckle and PSF reconstruction still face numerous challenges, such as noise-induced degradation of PSF estimation accuracy and the non-uniform and nonlinear characteristics of the PSF in complex optical systems. These factors complicate model construction and parameter solution. Therefore, developing efficient and robust speckle reconstruction techniques has become a key research focus. At present, imaging information reconstruction technology mainly includes inverse filtering, Wiener filtering, constrained restoration, iterative blind deconvolution, Bayesian prior blind restoration and TV regularized restoration methods.

[0003] Inverse filtering is an unconstrained restoration technique, the basic idea of ​​which is to restore the original image by performing a deconvolution operation on the blurred image. and degraded images , perform inverse Fourier transform to get the original image. In this case, the inverse filtering can be expressed as:

[0004]

[0005] The mathematical implementation of the above inverse filtering technique is relatively straightforward. Usually, only Fourier transform, division, and inverse Fourier transform are required for the image and blur kernel. This makes it easy to implement and does not require additional assumptions about the noise type or image content (such as noise models or signal priors). Image restoration can be performed with only the known blur kernel.

[0006] The Wiener filter technique restores the image based on the principle of minimizing the mean square error between the restored image and the original image, assuming that the image signal can be approximately regarded as a stationary random process. In digital image processing, the noise is usually considered to be white noise and the noise power spectrum is constant. The following formula can be used to approximate the Wiener filter process:

[0007] in is the signal-to-noise ratio, The value depends on the appropriate choice of experimental conditions, usually less than 1, and can be finally determined based on the visual effect of the restored image. The size of the value.

[0008] Wiener filter restoration technology offers advantages such as strong noise suppression, minimal mean square error, efficient frequency domain processing, and strong adaptability, making it suitable for image restoration in various noise types. By leveraging statistical information of both the signal and noise, it maintains good image quality during the restoration process, effectively improving image clarity in situations with high noise levels or severe blur.

[0009] Iterative blind deconvolution is a common method for blind image restoration. Its basic steps include estimating the initial image values ​​and performing alternating Fourier and inverse transforms, while also imposing constraints such as support domain restrictions and positivity restrictions on the target image and PSF. This technique requires no prior information and can estimate the degradation function and restore the image even with only a blurred image. It exhibits strong scene adaptability and can incorporate non-blind restoration techniques such as inverse filtering and Wiener filtering into the iterative process.

[0010] The Richardson-Lucy deconvolution technique combines iteration and maximum likelihood estimation. It assumes that noise follows a Poisson distribution and gradually restores image details and clarity through nonlinear iteration, while ensuring the non-negativity and total energy of the image, which helps maintain the global brightness of the image.

[0011] Total variation (TV) regularized restoration technology transforms the image restoration problem into a functional extremum problem of minimizing the energy function by constructing an energy functional based on the total variation of the image. This technology suppresses noise by minimizing the total variation of the image, maintaining the image's gradients, thereby effectively preserving edge and feature information and avoiding detail loss caused by oversmoothing. It also effectively smooths noisy areas and maintains the overall image structure.

[0012] While inverse filtering is simple to implement, in practice, the presence of noise and the small frequency components of the blur kernel can lead to noise amplification and artifacts. Furthermore, accurately estimating the blur kernel is often difficult in practice, and inaccurate estimation can significantly reduce the restoration effect.

[0013] Wiener filtering is based on a statistical minimization criterion, resulting in only optimal results on average. It also relies on estimating the power spectra of the signal and noise. In practical applications, these statistical properties are difficult to accurately capture, especially when noise and signal characteristics vary. This significantly reduces filtering effectiveness and increases computational complexity.

[0014] Iterative blind deconvolution techniques require no prior information, but they require numerous iterations, which is computationally intensive and time-consuming. Inaccurate initial estimates or improper parameter settings can slow convergence, and the technique is sensitive to noise, potentially causing artifacts or distortion in the restored image.

[0015] The Richardson-Lucy deconvolution technique requires a high initial estimate. Improper initial estimates can lead to slow or even non-convergence, heavy computational effort, and long processing time. Furthermore, the technique is sensitive to noise, which can cause edge artifacts in the restored image.

[0016] The performance of TV regularization restoration depends on the choice of regularization parameters. Improper parameter selection can lead to oversmoothing or residual noise. In the presence of strong noise, this can cause stair-stepping effects in smooth areas of the image and loss of details, especially in textured and gradient areas.

[0017] Existing techniques for recovering target information using point spread functions (PSFs) suffer from weak reconstruction adaptability. This problem makes the reconstruction sensitive to noise, prone to artifacts and noise amplification. Furthermore, they are poorly adaptable to target reconstruction in complex scattering scenes, resulting in low resolution and signal-to-noise ratio. Furthermore, in practical applications, the difficulty in accurately capturing the PSF results in low stability and reliability, making existing restoration algorithms inadequate to meet practical needs. Summary of the Invention

[0018] To address the challenges of existing technologies, this paper proposes a high-precision imaging information reconstruction method and system based on nonlinear deconvolution. By designing a reconstruction method that utilizes dual-parameter joint modulation, the method effectively improves its adaptability to complex speckle scenes. Adjusting these two parameters precisely controls the gain and attenuation of high- and low-frequency components in the image spectrum, achieving a good balance between target resolution and background noise. This method, capable of achieving high-fidelity imaging in complex scattering environments, has broad application prospects in a variety of fields, including underwater exploration, biological microscopy, astronomical imaging, computational photography, and satellite remote sensing.

[0019] The present invention is achieved through the following technical solutions: A high-precision imaging information reconstruction method based on nonlinear deconvolution, comprising: Obtaining a target diffusion map I of the target scene and a point spread function PSF of the optical system; Perform Fourier decomposition on the target diffusion map I and point spread function PSF respectively, and extract their amplitude and phase terms; The phase term is linearly reorganized into a new phase, and the amplitude term is nonlinearly reorganized into a new amplitude. The new phase and new amplitude are inverse Fourier transformed to obtain the reconstructed target.

[0020] Preferably, the target diffusion map I is formed on a camera when incoherent light is irradiated onto the target and modulated by the optical system.

[0021] Preferably, the point spread function PSF is formed on the camera as a response diagram of the optical system by replacing the target with a calibration point, which is recorded as the point spread function.

[0022] Preferably, the target diffusion map I and the point spread function PSF are subjected to Fourier decomposition respectively, and the amplitude term and phase term thereof are extracted, specifically: The speckle image I and the point spread function PSF are Fourier transformed to obtain the frequency domain expressions of the speckle image I and the point spread function PSF: and ; Extract the speckle image I and point spread function PSF amplitude terms of the frequency domain expression and , and the phase term and .

[0023] Preferably, the phase term is linearly reorganized into a new phase, specifically: Perform background noise suppression on the target diffusion map I to obtain the optimized target diffusion map ; The optimized target diffusion map The phase term The phase term of the point spread function PSF Perform linear interpolation operation, the expression is , and obtain the new phase after linear recombination.

[0024] Preferably, the amplitude term is reorganized nonlinearly to obtain a new amplitude, specifically: The optimized target diffusion map The amplitude term The amplitude term of the point spread function PSF | respectively perform nonlinear power adjustment, the expression is and , where α and β are adjustable parameters with a value range of [-1,1]. The multiplication of the two is expressed as , and obtain the new amplitude after nonlinear reorganization.

[0025] Preferably, the new phase and the new amplitude are subjected to inverse Fourier transform to obtain the reconstructed target, specifically: The reconstructed new phase and new amplitude are multiplied to obtain a composite frequency domain. The composite frequency domain is then inversely Fourier transformed to obtain the reconstructed target intensity distribution Ô(x,y). The resolution of the reconstructed image and the background noise suppression effect are balanced by adjusting the values ​​of the parameters α and β. The reconstructed target intensity distribution Ô(x,y) is expressed as:

[0026] in, and represent Fourier transform and inverse Fourier transform respectively; and Represents the target diffusion map after background noise suppression The phase value after Fourier transformation of the PSF of the scattering system.

[0027] A high-precision imaging information reconstruction system based on nonlinear deconvolution, comprising: A data acquisition module is used to obtain a target diffusion map I of a target scene and a point spread function PSF of an optical system; A decomposition module is used to perform Fourier decomposition on the target diffusion map I and the point spread function PSF, and extract their amplitude and phase terms; Linear and nonlinear recombination modules are used to linearly recombine phase terms into new phases and to nonlinearly recombine amplitude terms into new amplitudes; The inverse transform reconstruction module is used to perform inverse Fourier transform on the new phase and the new amplitude to obtain the reconstructed target.

[0028] A computer device comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method for reconstructing high-precision imaging information based on nonlinear deconvolution are implemented.

[0029] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a high-precision imaging information reconstruction method based on nonlinear deconvolution.

[0030] Compared with the prior art, the present invention has the following beneficial technical effects: This paper proposes a high-precision imaging information reconstruction method and system based on nonlinear deconvolution. By introducing parameters and establishing a system, it aims to more precisely control the gain and attenuation of high-frequency and low-frequency components in the image spectrum, thereby achieving a good balance between target resolution and background noise, ultimately achieving high-quality target information reconstruction. The proposed nonlinear reconstruction algorithm is applicable to a variety of imaging applications in complex scattering environments, such as underwater exploration, imaging through clouds and fog, astronomical imaging, and remote sensing imaging. Different parameters are selected to achieve optimal reconstruction results for different imaging applications, showing broad application prospects.

[0031] Furthermore, the present invention uses Fourier transform to transform the target diffusion map and system response, transforming the signal into the frequency domain for processing. In addition to the frequency domain, deconvolution can also be performed in other transform domains, such as the wavelet domain, the logarithmic domain, etc.

[0032] Furthermore, the present invention introduces dual-parameter modulation amplitude to perform deconvolution reconstruction of the target. Compared with traditional deconvolution methods, the method proposed in the present invention can adaptively change the resolution of the target and the background noise conditions. By adjusting the dual parameters, the resolution of the reconstructed image is improved, the contour is clearer, and the experimental noise is greatly suppressed. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0034] Figure 1 This is a flow chart of the nonlinear deconvolution reconstruction method of the present invention; Figure 2 Comparison of reconstruction results of deconvolution algorithm in the embodiment. DETAILED DESCRIPTION

[0035] The technical solution of the present invention will be described clearly and completely below. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0036] A high-precision imaging information reconstruction method based on nonlinear deconvolution, such as Figure 1 Shown, including, S1, obtain the target diffusion map I of the target scene and the point spread function PSF of the optical system; S2, perform Fourier decomposition on the target diffusion map I and the point spread function PSF, and extract their amplitude and phase terms; S3, linearly reorganizes the phase term into a new phase, and nonlinearly reorganizes the amplitude term into a new amplitude. S4, performing inverse Fourier transform on the new phase and new amplitude to obtain the reconstructed target.

[0037] Furthermore, the target diffusion map I in S1 is formed on the camera when the incoherent light is irradiated on the target and modulated by the optical system. The point spread function (PSF) is formed on the camera by replacing the target with a calibration point, and the response map of the optical system is recorded as the point spread function. Therefore, the target diffusion map detected by the camera is I Essentially, it is the convolution of the target intensity distribution and the system point spread function (PSF): (1) Among them, * represents the convolution operation.

[0038] Based on formula (1), the generalized mathematical expression of deconvolution technology is: (2) in" " represents a deconvolution operation. The present invention proposes a nonlinear reconstruction method, which performs Fourier decomposition on the target diffusion map I and the point spread function (PSF) of the optical system, extracts their amplitude and phase, directly linearly recombines the phase term into a new phase, and nonlinearly recombines the amplitude term to obtain a new phase. Finally, the new phase and new amplitude are inverse Fourier transformed to obtain the reconstructed target.

[0039] Furthermore, in S2, the target diffusion map I and the point spread function PSF are Fourier decomposed respectively, and their amplitude and phase terms are extracted, specifically: The speckle image I and the point spread function PSF are Fourier transformed to obtain the frequency domain expressions of the speckle image I and the point spread function PSF: and ; Extract the speckle image I and point spread function PSF amplitude terms of the frequency domain expression and , and the phase term and .

[0040] Furthermore, the phase term in S3 is linearly reorganized into a new phase, specifically: Background noise suppression processing is performed on the target diffusion map I to obtain an optimized target diffusion map Î; the background noise suppression processing includes at least one of frequency domain filtering, spatial domain filtering, or a noise removal algorithm based on deep learning.

[0041] The optimized target diffusion map The phase term The phase term of the point spread function PSF Perform linear interpolation operation, the expression is , and obtain the new phase after linear recombination.

[0042] The amplitude term in S3 is reorganized nonlinearly to obtain a new amplitude, specifically: The optimized target diffusion map The amplitude term The amplitude term of the point spread function PSF | respectively perform nonlinear power adjustment, the expression is and , where α and β are adjustable parameters with a value range of [-1,1]. The multiplication of the two is expressed as , obtaining a new amplitude after nonlinear reorganization. The amplitude modulation method used in this invention is specifically power function modulation, which can also be replaced by exponential modulation, logarithmic modulation, or even more complex function combination modulation. This invention only introduces dual parameters in amplitude modulation; if parameters can also be introduced in the phase domain, this approach can also be substituted for the present invention.

[0043] Furthermore, in S4, the new phase and new amplitude are subjected to inverse Fourier transform to obtain the reconstructed target, specifically: Multiplying the recombined new phase and new amplitude gives the composite frequency domain expression: ; Perform inverse Fourier transform on the composite frequency domain expression to obtain the reconstructed target intensity distribution Ô(x,y); the parameters α and β The value range of is [-1,1]. By adjusting its value, it can correspond to different filters, thereby balancing the resolution of the reconstructed image and the background noise suppression effect; the target diffusion map uses a speckle image.

[0044] The reconstructed target intensity distribution Ô(x,y) is expressed as: (3) This method introduces two parameters into the Fourier transform of speckle and PSF respectively. α and β ,parameter α and βThe value range of is [-1,1]. By adjusting the amplitude value, the balance between target resolution and background noise can be precisely controlled. and represent Fourier transform and inverse Fourier transform respectively; and Represent the speckle images after background noise suppression The phase value after Fourier transformation of the PSF of the scattering system. This nonlinear reconstruction technique can be regarded as a collection of multiple filters, for example: α = β =1, it corresponds to traditional matched filtering, which is suitable for ideal scenarios with low noise; when α=-1, β=1, it corresponds to inverse filtering, which may amplify noise and is only suitable for special scenarios with extremely weak noise; when α=1, β=0, it is phase-only filtering, which only uses phase information for reconstruction, and is suitable for scenarios where the contour needs to be preserved but the amplitude accuracy requirements are not high.

[0045] In practical applications, we typically optimize the parameters by traversing and searching within the interval α and β∈[0,1], and determine the optimal value by combining subjective visual effects with objective metrics (such as PSNR and SSIM). Based on our extensive experiments, we have found typical empirical parameters of α = 0.5 and β = 0.9, which achieve a balanced optimization of resolution and noise suppression in most complex scattering scenes. These parameters can be flexibly adjusted based on the specific scenario (such as lighting conditions, target complexity, and noise level).

[0046] A high-precision imaging information reconstruction system based on nonlinear deconvolution, comprising: A data acquisition module is used to obtain a target diffusion map I of a target scene and a point spread function PSF of an optical system; A decomposition module is used to perform Fourier decomposition on the target diffusion map I and the point spread function PSF, and extract their amplitude and phase terms; Linear and nonlinear recombination modules are used to linearly recombine phase terms into new phases and to nonlinearly recombine amplitude terms into new amplitudes; The inverse transform reconstruction module is used to perform inverse Fourier transform on the new phase and the new amplitude to obtain the reconstructed target.

[0047] The system is integrated into an optical imaging device, which includes at least one of a microscopic imaging system, a long-range scattering imaging system, or a holographic imaging system.

[0048] The method or system is applied to high-precision reconstruction scenarios of target information in the fields of biomedical imaging, industrial non-destructive testing, security monitoring or remote sensing detection.

[0049] The present invention introduces a reconstruction method based on nonlinear deconvolution of dual-parameter modulation amplitude.α and β , to fine-tune the amplitude of speckle and PSF and keep the phase value unchanged, so as to accurately control the balance between object resolution and background noise. α and β value, find the best parameters and achieve the best reconstruction effect.

[0050] The present invention constructs a speckle imaging system based on imaging target, scattering medium and detector, and combines it with nonlinear deconvolution algorithm to form a complete image reconstruction process. By collecting the scattered light field of the target and measuring the point spread function, the reconstruction is performed using formula (3) to obtain a clear restored image. This technical solution is applicable to various experimental conditions and can be dynamically adjusted. α and β The parameters are adjusted to adapt to different optical scenes and target characteristics, significantly improving the robustness and resolution of speckle reconstruction.

[0051] The nonlinear deconvolution reconstruction method proposed in this invention is compared with the restoration image of the existing deconvolution algorithm. Figure 2 Compared with the existing deconvolution method, it can be seen that the reconstructed image obtained by the nonlinear reconstruction method proposed in the present invention has the best resolution and contrast.

[0052] In another embodiment of the present invention, a computer device is provided, which includes a processor and a memory, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of the method for identifying false traffic of Internet advertisements.

[0053] In another embodiment of the present invention, a storage medium is provided, specifically a computer-readable storage medium (Memory). The computer-readable storage medium is a memory device in a computer device, used to store programs and data. It is understood that the computer-readable storage medium herein may include both built-in storage media in the computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides storage space, which stores the terminal's operating system. Furthermore, the storage space also stores one or more instructions suitable for being loaded and executed by a processor. These instructions may be one or more computer programs (including program code). It should be noted that the computer-readable storage medium herein may be a high-speed RAM memory or a non-volatile memory, such as at least one disk storage device. The processor may load and execute the one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the method for identifying fraudulent Internet advertising traffic in the above-mentioned embodiment.

[0054] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0055] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0056] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0057] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0058] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the rights of the present invention.

Claims

1. A high-precision imaging information reconstruction method based on nonlinear deconvolution, characterized in that: include, Obtaining a target diffusion map I of the target scene and a point spread function PSF of the optical system; Perform Fourier decomposition on the target diffusion map I and point spread function PSF respectively, and extract their amplitude and phase terms; The phase terms are linearly reorganized into new phases, and the amplitude terms are nonlinearly reorganized into new amplitudes; The new phase and new amplitude are inverse Fourier transformed to obtain the reconstructed target.

2. The high-precision imaging information reconstruction method based on nonlinear deconvolution according to claim 1, characterized in that: The target diffusion image I is formed on the camera when incoherent light is irradiated onto the target and modulated by the optical system.

3. The high-precision imaging information reconstruction method based on nonlinear deconvolution according to claim 1, characterized in that: The point spread function (PSF) forms a response diagram of the optical system on the camera by replacing the target with a calibration point, which is recorded as the point spread function.

4. The high-precision imaging information reconstruction method based on nonlinear deconvolution according to claim 1, characterized in that: The target diffusion map I and the point spread function PSF are decomposed by Fourier, and their amplitude and phase terms are extracted, specifically: The speckle image I and the point spread function PSF are Fourier transformed to obtain the frequency domain expressions of the speckle image I and the point spread function PSF: and ; Extract the speckle image I and point spread function PSF amplitude terms of the frequency domain expression and , and the phase term and .

5. The high-precision imaging information reconstruction method based on nonlinear deconvolution according to claim 1, characterized in that: The phase terms are linearly reorganized into new phases, specifically: Perform background noise suppression on the target diffusion map I to obtain the optimized target diffusion map Î; The optimized target diffusion map The phase term The phase term of the point spread function PSF Perform linear interpolation operation, the expression is , and obtain the new phase after linear recombination.

6. The high-precision imaging information reconstruction method based on nonlinear deconvolution according to claim 1, characterized in that: The amplitude terms are reorganized nonlinearly to obtain new amplitudes, specifically: The optimized target diffusion map The amplitude term The amplitude term of the point spread function PSF | respectively perform nonlinear power adjustment, the expression is and , where α and β are adjustable parameters with a value range of [-1,1]. The multiplication of the two is expressed as , and obtain the new amplitude after nonlinear reorganization.

7. The high-precision imaging information reconstruction method based on nonlinear deconvolution according to claim 1, characterized in that: Perform inverse Fourier transform on the new phase and new amplitude to obtain the reconstructed target, specifically: The reconstructed new phase and new amplitude are multiplied to obtain a composite frequency domain. The composite frequency domain is then inversely Fourier transformed to obtain the reconstructed target intensity distribution Ô(x,y). The resolution of the reconstructed image and the background noise suppression effect are balanced by adjusting the values ​​of the parameters α and β. The reconstructed target intensity distribution Ô(x,y) is expressed as: in, and represent Fourier transform and inverse Fourier transform respectively; and Represent the speckle images after background noise suppression The phase value after Fourier transformation of the PSF of the scattering system.

8. A high-precision imaging information reconstruction system based on nonlinear deconvolution, characterized in that: include: A data acquisition module is used to obtain a target diffusion map I of a target scene and a point spread function PSF of an optical system; A decomposition module is used to perform Fourier decomposition on the target diffusion map I and the point spread function PSF, and extract their amplitude and phase terms; Linear and nonlinear recombination modules are used to linearly recombine phase terms into new phases and to nonlinearly recombine amplitude terms into new amplitudes; The inverse transform reconstruction module is used to perform inverse Fourier transform on the new phase and the new amplitude to obtain the reconstructed target.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the processor implements the steps of a high-precision imaging information reconstruction method based on nonlinear deconvolution as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of a high-precision imaging information reconstruction method based on nonlinear deconvolution as described in any one of claims 1 to 7 are implemented.

Images