Two-dimensional projection potential reconstruction method and device based on local orbit

Abstract

The invention relates to a two-dimensional projection potential reconstruction method and device based on a local orbit. The method comprises the following steps: acquiring high-resolution image data and diffraction pattern data of a sample under different imaging conditions, and inputting the high-resolution image data and the diffraction pattern data into a pre-constructed two-dimensional projection potential reconstruction model for multi-slice simulation imaging calculation and image calculation processing to obtain two-dimensional projection reconstruction information of the sample. The diffraction pattern is more sensitive to the tilting angle, so that the tilting angle deviation of the sample can be corrected through an algorithm from the diffraction pattern, and high-frequency information which does not exist in the microscopic image can be obtained through the diffraction pattern. By using the local orbit, the solution space of the projection potential is constrained by using the prior information, namely the locality of the atomic electrostatic potential, so that the defects that the existing projection potential reconstruction technology depends on the periodicity of a known sample and is poor in convergence are overcome, and meanwhile, the signal-to-noise ratio is improved.

Description

Technical Field

[0001] The present application relates to the field of computational imaging technology, and particularly to a method and apparatus for reconstructing two-dimensional projection potential based on local orbits. Background Art

[0002] In the field of electron microscopy, improving spatial resolution has always been a key development goal. Transmission electron microscopy can improve resolution through hardware upgrades, such as reducing the electron beam wavelength and using aberration correctors; there are also computational imaging methods, such as exit wave reconstruction and electron holography imaging. Among them, computational imaging has attracted much attention because it can obtain higher resolution.

[0003] For the exit wave reconstruction technology of transmission electron microscopes, usually under the irradiation of a parallel electron beam, the objective lens defocus is changed to collect a series of underfocus images, and then an iterative algorithm is used to obtain the electron exit wave function on the lower surface of the sample. There are various existing exit wave reconstruction technologies. For example, the parabolic method is fast but only applicable to thin samples and has limited resolution; the maximum likelihood method considers non-linear imaging factors, has higher resolution and forms commercial software; the iterative wave function method is widely used in thin samples.

[0004] However, the current exit wave reconstruction technology has too high requirements for sample thickness and tilt angle, and the resolution is limited by the information limit of the objective lens. Summary of the Invention

[0005] Based on this, it is necessary to provide a method and apparatus for reconstructing two-dimensional projection potential based on local orbits that can solve the above problems for the above technical problems.

[0006] In the first aspect, the present application provides a method for reconstructing two-dimensional projection potential based on local orbits. The method includes:

[0007] Obtaining high-resolution image data and diffraction pattern data of a sample under different imaging conditions;

[0008] Inputting the high-resolution image data and diffraction pattern data into a pre-constructed two-dimensional projection potential reconstruction model for multi-slice simulation imaging calculation and image calculation processing to obtain two-dimensional projection reconstruction information of the sample; the two-dimensional projection potential reconstruction model is constructed based on a projection potential function and a multi-slice simulation imaging model; the projection potential function is constructed based on local functions of several types of atomic orbitals; the two-dimensional projection reconstruction information is used to reflect the two-dimensional projection potential distribution of the sample.

[0009] In the second aspect, the present application also provides an apparatus for reconstructing two-dimensional projection potential based on local orbits. The apparatus includes:

[0010] A data acquisition module, configured to obtain high-resolution image data and diffraction pattern data of a sample under different imaging conditions;

[0011] A reconstruction module is configured to input high-resolution image data and diffraction pattern data into a pre-constructed two-dimensional projection potential reconstruction model for multi-slice simulation imaging calculation and image calculation processing, so as to obtain two-dimensional projection reconstruction information of a sample; the two-dimensional projection potential reconstruction model is constructed based on a projection potential function and a multi-slice simulation imaging model; the projection potential function is constructed based on local functions of several types of atomic orbitals; the two-dimensional projection reconstruction information is used to reflect the two-dimensional projection potential distribution of the sample.

[0012] In a third aspect, the present application also provides a computer device. The computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of the method in the first aspect are implemented.

[0013] In a fourth aspect, the present application also provides a computer-readable storage medium. On the computer-readable storage medium, a computer program is stored, and when the computer program is executed by a processor, the steps of the method in the first aspect are implemented.

[0014] In a fifth aspect, the present application also provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the steps of the method in the first aspect are implemented.

[0015] The above two-dimensional projection potential reconstruction method and device based on local orbits obtain high-resolution image data and diffraction pattern data of a sample under different imaging conditions, and input the high-resolution image data and diffraction pattern data into a pre-constructed two-dimensional projection potential reconstruction model for multi-slice simulation imaging calculation and image calculation processing, so as to obtain two-dimensional projection reconstruction information of the sample. Since the diffraction pattern is more sensitive to the tilt angle, the tilt angle deviation of the sample can be corrected through an algorithm starting from the diffraction pattern, and high-frequency information not available in the microscopic image can be obtained from the diffraction pattern. Using local orbits utilizes the prior information of the locality of the atomic electrostatic potential to constrain the solution space of the projection potential, making up for the deficiencies of the existing projection potential reconstruction technology that relies on the periodicity of the known sample and has poor convergence, and at the same time improving the signal-to-noise ratio. Description of the Drawings

[0016] Figure 1 It is an application environment diagram of the two-dimensional projection potential reconstruction method based on local orbits in an embodiment;

[0017] Figure 2 It is a flowchart of the two-dimensional projection potential reconstruction method based on local orbits in an embodiment;

[0018] Figure 3 It is a flowchart of multi-slice simulation imaging and image calculation processing in an embodiment;

[0019] Figure 4Schematic diagram of the process for calculating the outgoing wave in an embodiment;

[0020] Figure 5 Schematic diagram of the process for determining the incident wave function value in an embodiment;

[0021] Figure 6 Schematic diagram of the process for performing image calculation processing in an embodiment;

[0022] Figure 7 Schematic diagram of the process for performing error calculation processing in an embodiment;

[0023] Figure 8 Schematic diagram of the process for optimizing the parameters to be updated in an embodiment;

[0024] Figure 9a High-resolution image with a defocus resolution of 7.2 nm in an embodiment;

[0025] Figure 9b High-resolution image with a defocus resolution of 47.2 nm in an embodiment;

[0026] Figure 9c High-resolution image with a defocus resolution of 87.2 nm in an embodiment;

[0027] Figure 9d Diffraction image of the zone axis under convergent beam illumination conditions in an embodiment;

[0028] Figure 10 Image of initializing the potential function with a Gaussian function array in an embodiment;

[0029] Figure 11a Distribution diagram of local orbital functions in the projection potential reconstruction result in an embodiment;

[0030] Figure 11b High-resolution image in the projection potential reconstruction result in an embodiment;

[0031] Figure 11c Diffraction pattern in the projection potential reconstruction result in an embodiment;

[0032] Figure 12 Structure block diagram of a two-dimensional projection potential reconstruction device based on local orbitals in an embodiment;

[0033] Figure 13 Internal structure diagram of a computer device in an embodiment. Detailed implementation manners

[0034] In order to make the objectives, technical solutions, and advantages of this application clearer, the following further elaborates on this application in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely used to explain this application and are not used to limit this application.

[0035] First, before specifically introducing the technical solutions of the embodiments of this application, the technical background on which the embodiments of this application are based will be introduced.

[0036] Improving spatial resolution has always been the core goal of the development of electron microscopes. The ways to improve resolution in transmission electron microscopy include reducing the electron beam wavelength, aberration corrector, exit wave reconstruction, electron holography imaging, etc. Among them, the first two intuitively improve the image quality through hardware upgrades, and the latter two belong to computational imaging methods, which require algorithm processing after taking images to restore the structural information of the sample. Computational imaging methods in electron microscopy have been widely studied because they can obtain higher resolution.

[0037] A typical scheme of the exit wave reconstruction technology in a transmission electron microscope is as follows: under the irradiation of a parallel electron beam, by changing the defocus amount of the objective lens, several images (referred to as a series of underfocus images) are collected for the same region of the sample, and then the electron exit wave function on the lower surface of the sample is obtained from these images through an iterative algorithm. Specifically, the iterative algorithms include the parabolic approximation method (PAM), the maximum likelihood method (MAL), and the iterative wave function reconstruction method (IWFR).

[0038] In the field of electron microscopy exit wave reconstruction, Related Art 1 discloses a parabolic approximation method for exit wave reconstruction technology. This method stacks a series of underfocus images into a three-dimensional matrix and performs a three-dimensional Fourier transform, then constructs a filter according to the contrast transfer function of the objective lens and filters the transformed image, and finally performs an inverse Fourier transform on the filtered image to obtain the amplitude and phase of the exit wave. Its characteristic is fast calculation speed, but it only considers the process of linear imaging, is only effective for thin samples, and the resolution is also limited.

[0039] Related Art 2 discloses a maximum likelihood method for exit wave reconstruction technology. This method designs a set of exit wave reconstruction algorithms based on the mathematical framework of maximum likelihood estimation. First, a physical model from the exit wave to the image plane is built, then a loss function is defined to calculate the difference between the simulated image and the experimental image, and finally the minimized loss function is obtained through iteration to get the best estimate of the exit wave. This method is close to the mainstream framework of modern computational imaging methods, considers the factors of non-linear imaging, has a relatively high reconstruction resolution, and also forms the mainstream commercial software TrueImage.

[0040] Related Art 3 discloses an existing iterative wave function method for exit wave reconstruction technology. This method obtains the gradient of the updated wave function by imposing an amplitude constraint in the image space, has strong robustness, and is widely used in thin samples.

[0041] Related art 4 discloses an existing local orbital stacking imaging technique. This method is based on four-dimensional scanning transmission electron microscopy data and represents the potential function information of the sample with local orbital functions, thus improving the information limit of microscope imaging to 14 pm.

[0042] Related art 5 discloses a local orbital tomography technique. This method represents the three-dimensional potential function information of the sample with local orbital functions, improves atomic-resolution tomography to the million-atom level, and simultaneously improves the position accuracy of atomic coordinates.

[0043] Based on this, the present application provides a two-dimensional projection potential reconstruction method and apparatus based on local orbitals, aiming to solve the above technical problems.

[0044] The two-dimensional projection potential reconstruction method based on local orbitals provided by the embodiments of the present application can be applied to, for example, Figure 1 the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or placed in the cloud or other network servers. The terminal 102 inputs the high-resolution image data and diffraction pattern data of the sample under different imaging conditions into a pre-constructed two-dimensional projection potential reconstruction model for multi-slice simulation imaging calculation and image calculation processing to obtain the two-dimensional projection reconstruction information of the sample. Among them, the terminal 102 can be, but is not limited to, various personal computers and laptop computers. The server 104 can be implemented with an independent server or a server cluster composed of multiple servers.

[0045] In an exemplary embodiment, as Figure 2 shown, the present application provides a two-dimensional projection potential reconstruction method based on local orbitals. Taking the terminal in Figure 1 as an example, the method includes the following steps:

[0046] S201, obtain the high-resolution image data and diffraction pattern data of the sample under different imaging conditions.

[0047] Among them, the high-resolution image data is the image information of the sample obtained by the transmission electron microscope under various imaging conditions, and these imaging conditions include, but are not limited to, using convergent beam illumination, changing the objective lens defocus amount, changing the incident angle of the electron beam, etc. The diffraction pattern data is the pattern data formed by the diffraction phenomenon generated after the interaction between the electron beam and the sample. When the electron beam irradiates the sample, due to the scattering of electrons by the sample atoms, a specific diffraction pattern will be formed.

[0048] In the embodiments of the present application, the terminal controls a transmission electron microscope to perform imaging operations on a sample. During the imaging process, the terminal sets different imaging conditions, including but not limited to using convergent beam illumination, changing the objective lens defocus amount, changing the incident angle of the electron beam, etc. The terminal records the relevant imaging parameters for each imaging, such as the sample spherical aberration coefficient, defocus amount, incident electron wave function, etc. Through these operations, the terminal obtains high-resolution image data and diffraction pattern data of the sample under different imaging conditions.

[0049] Another implementation: The terminal is connected to the transmission electron microscope through a network. The terminal sends imaging instructions and parameter setting information to the microscope. The microscope performs imaging according to the instructions and transmits the collected high-resolution image data and diffraction pattern data back to the terminal in real time. While receiving the data, the terminal records the relevant imaging parameters. For example, the terminal communicates with the microscope through a wireless local area network, remotely controls the imaging process, and obtains data of multiple samples from different regions under different imaging conditions.

[0050] S202, input the high-resolution image data and diffraction pattern data into a pre-constructed two-dimensional projection potential reconstruction model for multi-slice simulation imaging calculation and image calculation processing to obtain two-dimensional projection reconstruction information of the sample.

[0051] Among them, the two-dimensional projection potential reconstruction model is constructed based on a projection potential function and a multi-slice simulation imaging model; the projection potential function is constructed based on local functions of several types of atomic orbitals; the two-dimensional projection reconstruction information is used to reflect the two-dimensional projection potential distribution of the sample. The two-dimensional projection reconstruction information contains the two-dimensional projection potential distribution of the sample and can be used to analyze the microscopic structural characteristics of the sample, such as atomic arrangement, electron cloud distribution, etc.

[0052] In the embodiments of the present application, the terminal inputs the obtained high-resolution image data and diffraction pattern data into a two-dimensional projection potential reconstruction model pre-constructed locally. The terminal first preprocesses the data according to the model requirements, such as data format conversion, normalization, etc. Then, the terminal calls the projection potential function in the model and initializes the projection potential using local functions similar to atomic orbitals (such as Gaussian functions). Next, the terminal uses the multi-slice simulation imaging model to calculate the outgoing wave according to the imaging parameters, and then obtains the simulated TEM diffraction pattern and image. After that, the terminal compares the simulated image with the experimental image, calculates the loss function, and uses the numerical derivative method to calculate the gradient of each parameter from the loss function, and updates the parameters in the projection potential function according to the gradient, such as the intensity, radius, center coordinates of the local orbital function, etc. The terminal continuously repeats the above processes of simulated imaging, calculating the loss function, and updating the parameters until the convergence condition is met (such as the number of iterations reaches the set value or the loss function reaches the convergence set value), and finally obtains the two-dimensional projection reconstruction information of the sample.

[0053] Another implementation: The terminal uploads data to the cloud server and uses the computing resources of the cloud to run the two-dimensional projection potential reconstruction model. The cloud server returns the calculation result, that is, the two-dimensional projection reconstruction information of the sample. The terminal performs post-processing and analysis on the returned information locally.

[0054] In the above two-dimensional projection potential reconstruction method based on local orbits, by obtaining the high-resolution image data and diffraction pattern data of the sample under different imaging conditions, the high-resolution image data and diffraction pattern data are input into a pre-constructed two-dimensional projection potential reconstruction model for multi-slice simulated imaging calculation and image calculation processing to obtain the two-dimensional projection reconstruction information of the sample. Since the diffraction pattern is more sensitive to the tilting angle, the tilting angle deviation of the sample can be corrected by an algorithm starting from the diffraction pattern, and high-frequency information not available in the microscopic image can be obtained from the diffraction pattern. Using local orbits makes use of the prior information of the locality of the atomic electrostatic potential to constrain the solution space of the projection potential, making up for the deficiencies of the existing projection potential reconstruction techniques that rely on the periodicity of the known sample and have poor convergence, and at the same time improving the signal-to-noise ratio.

[0055] In an exemplary embodiment, based on the above embodiment, please refer to Figure 3 This application embodiment involves the process of inputting high-resolution image data and diffraction pattern data into a pre-constructed two-dimensional projection potential reconstruction model for multi-slice simulated imaging calculation and image calculation processing to obtain the two-dimensional projection reconstruction information of the sample, including the following steps:

[0056] S301, input the high-resolution image data and diffraction pattern data into the projection potential function for projection potential calculation processing to obtain the total projection potential.

[0057] Among them, the total projection potential is the result obtained by calculating the high-resolution image data and diffraction pattern data input into the projection potential function, which comprehensively reflects the projection potential distribution of the sample on the two-dimensional plane.

[0058] In the embodiment of this application, the terminal inputs the obtained high-resolution image data and diffraction pattern data into a pre-constructed projection potential function. The projection potential function is a linear combination of local functions of several types of atomic orbitals, and the terminal adjusts and calculates the parameters in the projection potential function according to these data. During the calculation process, the terminal considers the characteristics of the data and the characteristics of the projection potential function, and obtains the total projection potential through a series of mathematical operations.

[0059] The calculation process of the total projection potential is shown in formula (1):

[0060] (1)

[0061] Among them, represents the total projection potential, which is the potential function of all local orbit functions Linear summation.

[0062] S302. Perform outgoing wave calculation and processing based on the total projection potential and the multi-slice simulation imaging model to obtain the outgoing wave.

[0063] In the embodiments of the present application, after obtaining the total projection potential, the terminal substitutes it into the multi-slice simulation imaging model. The multi-slice simulation imaging model simulates the process of electrons propagating layer by layer in the sample, considering the interaction between electrons and sample atoms. The terminal calculates the outgoing wave of the electrons after passing through the sample according to the algorithm and related parameters of the model.

[0064] Another implementation: The terminal can use parallel computing to accelerate the calculation process of the outgoing wave. For example, allocate the calculation tasks of the multi-slice simulation imaging model to multiple processor cores or computing nodes for simultaneous calculation.

[0065] S303. Perform image calculation and processing based on the outgoing wave to obtain two-dimensional projection reconstruction information.

[0066] In the embodiments of the present application, the terminal performs image calculation and processing based on the calculated outgoing wave. This includes operations such as Fourier transform and intensity calculation on the outgoing wave to obtain image information reflecting the two-dimensional projection potential distribution of the sample, that is, two-dimensional projection reconstruction information.

[0067] Another implementation: The terminal can perform post-processing on the calculated two-dimensional projection reconstruction information, such as operations like contrast enhancement and sharpening, to improve the quality and visualization effect of the image.

[0068] Through the steps of the above embodiments, with the terminal as the execution entity, the two-dimensional projection potential of the sample can be effectively reconstructed. Utilizing the sensitivity of the diffraction pattern to the tilting angle, the terminal can correct the tilting angle deviation of the sample through an algorithm and obtain high-frequency information not available in the microscopic image, thereby more comprehensively understanding the microscopic structure of the sample. Using local orbitals to construct the projection potential function and constraining the solution space of the projection potential with the prior information of the locality of the atomic electrostatic potential makes up for the deficiencies of the existing projection potential reconstruction technology that relies on the known periodicity of the sample and has poor convergence, improving the accuracy and efficiency of the reconstruction.

[0069] In an exemplary embodiment, based on the above embodiments, please refer to Figure 4 , the multi-slice simulation imaging model of the embodiments of the present application includes an object function, a Fresnel propagation factor function, and an outgoing wave function; the process of performing outgoing wave calculation and processing based on the total projection potential and the multi-slice simulation imaging model to obtain the outgoing wave includes the following steps:

[0070] S401. Input the total projection potential into the object function to perform linear summation processing of the potential functions of several local orbital functions to obtain the object function value;

[0071] In the embodiments of the present application, after obtaining the total projection potential, the terminal inputs it into the object function. The object function is constructed by linearly adding the potential functions of several local orbital functions. The terminal decomposes the total projection potential into the contributions of each local orbital function according to the preset form of the local orbital function, and linearly adds the potential functions of these local orbital functions to obtain the object function value.

[0072] Among them, the form of the local orbital function includes, but is not limited to, the Gaussian function as shown in formula (2):

[0073] (2)

[0074] and the quartic polynomial function as shown in formula (3):

[0075] (3)

[0076] where the relative distance is as shown in formula (4):

[0077] (4)

[0078] while and represent the strength and radius of the local orbital function respectively, represent the coordinates of the local orbital function and any point in real space respectively.

[0079] The object function is as shown in formula (5):

[0080] (5)

[0081] S402: Input the thickness of each layer of the sample, the tilting angle of the sample relative to the incident electron beam, and the frequency coordinates of any point in the reciprocal space, which are obtained in advance, into the Fresnel propagation factor function for Fourier transform and exponential operation processing to obtain the Fresnel propagation factor value.

[0082] In the embodiments of the present application, the terminal pre-obtains the thickness of each layer of the sample, the tilting angle of the sample relative to the incident electron beam, and the frequency coordinates of any point in the reciprocal space. These parameters are input into the Fresnel propagation factor function, and the Fresnel propagation factor function includes Fourier transform and exponential operation. The terminal first performs Fourier transform on the relevant function to convert it from real space to reciprocal space, and then performs exponential operation to finally obtain the Fresnel propagation factor value.

[0083] Among them, the Fresnel propagation factor function is as shown in formula (6) and formula (7):

[0084] (6)

[0085] (7)

[0086] wherein represents the interaction coefficient, represents the phenomenological absorption coefficient, represents the object function, represents the Fresnel propagation factor, represents the thickness of each layer, represents the tilt angle of the sample relative to the incident electron beam, and respectively represent two-dimensional Fourier transform and inverse transform, represents the frequency coordinate of any point in the reciprocal space, represents the wavelength of the electron beam.

[0087] In S403, the Fresnel propagation factor value, the object function value, and the predetermined incident wave function value are input into the outgoing wave function for multiplication operation to obtain the outgoing wave.

[0088] In the embodiment of the present application, the terminal inputs the Fresnel propagation factor value, the object function value, and the predetermined incident wave function value obtained in the steps into the outgoing wave function. The outgoing wave function is obtained by multiplying the Fresnel propagation factor function with the object function and the incident wave function multiple times. The terminal performs multiplication operations in sequence according to the calculation rule of the outgoing wave function, and finally obtains the outgoing wave.

[0089] The outgoing wave function is shown in formula (8):

[0090] (8)

[0091] wherein, in formula (8), layer represents the set number of layers.

[0092] Through the steps of the above embodiments, the method of the embodiment of the present application can accurately calculate the outgoing wave according to the total projection potential and the multi-layer simulation imaging model. In the calculation of the object function, the potential functions of the local orbital functions are linearly added, fully considering the locality of the sample projection potential, and improving the accuracy of the object function calculation. In the calculation of the Fresnel propagation factor function, by combining parameters such as the layer thickness, tilt angle, and reciprocal space frequency coordinate of the sample, the propagation process of electrons in the sample can be more realistically simulated. In the calculation of the outgoing wave function, by multiplying the Fresnel propagation factor function with the object function and the incident wave function multiple times, the interaction between electrons and the sample is comprehensively considered, and the obtained outgoing wave can more accurately reflect the internal structure information of the sample.

[0093] In an exemplary embodiment, based on the above embodiment, please refer to Figure 5, the determination process of the incident wave function value in the embodiments of the present application includes the following steps:

[0094] S501, input the frequency coordinate of any point in the reciprocal space into the condenser aperture function for aperture calculation processing to obtain the condenser aperture value.

[0095] In the implementation of the present application, the terminal obtains the frequency coordinate k of any point in the reciprocal space and inputs it into the condenser aperture function. The terminal performs operations on the frequency coordinate k according to the mathematical expression of the condenser aperture function.

[0096] S502, input the frequency coordinate of any point in the reciprocal space, the wavelength of the electron beam obtained in advance, and the aberration coefficient into the lens aberration function for lens aberration calculation processing to obtain the lens aberration value.

[0097] In the embodiments of the present application, the terminal inputs the frequency coordinate of any point in the reciprocal space, the wavelength of the electron beam measured in advance, and the aberration coefficient into the lens aberration function. According to the meanings of the parameters in the formula, the terminal first calculates and , then performs a double summation operation, and finally takes the real part to obtain the lens aberration value. For example, when analyzing a certain crystal sample, the terminal obtains the values of k, and , and calculates step by step according to the formula to finally obtain the lens aberration value.

[0098] S503, input the condenser aperture value and the lens aberration value into the incident wave function for incident wave calculation processing to obtain the incident wave function value.

[0099] Among them, the above-mentioned incident wave function can have different expression forms according to the imaging conditions, including but not limited to the expression forms shown in formulas (9), (10), and (11):

[0100] (9)

[0101] (10)

[0102] (11)

[0103] Among them, represents the condenser aperture function, represents the lens aberration function, respectively represent the condenser and the objective lens, represents the real part of the complex function, ( ) represents the Zernike aberration coefficient, and respectively represent the dimensionless complex coordinate in the reciprocal space and its complex conjugate value. In particular, That is, it represents the defocus amount.

[0104] By determining the incident wave function value through the steps of the above embodiments, the influence of the condenser aperture and lens aberration on the electron beam wave function can be fully considered, making the determined incident wave function value more in line with the actual imaging situation. This helps to improve the accuracy of electron microscope imaging simulation. In the field of materials science, the microscopic structure of materials can be studied more precisely.

[0105] In an exemplary embodiment, based on the above embodiments, please refer to Figure 6 , the two-dimensional projection reconstruction information of the embodiments of the present application includes the simulated diffraction pattern intensity and the simulated image data under the defocus amount; the embodiments of the present application relate to the process of performing image calculation processing based on the exit wave to obtain the two-dimensional projection reconstruction information, including the following steps:

[0106] S601, input the exit wave into the diffraction pattern intensity function for intensity calculation processing to obtain the simulated diffraction pattern intensity.

[0107] In the embodiments of the present application, after obtaining the exit wave, the terminal inputs it into the diffraction pattern intensity function. First, the terminal performs a two-dimensional Fourier transform on the exit wave to convert it from real space to reciprocal space. Then, take the modulus of the result of the Fourier transform, and then perform a square operation on the modulus value to finally obtain the simulated diffraction pattern intensity.

[0108] S602, input the exit wave, the pre-determined objective lens transfer function value, and different defocus amounts into the image function for image calculation processing to obtain the simulated image data under the defocus amount.

[0109] In the embodiments of the present application, the terminal inputs the exit wave, the pre-determined objective lens transfer function value, and different defocus amounts into the image function. First, perform a two-dimensional Fourier transform on the exit wave, and then multiply the transformation result by the objective lens transfer function value. Then, perform a two-dimensional inverse Fourier transform on the multiplied result, and finally take the modulus and square it to obtain the simulated image data under different defocus amounts.

[0110] Among them, the diffraction pattern intensity function is shown in formula (12), the image function intensity function is shown in formula (13), and the objective lens transfer function is shown in formula (14):

[0111] (12)

[0112] (13)

[0113] (14)

[0114] Among them represents the diffraction pattern intensity, represents the defocus amount The calculated image below represents the objective lens transfer function represents the convergence semi-angle of the electron incident beam represents the focal length fluctuation of the objective lens represents the objective aperture function

[0115] Through the steps of the above embodiments, the simulated diffraction pattern intensity and the simulated image data at different defocus amounts can be accurately obtained from the outgoing wave, thereby constituting two-dimensional projection reconstruction information. The simulated diffraction pattern intensity can provide the crystal structure information of the sample, helping researchers understand the arrangement and periodic characteristics of atoms in the sample. The simulated image data at different defocus amounts can show the imaging effects of the sample under different defocus conditions, enabling researchers to observe the microscopic structure of the sample more comprehensively. This method combines the information of the diffraction pattern and the images at different defocus amounts, improving the accuracy and reliability of the analysis of the microscopic structure of the sample.

[0116] In an exemplary embodiment, based on the above embodiments, please refer to Figure 7 , the method of the embodiment of the present application further includes the following steps:

[0117] S701, Obtain experimental image data and experimental diffraction pattern intensity.

[0118] Among them, the experimental image data is the image data obtained through actual experimental operations, such as imaging the sample using equipment such as a transmission electron microscope. It truly reflects the imaging situation of the sample under experimental conditions. The experimental diffraction pattern intensity is the intensity information of the diffraction pattern recorded by the detector during the experiment when the electron beam irradiates the sample to generate a diffraction phenomenon, and it contains characteristics such as the crystal structure of the sample.

[0119] In the embodiment of the present application, the terminal establishes a communication connection with the experimental equipment (such as a transmission electron microscope) and receives the experimental image data and experimental diffraction pattern intensity of the sample collected by the experimental equipment. The terminal can perform format conversion and preprocessing on the received data to make it meet the requirements of subsequent calculations.

[0120] S702, Input the experimental image, experimental diffraction pattern intensity, simulated diffraction pattern intensity, and simulated image data at different defocus amounts into a pre-constructed loss function for error calculation processing to obtain an error value.

[0121] In the embodiment of the present application, the terminal inputs the experimental image data, experimental diffraction pattern intensity, simulated diffraction pattern intensity, and simulated image data at different defocus amounts into a pre-constructed loss function. The loss function will compare and calculate these data to obtain the error value between them. During the calculation process, the terminal will perform corresponding mathematical operations according to the specific form of the loss function, such as squaring, summing, averaging, etc.

[0122] Among them, the methods for calculating the loss function include but are not limited to the following forms, as shown in formulas (15), (16), (17), and (18):

[0123] (15)

[0124] (16)

[0125] (17)

[0126] (18)

[0127] where represents the summation over all incident electron beam conditions, represents the summation over all defocus amounts, represents at the incident beam and the defocus amount the image intensity acquired under the imaging conditions, represents the weight of the diffraction pattern residual relative to the image residual, represents at the incident beam the diffraction pattern intensity acquired under the imaging conditions, represents the function in the region the integral within.

[0128] S703. Optimize the parameters to be updated in the two-dimensional projection potential reconstruction model according to the error value, and obtain the optimized two-dimensional projection potential reconstruction model.

[0129] Among them, the parameters to be updated are the parameters in the two-dimensional projection potential reconstruction model that need to be adjusted and optimized.

[0130] In the embodiments of the present application, the terminal adjusts the parameters to be updated in the two-dimensional projection potential reconstruction model by using an optimization algorithm according to the calculated error value. The terminal calculates the gradient of the error value with respect to the parameters to be updated, and updates the values of the parameters according to the direction and magnitude of the gradient, so that the error value gradually decreases. The terminal continuously repeats this process until the convergence condition is met (such as the error value is less than the set threshold or the number of iterations reaches the upper limit).

[0131] Through the steps of the above embodiments, the terminal can continuously optimize the two-dimensional projection potential reconstruction model, making its prediction results closer to the actual experimental data. Obtaining the experimental image data and the experimental diffraction pattern intensity provides a real reference basis for model optimization, and calculating the error value through the loss function can accurately measure the prediction deviation of the model.

[0132] In an exemplary embodiment, based on the above embodiment, please refer to Figure 8, the embodiment of the present application relates to the process of optimizing the parameters to be updated in the two-dimensional projection potential reconstruction model according to the error value to obtain the optimized two-dimensional projection potential reconstruction model, including the following steps:

[0133] S801, calculate the gradient of the parameter to be updated.

[0134] Among them, the parameter to be updated includes at least one of the local orbital function intensity, local orbital function radius, local orbital function center coordinate, sample tilt angle, slice thickness, defocus amount, and aberration coefficient.

[0135] In the embodiment of the present application, after obtaining the error value, the terminal calculates the gradient of the parameter to be updated according to the functional relationship between the error value and the parameter to be updated by using the derivative rule. The parameters to be updated include the local orbital function intensity, local orbital function radius, local orbital function center coordinate, sample tilt angle, slice thickness, defocus amount, and aberration coefficient, etc. The terminal will find the partial derivative of the error value with respect to each parameter to be updated through methods such as the chain rule, so as to obtain the corresponding gradient.

[0136] S802, input the gradient into the update function for parameter update processing to obtain the optimized two-dimensional projection potential reconstruction model.

[0137] In the embodiment of the present application, the terminal inputs the gradient calculated in the previous step into the update function. The update function will adjust each parameter to be updated according to the direction and magnitude of the gradient, so as to achieve parameter update.

[0138] Among them, after calculating the gradient, the required parameters are updated in the following manner, as shown in formulas (19) to (25):

[0139] (19)

[0140] (20)

[0141] (21)

[0142] (22)

[0143] (23)

[0144] (24)

[0145] (25)

[0146] Among them and respectively represent two consecutive iteration times, 、 , , , , , are the learning rates of each parameter at the -th iteration, , , , , , , respectively represent the gradients of the loss function with respect to each parameter at the -th iteration.

[0147] Through the steps of the above embodiments, the terminal can effectively optimize the parameters to be updated in the two-dimensional projection potential reconstruction model according to the error value. Accurately calculating the gradients of the parameters to be updated provides a direction and basis for subsequent parameter updates. Updating the parameters through the update function enables the model to gradually converge to a better state. This optimization process can significantly improve the accuracy and reliability of the two-dimensional projection potential reconstruction model, enabling the model to more accurately reconstruct the two-dimensional projection potential of the sample.

[0148] In an exemplary embodiment, based on the above embodiments, the method of the embodiments of the present application further includes the following steps:

[0149] Step 1: Obtain experimental image data and experimental diffraction pattern intensity; input the experimental image, experimental diffraction pattern intensity, simulated diffraction pattern intensity, and simulated image data into a pre-constructed loss function for error calculation processing to obtain an error value;

[0150] Step 2: Calculate the gradients of the parameters to be updated; input the gradients into an update function for parameter update processing to obtain an optimized two-dimensional projection potential reconstruction model.

[0151] Step 3: Obtain high-resolution image data and diffraction pattern data of the sample under different imaging conditions; input the high-resolution image data and diffraction pattern data into the projection potential function in the optimized two-dimensional projection potential reconstruction model for projection potential calculation processing to obtain the total projection potential;

[0152] Step 4: Input the frequency coordinates of any point in the reciprocal space into the condenser aperture function for aperture calculation processing to obtain the condenser aperture value; input the frequency coordinates of any point in the reciprocal space, the wavelength of the electron beam obtained in advance, and the aberration coefficients into the lens aberration function for lens aberration calculation processing to obtain the lens aberration value; input the condenser aperture value and the lens aberration value into the incident wave function for incident wave calculation processing to obtain the incident wave function value;

[0153] Step 5: Input the total projected potential into the object function in the multi-slice simulation imaging model for linear summation of the potential functions of several local orbital functions to obtain the object function value; input the thickness of each slice of the sample, the tilting angle of the sample relative to the incident electron beam, and the frequency coordinates of any point in the reciprocal space, which are obtained in advance, into the Fresnel propagation factor function in the optimized multi-slice simulation imaging model for Fourier transform and exponential operation processing to obtain the Fresnel propagation factor value; input the Fresnel propagation factor value, the object function value, and the incident wave function value determined in advance into the outgoing wave function in the multi-slice simulation imaging model for multiplication operation processing to obtain the outgoing wave;

[0154] Step 6: Input the outgoing wave into the diffraction pattern intensity function for intensity calculation processing to obtain the optimized simulated diffraction pattern intensity; input the outgoing wave, the objective lens transfer function value determined in advance, and different defocus amounts into the image function for image calculation processing to obtain the optimized simulated image data; determine the optimized two-dimensional projected potential distribution according to the simulated diffraction pattern intensity and the simulated image data at different defocus amounts.

[0155] Next, a specific embodiment of the present application will be introduced:

[0156] 1. Collect high-resolution series of underfocus images and diffraction patterns of the Si

[110] zone axis under convergent beam illumination conditions, as Figures 9a to 9d shown. Only some series of underfocus images are shown in the figure. The actual defocus amount sequence is 7.2 - 127.2 nm, the defocus amount step is 10 nm, and there are 11 pictures in total.

[0157] 2. Initialize the potential function with a Gaussian function array, as Figure 10 shown.

[0158] 3. Simulate the series of underfocus images and diffraction patterns using the multi-slice method according to the imaging parameters.

[0159] 4. Calculate the loss function according to the above formula (15).

[0160] 5. Use the automatic differentiation mechanism to obtain the gradients of each parameter and update the parameters using an optimizer.

[0161] 6. Stop after reaching the set number of iterations, and output the reconstructed projected potential information, as well as the series of underfocus images and diffraction patterns, as Figure 11a 、 11b and shown in 11c.

[0162] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear indication in this document, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0163] Based on the same inventive concept, an embodiment of the present application also provides a two-dimensional projection potential reconstruction device based on local orbits for implementing the above-mentioned two-dimensional projection potential reconstruction method based on local orbits. The implementation solutions provided by this device to solve problems are similar to the implementation solutions described in the above method. Therefore, the specific limitations in one or more embodiments of the two-dimensional projection potential reconstruction device based on local orbits provided below can refer to the limitations on the two-dimensional projection potential reconstruction method in the above text, and will not be repeated here.

[0164] In one embodiment, as Figure 12 shown, a two-dimensional projection potential reconstruction device 900 based on local orbits is provided, including:

[0165] A data acquisition module 901, configured to acquire high-resolution image data and diffraction pattern data of a sample under different imaging conditions;

[0166] A reconstruction module 902, configured to input the high-resolution image data and diffraction pattern data into a pre-constructed two-dimensional projection potential reconstruction model for multi-slice simulated imaging calculation and image calculation processing to obtain two-dimensional projection reconstruction information of the sample; the two-dimensional projection potential reconstruction model is constructed based on a projection potential function and a multi-slice simulated imaging model; the projection potential function is constructed based on local functions of several types of atomic orbitals; the two-dimensional projection reconstruction information is used to reflect the two-dimensional projection potential distribution of the sample.

[0167] In one of the embodiments, the above reconstruction module is specifically configured to input the high-resolution image data and diffraction pattern data into the projection potential function for projection potential calculation processing to obtain a total projection potential; perform outgoing wave calculation processing based on the total projection potential and the multi-slice simulated imaging model to obtain an outgoing wave; perform image calculation processing based on the outgoing wave to obtain two-dimensional projection reconstruction information.

[0168] In one embodiment, the above-mentioned multi-layer simulation imaging model includes an object function, a Fresnel propagation factor function, and an exit wave function; the above-mentioned reconstruction module is further specifically configured to input the total projection potential into the object function for linearly adding the potential functions of a plurality of local orbital functions to obtain an object function value; input the thickness of each layer of the sample, the tilting angle of the sample relative to the incident electron beam, and the frequency coordinates of any point in the reciprocal space, which are pre-acquired, into the Fresnel propagation factor function for Fourier transform and exponential operation processing to obtain a Fresnel propagation factor value; input the Fresnel propagation factor value, the object function value, and the pre-determined incident wave function value into the exit wave function for multiplication operation processing to obtain an exit wave.

[0169] In one embodiment, the above-mentioned reconstruction module is further specifically configured to input the frequency coordinates of any point in the reciprocal space into the condenser aperture function for aperture calculation processing to obtain a condenser aperture value; input the frequency coordinates of any point in the reciprocal space, the wavelength of the electron beam pre-acquired, and the aberration coefficients into the lens aberration function for lens aberration calculation processing to obtain a lens aberration value; input the condenser aperture value and the lens aberration value into the incident wave function for incident wave calculation processing to obtain an incident wave function value.

[0170] In one embodiment, the above-mentioned two-dimensional projection reconstruction information includes the simulated diffraction pattern intensity and the simulated image data under defocus; the above-mentioned reconstruction module is further specifically configured to input the exit wave into the diffraction pattern intensity function for intensity calculation processing to obtain the simulated diffraction pattern intensity; input the exit wave, the pre-determined objective lens transfer function value, and different defocus values into the image function for image calculation processing to obtain the simulated image data under defocus.

[0171] In one embodiment, the above-mentioned device further includes:

[0172] An experimental data acquisition module, configured to acquire experimental image data and experimental diffraction pattern intensity;

[0173] An error calculation module, configured to input the experimental image, the experimental diffraction pattern intensity, the simulated diffraction pattern intensity, and the simulated image data under defocus into a pre-constructed loss function for error calculation processing to obtain an error value;

[0174] A parameter optimization module, configured to optimize the parameters to be updated in the two-dimensional projection potential reconstruction model according to the error value to obtain an optimized two-dimensional projection potential reconstruction model.

[0175] In one embodiment, the above parameter optimization module is specifically configured to calculate the gradient of the parameter to be updated; the parameter to be updated includes at least one of the intensity of the local orbital function, the radius of the local orbital function, the central coordinates of the local orbital function, the sample tilting angle, the layer thickness, the defocus amount, and the aberration coefficient; and input the gradient into an update function for parameter update processing to obtain an optimized two-dimensional projection potential reconstruction model.

[0176] Each module in the above two-dimensional projection potential reconstruction device based on local orbitals can be implemented in whole or in part by software, hardware, or a combination thereof. The above modules can be embedded in or independent of a processor in a computer device in the form of hardware, or stored in a memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0177] In one embodiment, a computer device is provided. The computer device can be a terminal, and its internal structure diagram can be as Figure 13 shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface, the display unit, and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used for exchanging information between the processor and external devices. The communication interface of the computer device is used for communicating with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a two-dimensional projection potential reconstruction method based on local orbitals. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the computer device housing, or an external keyboard, touchpad, or mouse, etc.

[0178] Those skilled in the art can understand that Figure 13 the structure shown in

[0179] In one embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory. When the processor executes the computer program, the following steps are implemented:

[0180] Obtain high-resolution image data and diffraction pattern data of a sample under different imaging conditions;

[0181] Input the high-resolution image data and the diffraction pattern data into a pre-constructed two-dimensional projection potential reconstruction model for multi-slice simulation imaging calculation and image calculation processing to obtain two-dimensional projection reconstruction information of the sample; the two-dimensional projection potential reconstruction model is constructed based on a projection potential function and a multi-slice simulation imaging model; the projection potential function is constructed based on local functions of several types of atomic orbitals; the two-dimensional projection reconstruction information is used to reflect the two-dimensional projection potential distribution of the sample.

[0182] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0183] Input the high-resolution image data and the diffraction pattern data into the projection potential function for projection potential calculation processing to obtain the total projection potential;

[0184] Perform outgoing wave calculation processing according to the total projection potential and the multi-slice simulation imaging model to obtain the outgoing wave;

[0185] Perform image calculation processing according to the outgoing wave to obtain two-dimensional projection reconstruction information.

[0186] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0187] Input the total projection potential into the object function for potential function linear summation processing of several local orbital functions to obtain the object function value;

[0188] Input the thickness of each slice of the sample, the tilt angle of the sample relative to the incident electron beam, and the frequency coordinates of any point in the reciprocal space obtained in advance into the Fresnel propagation factor function for Fourier transform and exponential operation processing to obtain the Fresnel propagation factor value;

[0189] Input the Fresnel propagation factor value, the object function value, and the incident wave function value determined in advance into the outgoing wave function for multiplication operation processing to obtain the outgoing wave.

[0190] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0191] Input the frequency coordinates of any point in the reciprocal space into the condenser aperture function for aperture calculation processing to obtain the condenser aperture value;

[0192] Input the frequency coordinates of any point in the reciprocal space, the wavelength of the pre-acquired electron beam, and the aberration coefficients into the lens aberration function for lens aberration calculation processing to obtain the lens aberration value;

[0193] Input the condenser aperture value and the lens aberration value into the incident wave function for incident wave calculation processing to obtain the incident wave function value.

[0194] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0195] Input the exit wave into the diffraction pattern intensity function for intensity calculation processing to obtain the simulated diffraction pattern intensity;

[0196] Input the exit wave, the pre-determined objective lens transfer function value, and different defocus amounts into the image function for image calculation processing to obtain the simulated image data at the defocus amount.

[0197] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0198] Obtain the experimental image data and the experimental diffraction pattern intensity;

[0199] Input the experimental image, the experimental diffraction pattern intensity, the simulated diffraction pattern intensity, and the simulated image data at the defocus amount into the pre-constructed loss function for error calculation processing to obtain the error value;

[0200] Optimize the parameters to be updated in the two-dimensional projection potential reconstruction model according to the error value to obtain the optimized two-dimensional projection potential reconstruction model.

[0201] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0202] Calculate the gradient of the parameters to be updated; the parameters to be updated include at least one of the local orbital function intensity, the local orbital function radius, the local orbital function center coordinates, the sample tilt angle, the slice thickness, the defocus amount, and the aberration coefficients;

[0203] Input the gradient into the update function for parameter update processing to obtain the optimized two-dimensional projection potential reconstruction model.

[0204] According to some embodiments of the present application, a computer program product is further provided. When the computer program is executed by a processor, the above method can be implemented. The computer program product includes one or more computer instructions. When these computer instructions are loaded and executed on a computer, part or all of the above method can be implemented in accordance with the process or function described in the embodiments of the present application.

[0205] According to some embodiments of the present application, there is also provided a non-transitory computer-readable storage medium including instructions, such as a memory including instructions, and the above instructions can be executed by a processor of an electronic device to complete the above method. For example, the non-transitory computer-readable storage medium may be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0206] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data that have been authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant laws, regulations, and standards of relevant countries and regions.

[0207] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.

[0208] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, 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, it should be considered as the scope described in this specification.

[0209] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A two-dimensional projection potential reconstruction method based on local orbits, characterized in that: The method comprises: Obtain high-resolution image data and diffraction pattern data of samples under different imaging conditions; The high-resolution image data and diffraction pattern data are input into a pre-constructed two-dimensional projection potential reconstruction model to perform multi-layer simulation imaging calculation and image calculation processing to obtain the two-dimensional projection reconstruction information of the sample; the two-dimensional projection potential reconstruction model is constructed based on the projection potential function and the multi-layer simulation imaging model; the projection potential function is constructed based on the local functions of several types of atomic orbitals; the two-dimensional projection reconstruction information is used to reflect the two-dimensional projection potential distribution of the sample.

2. The method according to claim 1, characterized in that: The step of inputting the high-resolution image data and the diffraction pattern data into a pre-constructed two-dimensional projection potential reconstruction model to perform multi-layer simulation imaging calculation and image calculation processing to obtain two-dimensional projection reconstruction information of the sample includes: Inputting the high-resolution image data and the diffraction pattern data into the projection potential function to perform projection potential calculation processing to obtain a total projection potential; Performing exit wave calculation processing according to the total projection potential and the multi-slice simulated imaging model to obtain an exit wave; Image calculation processing is performed according to the outgoing wave to obtain the two-dimensional projection reconstruction information.

3. The method according to claim 2, characterized in that The multi-layer simulated imaging model includes a physical function, a Fresnel propagation factor function and an exit wave function; the exit wave calculation processing is performed according to the total projection potential and the multi-layer simulated imaging model to obtain the exit wave including: Input the total projection potential into the physical function to perform linear addition processing on the potential functions of several local orbital functions to obtain the physical function value; Input the pre-acquired thickness of each layer of the sample, the tilt angle of the sample relative to the incident electron beam, and the frequency coordinate of any point in the reciprocal space into the Fresnel propagation factor function for Fourier transformation and exponential operation processing to obtain the Fresnel propagation factor value; The Fresnel propagation factor value, the material function value and a predetermined incident wave function value are input into the output wave function for multiplication processing to obtain the output wave.

4. The method according to claim 3, characterized in that: The process of determining the incident wave function value includes: Inputting the frequency coordinate of any point in the reciprocal space into the condenser aperture function to perform aperture calculation processing to obtain the condenser aperture value; The frequency coordinate of any point in the reciprocal space, the wavelength of the electron beam acquired in advance, and the aberration coefficient are input into the lens aberration function to perform lens aberration calculation processing to obtain the lens aberration value; The condenser aperture value and the lens aberration value are input into the incident wave function to perform incident wave calculation processing to obtain the incident wave function value.

5. The method according to claim 2, characterized in that: The two-dimensional projection reconstruction information includes simulated image data under simulated diffraction pattern intensity and defocus amount; the image calculation processing is performed according to the outgoing wave to obtain the two-dimensional projection reconstruction information, including: Inputting the outgoing wave into a diffraction pattern intensity function to perform intensity calculation processing to obtain a simulated diffraction pattern intensity; The output wave, a predetermined objective lens transfer function value and different defocus amounts are input into an image function for image calculation processing to obtain simulated image data under the defocus amount.

6. The method according to claim 5, characterized in that The method further comprises: Acquire experimental image data and experimental diffraction pattern intensity; Inputting the experimental image, the experimental diffraction pattern intensity, the simulated diffraction pattern intensity and the simulated image data under the defocus amount into a pre-constructed loss function for error calculation processing to obtain an error value; The parameters to be updated in the two-dimensional projection potential reconstruction model are optimized according to the error value to obtain an optimized two-dimensional projection potential reconstruction model.

7. The method according to claim 6, characterized in that The step of optimizing the parameters to be updated in the two-dimensional projection potential reconstruction model according to the error value to obtain an optimized two-dimensional projection potential reconstruction model comprises: Calculating the gradient of the parameter to be updated; the parameter to be updated includes at least one of the local orbit function intensity, the local orbit function radius, the local orbit function center coordinates, the sample tilt angle, the slice thickness, the defocus amount and the aberration coefficient; The gradient is input into an update function to perform parameter update processing to obtain the optimized two-dimensional projection potential reconstruction model.

8. A two-dimensional projection potential reconstruction device based on local orbits, characterized in that: The device comprises: A data acquisition module, used to acquire high-resolution image data and diffraction pattern data of the sample under different imaging conditions; A reconstruction module is used to input the high-resolution image data and diffraction pattern data into a pre-constructed two-dimensional projection potential reconstruction model to perform multi-layer simulation imaging calculation and image calculation processing to obtain the two-dimensional projection reconstruction information of the sample; the two-dimensional projection potential reconstruction model is constructed based on the projection potential function and the multi-layer simulation imaging model; the projection potential function is constructed based on the local functions of several types of atomic orbitals; the two-dimensional projection reconstruction information is used to reflect the two-dimensional projection potential distribution of the sample.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images