CN117451626A - Stacked imaging method and apparatus including sample shape optimization - Google Patents
Stacked imaging method and apparatus including sample shape optimization Download PDFInfo
- Publication number
- CN117451626A CN117451626A CN202311414552.2A CN202311414552A CN117451626A CN 117451626 A CN117451626 A CN 117451626A CN 202311414552 A CN202311414552 A CN 202311414552A CN 117451626 A CN117451626 A CN 117451626A
- Authority
- CN
- China
- Prior art keywords
- sample
- parameters
- optimized
- parameter
- function
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000003384 imaging method Methods 0.000 title claims abstract description 43
- 238000005457 optimization Methods 0.000 title claims abstract description 43
- 238000000034 method Methods 0.000 claims abstract description 56
- 230000008569 process Effects 0.000 claims abstract description 33
- 238000002447 crystallographic data Methods 0.000 claims abstract description 13
- 230000006870 function Effects 0.000 claims description 145
- 238000004590 computer program Methods 0.000 claims description 7
- 238000003860 storage Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000005428 wave function Effects 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 12
- 230000003287 optical effect Effects 0.000 abstract description 11
- 238000009826 distribution Methods 0.000 abstract description 8
- 238000010894 electron beam technology Methods 0.000 description 16
- 239000013078 crystal Substances 0.000 description 12
- 239000000463 material Substances 0.000 description 12
- 238000004422 calculation algorithm Methods 0.000 description 11
- 238000004891 communication Methods 0.000 description 8
- 230000008859 change Effects 0.000 description 7
- 238000002474 experimental method Methods 0.000 description 7
- 238000004088 simulation Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 238000002360 preparation method Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000037303 wrinkles Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 2
- 238000003491 array Methods 0.000 description 2
- 125000004429 atom Chemical group 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 238000003475 lamination Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 125000004434 sulfur atom Chemical group 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/01—Arrangements or apparatus for facilitating the optical investigation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
Abstract
The application relates to a laminated imaging method and device comprising sample shape optimization, wherein the method comprises the following steps: scanning the sample to acquire diffraction information of a corresponding scanning point, and generating a diffraction data set according to the diffraction information; when a forward propagation process is constructed, the shape parameters of the sample are contained as variable parameters in the forward propagation process, and a loss function of a forward propagation model is calculated; solving the gradient of the loss function with respect to the parameter to be optimized, and optimizing the parameter to be optimized according to the gradient; and recalculating forward propagation based on the parameters to be optimized, calculating a loss function until a preset iteration termination condition is met, and outputting final parameters to be optimized, wherein the sample shape parameters according to the final parameters to be optimized are used as parameters for reflecting the shape of the sample in a three-dimensional space. Therefore, the problems that in the related art, the optical path difference is changed due to the complexity of experimental conditions, the decoupling effect and the reconstruction accuracy are further reduced, the error is increased, the spatial distribution information is difficult to acquire and the like are solved.
Description
Technical Field
The present disclosure relates to the field of computational imaging, and in particular, to a stacked imaging method and apparatus including sample shape optimization.
Background
In the related art, the stacked imaging technique refers to a computational imaging technique, through which sample information and light source information can be obtained simultaneously, and in an ideal case, a sample function and a light source function can be completely decoupled, so as to realize zero aberration imaging.
However, in the related art, the conditions in the experiment are often complex, the normal direction of the sample surface and the direction of the crystal band axis always have deviation, the sample needs to be rotated to ensure that the crystal band axis of the sample is parallel to the incident beam, so that the normal direction of the sample surface deviates from the direction of the incident beam, and the optical path difference when the incident beam reaches different positions on the sample surface is further different, thereby affecting the decoupling effect of the sample information and the light source information, reducing the accuracy of the sample information obtained by reconstruction.
Disclosure of Invention
The application provides a laminated imaging method and device comprising sample shape optimization, which are used for solving the problems that in the related technology, because conditions in experiments are often complex, deviation of a normal direction of a sample surface and a crystal band axis direction possibly causes change of an optical path difference, and further decoupling effect and reconstruction accuracy are reduced, and in addition, because two-dimensional materials are wrinkled, errors are increased, spatial distribution information is difficult to obtain and the like.
An embodiment of a first aspect of the present application provides a stack imaging method including sample shape optimization, comprising the steps of: scanning a sample to acquire diffraction information of a corresponding scanning point, and generating a diffraction data set according to the diffraction information; when a forward propagation process is constructed, taking the shape parameter of the sample as a variable parameter, including the variable parameter into the forward propagation process, and calculating a loss function of a forward propagation model; solving the gradient of the loss function with respect to the parameter to be optimized, and optimizing the parameter to be optimized according to the gradient; and based on the parameters to be optimized, recalculating forward propagation, calculating the loss function until a preset iteration termination condition is met, and outputting final parameters to be optimized, wherein the sample shape parameters according to the final parameters to be optimized are taken as parameters for reflecting the shape of the sample in a three-dimensional space.
Optionally, in one embodiment of the present application, the parameters to be optimized include at least one of an objective function, a light source function, and the sample shape parameter.
Optionally, in an embodiment of the present application, before the sample shape parameter is taken as the variable parameter, the method further includes: initializing the object function, the light source function, and the sample shape parameter.
Alternatively, in one embodiment of the present application, the outgoing wave function of the forward propagation model may be:
wherein Δz represents the thickness of each layer function, j represents the number of scan positions, r represents the coordinates in real space, P (r-r) j ) For an incident beam scanned to the j-th position,for the distance of pre-propagation of the incident beam scanned to the jth position, O N (r) is an object function of the nth layer.
Alternatively, in one embodiment of the present application, the preset iteration termination condition may be, but is not limited to, converging on the loss function or reaching a preset number of iterations.
Embodiments of a second aspect of the present application provide a stacked imaging apparatus including sample shape optimization, comprising: the generation module is used for scanning the sample to acquire diffraction information of a corresponding scanning point and generating a diffraction data set according to the diffraction information; the calculation module is used for taking the shape parameter of the sample as a variable parameter when constructing a forward propagation process, and calculating a loss function of a forward propagation model in the forward propagation process; the solving module is used for solving the gradient of the loss function on the parameter to be optimized and optimizing the parameter to be optimized according to the gradient; and the iteration module is used for recalculating forward propagation based on the parameters to be optimized, calculating the loss function until a preset iteration termination condition is met, and outputting final parameters to be optimized, wherein the sample shape parameters according to the final parameters to be optimized are used as parameters for reflecting the shape of the sample in a three-dimensional space.
Optionally, in one embodiment of the present application, the parameters to be optimized include at least one of an objective function, a light source function, and the sample shape parameter.
Optionally, in one embodiment of the present application, further includes: an initialization module for initializing the object function, the light source function and the sample shape parameter before taking the sample shape parameter as the variable parameter.
An embodiment of a third aspect of the present application provides an electronic device, including: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor executing the program to implement a stacked imaging method including sample shape optimization as described in the above embodiments.
A fourth aspect of the present application provides a computer readable storage medium storing a computer program which when executed by a processor implements a stacked imaging method as above comprising sample shape optimization.
According to the method and the device, under the condition that the diffraction data set is utilized to recover the shape parameters of the sample in the three-dimensional space, the shape parameters of the sample are introduced into the algorithm, and repeated iterative optimization is carried out in the reconstruction process, so that deviation between the normal direction of the sample and the direction of the band axis of the crystal is avoided, high-precision sample information is obtained, the preparation difficulty of the sample is reduced, the stability of the stacked imaging algorithm is improved, and meanwhile, the shape parameters of the sample in the three-dimensional space can be directly obtained. Therefore, the problems that in the related art, due to the fact that conditions in experiments are complex, the deviation of the normal direction of the sample surface and the direction of the crystal band axis possibly causes the change of optical path difference, the decoupling effect and the reconstruction accuracy are further reduced, in addition, due to the fact that wrinkles exist in the two-dimensional material, errors are increased, space distribution information is difficult to obtain and the like are solved.
Additional aspects and advantages of the application will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the application.
Drawings
The foregoing and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings, in which:
FIG. 1 is a flow chart of a stack imaging method including sample shape optimization according to an embodiment of the present application;
FIG. 2 is a flow chart of a stack imaging method including sample shape optimization according to one embodiment of the present application;
FIG. 3 is a surface tilted bilayer MoS for use in a stacked imaging algorithm according to one embodiment of the present application 2 A structural schematic;
FIG. 4 is a schematic representation of a reconstructed wide range of objective functions according to one embodiment of the present application;
FIG. 5 is a schematic representation of a reconstructed input beam function according to one embodiment of the present application;
FIG. 6 is a schematic representation of a reconstructed sample shape function according to one embodiment of the present application;
FIG. 7 is a schematic diagram showing a comparison of the reconstruction effect of the present application and related art according to one embodiment of the present application;
FIG. 8 is a broken line plot of the values of a loss function with an iterative process according to one embodiment of the present application;
FIG. 9 is a schematic structural view of a stacked imaging apparatus including sample shape optimization according to an embodiment of the present application;
fig. 10 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
Embodiments of the present application are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary and intended for the purpose of explaining the present application and are not to be construed as limiting the present application.
A stack imaging method, apparatus, electronic device, and storage medium including sample shape optimization according to embodiments of the present application are described below with reference to the accompanying drawings. In the related art mentioned in the background technology center, because conditions in experiments are often complex, deviation between a normal direction of a sample surface and a crystal band axis direction may cause variation of an optical path difference, so that decoupling effect and reconstruction accuracy are reduced, in addition, because two-dimensional materials have folds, errors are increased, spatial distribution information is difficult to obtain and the like, the application provides a laminated imaging method comprising sample shape optimization, in the method, under the condition that a diffraction data set is utilized to recover shape parameters of a sample in a three-dimensional space, the shape parameters of the sample can be introduced into an algorithm, repeated iterative optimization is performed in a reconstruction process, so that deviation between the normal direction of the sample and the crystal band axis direction is avoided, high-accuracy sample information is obtained, preparation difficulty of the sample is reduced, stability of a laminated imaging algorithm is improved, and meanwhile, the shape parameters of the sample in the three-dimensional space can be directly obtained. Therefore, the problems that in the related art, due to the fact that conditions in experiments are complex, the deviation of the normal direction of the sample surface and the direction of the crystal band axis possibly causes the change of optical path difference, the decoupling effect and the reconstruction accuracy are further reduced, in addition, due to the fact that wrinkles exist in the two-dimensional material, errors are increased, space distribution information is difficult to obtain and the like are solved.
Specifically, fig. 1 is a schematic flow chart of a stacked imaging method including sample shape optimization according to an embodiment of the present application.
As shown in fig. 1, the stack imaging method including sample shape optimization includes the steps of:
in step S101, a sample is scanned to collect diffraction information corresponding to the scanned points, and a diffraction data set is generated according to the diffraction information.
It is understood that diffraction information refers to information about objects contained in interference and scattering effects of light waves generated by diffraction phenomena, including information about spatial distribution of diffraction patterns, positions of interference fringes, diffraction angles, phase differences, and the like.
Specifically, the embodiment of the application collects the convergence beam diffraction information of the scanning points which are regularly arranged or randomly distributed in the two-dimensional sample plane, can obtain experimental data containing the scanning positions and diffraction information corresponding to the scanning positions, can include information such as diffraction angles and diffraction intensities of each scanning point, and provides a data base for subsequent analysis and processing.
In step S102, in constructing the forward propagation process, the sample shape parameter is included as a variable parameter into the forward propagation process, and a loss function of the forward propagation model is calculated.
It is understood that sample shape parameters refer to variable parameters describing the shape characteristics of the sample to control the geometric properties and characteristics of the shape.
In actual implementation, the embodiment of the application takes the shape parameter of the sample as a variable parameter when constructing the forward propagation process, and comprises the steps of calculating a loss function of a forward propagation model in the forward propagation process, wherein a light source used for imaging comprises but is not limited to electrons, X rays and visible light, and furthermore, the setting method of the shape parameter of the sample comprises describing the shape of the sample by using two inclination angles after approximating the sample as a perfect sheet (for example, adding surface inclination determined by two angles to the sample or describing the shape of the sample by using differential incident beams, adding free propagation distance to the incident beam at each scanning position, determining the free propagation distance by two inclination angles of the sample) and taking fluctuation of the surface of the sample into consideration, describing the shape of the sample by using one fluctuation surface (for example, adding surface inclination to the sample, the inclination degree of each position can be freely changed or differential incident beams can be used for describing the shape of the sample, the free propagation distance is added for the incident beams of each scanning position, the free propagation distance of the incident beams of each scanning position is independently optimized, the shape parameters of the sample are taken as variable parameters and are included in the forward propagation process, the loss function of a forward propagation model is calculated, for example, the shape parameters (for example, coordinate values or local geometric features) of the sample can be input, the prediction result of the three-dimensional shape or structure of the sample is output, the error between the prediction result and a real sample can be obtained by using the loss function of the forward propagation model, and a regularization term such as smoothness or topological relation of the shape parameters can be added to restrict the optimization process.
According to the embodiment of the application, the shape parameters of the sample are used as the variable parameters and are included in the forward propagation process, the loss function of the forward propagation model is calculated, the shape parameters can be adjusted to generate a better prediction result, the fineness and the flexibility of controlling the shape parameters are improved, and therefore accurate reconstruction and optimization can be achieved.
Optionally, in an embodiment of the present application, before taking the sample shape parameter as the variable parameter, the method further includes: initializer function, light source function and sample shape parameters.
It will be understood that the object function refers to a function or model describing the material properties of the object surface, the light source function refers to a function or model describing the light source properties, and the sample shape parameters refer to variable parameters describing the shape characteristics of the sample.
Specifically, before the sample shape parameter is used as the variable parameter, the embodiment of the application further comprises an initialization function, a light source function and the sample shape parameter, wherein the initialization function can provide a good starting point for the model, the generated image or shape can have a realistic illumination effect in the initial stage by reasonably selecting the initial value of the light source function, and the model can be quickly converged to the expected shape space by proper sample shape parameter initialization, so that the stability of the model can be improved, the authenticity of a prediction result is improved, better shape control and flexibility can be provided, the parameter change can be responded better, and the accurate control and adjustment of the generated shape can be realized.
Alternatively, in one embodiment of the present application, the exit wave function of the forward propagation model may be:
wherein Δz represents the thickness of each layer of the function, j representsThe number of the table scanning position, r, represents the coordinates in real space, P (r-r j ) For an incident beam scanned to the j-th position,for the distance of pre-propagation of the incident beam scanned to the jth position, O N (r) is an object function of the nth layer.
Specifically, Δz represents the thickness of each layer function, j represents the serial number of the scanning position, r represents the coordinates in real space, P (r-r) j ) For an incident beam scanned to the j-th position,for the distance of pre-propagation of the incident beam scanned to the jth position, O N (r) is an object function of the nth layer, specifically expressed as:
p(k；z)＝exp(-iπzλk 2 )，
wherein P is z {.cndot } is the Fresnel near field diffraction acting factor, and z is the propagation distance of near field diffraction.
The embodiment of the application can help to rapidly process a large amount of data, better predict and adjust, effectively understand and evaluate the influence of the shape parameters of the sample, optimize the generated result and make necessary decisions and adjustments by using the forward propagation model.
In step S103, the gradient of the loss function with respect to the parameter to be optimized is solved, and the parameter to be optimized is optimized according to the gradient.
It will be appreciated that the method of optimizing parameters to be optimized according to the gradient comprises: the derivative of the error function with respect to the sample shape parameter is directly solved, updated by methods including but not limited to gradient descent or obtained by automatic derivative techniques, updated and optimized by methods including but not limited to gradient descent.
Specifically, the gradient optimization parameters to be optimized include:
wherein O is i (r) is an object function of the i-th layer,as object function O i (r)，α P For the incident beam function P->For the learning rate of the vacuum of the upper surface of the sample (i.e. the distance of the pre-propagation of the incident beam for each scanning position)>Gradient as object function->Gradient of the input beam function->Is the gradient of the upper surface of the sample.
According to the method and the device for optimizing the parameters, the gradient of the loss function on the parameters to be optimized is solved, the parameters to be optimized are optimized through the gradient, optimization, acceleration convergence, parameter control and model trainability increase of the model can be achieved, performance and flexibility of the model are improved, and generation results or prediction accuracy are improved.
Optionally, in an embodiment of the present application, the parameters to be optimized include at least one of an objective function, a light source function, and a sample shape parameter.
It is understood that the parameters to be optimized include at least one of an object function, a light source function and a sample shape parameter, wherein the object function refers to a function or model describing the material characteristics of the surface of the object, and can be used for defining how the object reflects, absorbs or transmits light, sound and other fluctuation or signals, and by adjusting the parameters of the object function, the appearance, the material and the optical or acoustic characteristics of the object can be changed; the light source function refers to a function or model describing the characteristics of the light source, can be used for simulating and describing the intensity, color, direction and other attributes of the light source, and can realize the control of illumination conditions by adjusting the parameters of the light source function so as to influence the appearance of the generated image or shape; the sample shape parameter refers to a parameter describing the shape characteristics of the sample, and can be used for controlling the characteristics of the shape, such as the size, the geometric structure, the curvature and the like, and the generated shape can be deformed, reconstructed or edited by adjusting the value of the sample shape parameter.
According to the embodiment of the application, the quality and fidelity of the generated result are improved through optimizing the object function, the light source function and the sample shape parameters, and the controllability and the flexibility of the shape are improved, so that the reconstruction and the optimization of the shape are realized.
In step S104, based on the parameters to be optimized, forward propagation is recalculated, and a loss function is calculated until a preset iteration termination condition is satisfied, and the final parameters to be optimized are output, so that the sample shape parameters according to the final parameters to be optimized are used as parameters for reflecting the shape of the sample in the three-dimensional space.
It can be appreciated that the embodiment of the present application may scan the sample by using an electron beam to obtain an object function, an electron beam function, and a sample shape function, and the steps are as follows, in conjunction with fig. 2:
step S1, scanning a sample by an electron beam, and collecting a diffraction pattern of each scanning point of the sample.
Step S2, initializing an object function, an electron beam function and a sample shape function.
And step S3, carrying out pre-propagation treatment on the electron beam according to the shape parameters of the sample.
And S4, calculating forward propagation simulation data according to the object and the electron beam function after the pre-propagation.
And S5, calculating a loss function of simulation and experimental data of the forward propagation process.
And S6, solving the gradient of the loss function with respect to the optimization parameters, and optimizing the object function, the electron beam function and other parameters to be optimized according to the gradient.
And S7, judging that the iteration convergence termination condition is met.
Step S8, obtaining an object function, an electron beam function and a sample shape function.
In the optimization process, the parameters to be optimized are input into forward propagation to recalculate the loss function, and the parameters are continuously optimized according to the preset iteration termination condition (such as the convergence of the loss function or the preset iteration times) until the termination condition is met, and finally the output parameters to be optimized comprise the shape parameters of the sample and can be used as the parameters for reflecting the shape of the sample in the three-dimensional space, so that the quality and accuracy of a generated result are improved, the diversity and the adaptability of the shape are controlled, and the generation efficiency and the shape controllability are enhanced.
Optionally, in an embodiment of the present application, the preset iteration termination condition is that the loss function converges or reaches a preset number of iterations.
Specifically, the preset iteration termination condition is that the loss function converges or reaches a preset iteration number, where when the change of the loss function is small, it can be understood that a relatively stable state is reached, it can be judged that the loss function has converged, a relatively optimal state is reached, at this time, the iteration process can be stopped, in addition, a preset iteration number can be set, when the iteration number reaches a set value, the iteration process can be stopped, for example, by optimizing a sample shape parameter to realize sample reconstruction, the preset iteration number can be set to 1000, after 1000 iterations, even though the loss function may not completely converge, we can stop the iteration, and consider that the obtained generated result has higher quality and accuracy.
According to the method and the device, the generated result can be ensured to be close to the target result enough through the convergence of the loss function as the termination condition, so that the quality and the accuracy of the generated result are improved, and the acceptable generated result can be ensured to be obtained within the specified iteration times by setting the preset iteration times as the termination condition, so that the time and the resource use of the optimization process are effectively controlled.
A stack imaging method including sample shape optimization according to an embodiment of the present application is described in detail in one embodiment with reference to fig. 2, 3, 4, 5, 6, 7, and 8.
It can be appreciated that the embodiment of the present application may scan the sample by using an electron beam to obtain an object function, an electron beam function, and a sample shape function, and the steps are as follows, in conjunction with fig. 2:
step S1, scanning a sample by an electron beam, and collecting a diffraction pattern of each scanning point of the sample.
Step S2, initializing an object function, an electron beam function and a sample shape function.
And step S3, carrying out pre-propagation treatment on the electron beam according to the shape parameters of the sample.
And S4, calculating forward propagation simulation data according to the object and the electron beam function after the pre-propagation.
And S5, calculating a loss function of simulation and experimental data of the forward propagation process.
And S6, solving the gradient of the loss function with respect to the optimization parameters, and optimizing the object function, the electron beam function and other parameters to be optimized according to the gradient.
And S7, judging that the iteration convergence termination condition is met.
Step S8, obtaining an object function, an electron beam function and a sample shape function.
It can be understood that, the sample is scanned by the light source to acquire diffraction information of a corresponding scanning point, a diffraction data set is obtained, in the process of constructing forward propagation, an initializer function, a light source function and a sample shape parameter are included as variable parameters in the forward propagation process of a lamination algorithm, a loss function of a forward propagation model is calculated, then, a gradient of a parameter to be optimized is solved according to the loss function, the parameter to be optimized is optimized through the gradient, wherein the parameter to be optimized comprises the initializer function, the light source function and the sample shape parameter, the forward propagation recalculation is continuously executed, the loss function is calculated until a preset iteration termination condition is met, the optimized parameter to be optimized is output, and accurate reconstruction and analysis of the sample shape can be realized.
Specifically, the outgoing wave function in the forward propagation model is:
wherein Δz represents the thickness of each layer function, j represents the number of scan positions, r represents the coordinates in real space, P (r-r) j ) For an incident beam scanned to the j-th position,for the distance of pre-propagation of the incident beam scanned to the jth position, O N (r) is an object function of the nth layer, specifically expressed as:
wherein,for fresnel near field diffraction, z is the propagation distance of near field diffraction.
Optimizing parameters to be optimized according to the gradient comprises:
wherein O is i (r) is an object function of the i-th layer,as object function O i (r)，α P For the incident beam function P->For the learning rate of the vacuum of the upper surface of the sample (i.e. the distance of the pre-propagation of the incident beam for each scanning position)>Gradient as object function->Gradient of the input beam function->Is the gradient of the upper surface of the sample.
Specifically, in embodiments of the present application, as shown in FIG. 3, a surface tilted bilayer MoS is used in a stacked imaging algorithm 2 The structure, wherein the left image is a structural projection along the incident beam, the right image is a structural model, and a significant surface tilt is seen, which causes the object and incident beam to intermix, and the resulting object and incident beam present artifacts, requiring reconstruction of the two-dimensional material MoS with shear 2 Simulated diffraction data obtained by simulation calculation, wherein the simulation calculation conditions are as follows: accelerated electron energy: 300keV; convergence half angle: 25mrad; scanning step length:amount of under-focus: 30nm, as shown in fig. 2, the negative effect of the sample shape on the stack imaging is large, high quality reconstruction results cannot be obtained in the related art, the stack imaging thought of the present application is used to optimize the object function, the input beam function and the object shape parameters simultaneously, and simultaneously obtain parameters to be optimized, so that the reconstruction obtains a wide range of object functions as shown in fig. 4, projections of the corresponding structures in the input beam direction, as shown in fig. 5, the input beam functions as shown in fig. 5, wherein no artifacts related to the object function and the sample shape functions as shown in fig. 6 exist, wherein the structures arranged in the structure of fig. 3 can be exactly matched, as shown in fig. 7 and 8, the reconstruction results of the embodiment of the present application are obtained, and MoS is superimposed 2 Along the projection of the incident beam direction, each atom column of the six-membered ring contains one Mo atom and two S atoms, and compared with the related technology, the object function obtained by the method has better contrast and higher signal-to-noise ratio.
According to the laminated imaging method comprising sample shape optimization, under the condition that the shape parameters of the sample in the three-dimensional space are restored by utilizing the diffraction data set, the shape parameters of the sample can be introduced into an algorithm, and repeated iterative optimization is performed in the reconstruction process, so that deviation between the normal direction of the sample and the direction of the band axis of the crystal is avoided, high-precision sample information is obtained, the preparation difficulty of the sample is reduced, the stability of a laminated imaging algorithm is improved, and meanwhile, the shape parameters of the sample in the three-dimensional space can be directly obtained. Therefore, the problems that in the related art, due to the fact that conditions in experiments are complex, the deviation of the normal direction of the sample surface and the direction of the crystal band axis possibly causes the change of optical path difference, the decoupling effect and the reconstruction accuracy are further reduced, in addition, due to the fact that wrinkles exist in the two-dimensional material, errors are increased, space distribution information is difficult to obtain and the like are solved.
Next, a stacked imaging apparatus including sample shape optimization according to an embodiment of the present application will be described with reference to the accompanying drawings.
Fig. 9 is a block schematic diagram of a stacked imaging apparatus incorporating sample shape optimization in accordance with an embodiment of the present application.
As shown in fig. 9, the stacked imaging apparatus 10 including sample shape optimization includes: a generation module 100, a calculation module 200, a solution module 300, and an iteration module 400.
Specifically, the generating module 100 is configured to scan a sample to collect diffraction information corresponding to a scan point, and generate a diffraction data set according to the diffraction information.
The calculation module 200 is configured to calculate a loss function of the forward propagation model by including the sample shape parameter as a variable parameter in the forward propagation process.
The solving module 300 is configured to solve a gradient of the loss function with respect to the parameter to be optimized, and optimize the parameter to be optimized according to the gradient.
The iteration module 400 is configured to recalculate the forward propagation based on the parameter to be optimized, calculate the loss function until a preset iteration termination condition is satisfied, and output a final parameter to be optimized, where the sample shape parameter according to the final parameter to be optimized is used as a parameter for reflecting the shape of the sample in the three-dimensional space.
Optionally, in an embodiment of the present application, the parameters to be optimized include at least one of an objective function, a light source function, and a sample shape parameter.
Optionally, in one embodiment of the present application, further includes: and initializing a module.
Wherein the initialization module is used for initializing the object function, the light source function and the sample shape parameter before taking the sample shape parameter as the variable parameter.
It should be noted that the foregoing explanation of the embodiment of the stacked imaging method including the optimization of the shape of the sample is also applicable to the stacked imaging apparatus including the optimization of the shape of the sample in this embodiment, and will not be repeated here.
According to the laminated imaging device comprising sample shape optimization, under the condition that the shape parameters of the sample in the three-dimensional space are restored by utilizing the diffraction data set, the shape parameters of the sample can be introduced into an algorithm, and repeated iterative optimization is performed in the reconstruction process, so that deviation between the normal direction of the sample and the direction of the band axis of the crystal is avoided, high-precision sample information is obtained, the preparation difficulty of the sample is reduced, the stability of a laminated imaging algorithm is improved, and meanwhile, the shape parameters of the sample in the three-dimensional space can be directly obtained. Therefore, the problems that in the related art, due to the fact that conditions in experiments are complex, the deviation of the normal direction of the sample surface and the direction of the crystal band axis possibly causes the change of optical path difference, the decoupling effect and the reconstruction accuracy are further reduced, in addition, due to the fact that wrinkles exist in the two-dimensional material, errors are increased, space distribution information is difficult to obtain and the like are solved.
Fig. 10 is a schematic structural diagram of an electronic device according to an embodiment of the present application. The electronic device may include:
memory 1001, processor 1002, and a computer program stored on memory 1001 and executable on processor 1002.
The processor 1002, when executing the program, implements the stack imaging method including sample shape optimization provided in the above-described embodiments.
Further, the electronic device further includes:
a communication interface 1003 for communication between the memory 1001 and the processor 1002.
Memory 1001 for storing computer programs that may be run on processor 1002.
Memory 1001 may include high-speed RAM memory and may also include non-volatile memory (non-volatile memory), such as at least one disk memory.
If the memory 1001, the processor 1002, and the communication interface 1003 are implemented independently, the communication interface 1003, the memory 1001, and the processor 1002 may be connected to each other through a bus and perform communication with each other. The bus may be an industry standard architecture (Industry Standard Architecture, abbreviated ISA) bus, an external device interconnect (Peripheral Component, abbreviated PCI) bus, or an extended industry standard architecture (Extended Industry Standard Architecture, abbreviated EISA) bus, among others. The buses may be divided into address buses, data buses, control buses, etc. For ease of illustration, only one thick line is shown in fig. 10, but not only one bus or one type of bus.
Alternatively, in a specific implementation, if the memory 1001, the processor 1002, and the communication interface 1003 are integrated on a single chip, the memory 1001, the processor 1002, and the communication interface 1003 may perform communication with each other through internal interfaces.
The processor 1002 may be a central processing unit (Central Processing Unit, abbreviated as CPU), or an application specific integrated circuit (Application Specific Integrated Circuit, abbreviated as ASIC), or one or more integrated circuits configured to implement embodiments of the present application.
The embodiments also provide a computer readable storage medium having stored thereon a computer program which when executed by a processor implements a stack imaging method as above comprising sample shape optimization.
In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present application. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or N embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.
Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In the description of the present application, the meaning of "N" is at least two, such as two, three, etc., unless explicitly defined otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and additional implementations are included within the scope of the preferred embodiment of the present application in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order from that shown or discussed, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the embodiments of the present application.
Logic and/or steps represented in the flowcharts or otherwise described herein, e.g., a ordered listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or N wires, a portable computer cartridge (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer readable medium may even be paper or other suitable medium on which the program is printed, as the program may be electronically captured, via optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It is to be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the N steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, may be implemented using any one or combination of the following techniques, as known in the art: discrete logic circuits having logic gates for implementing logic functions on data signals, application specific integrated circuits having suitable combinational logic gates, programmable Gate Arrays (PGAs), field Programmable Gate Arrays (FPGAs), and the like.
Those of ordinary skill in the art will appreciate that all or a portion of the steps carried out in the method of the above-described embodiments may be implemented by a program to instruct related hardware, where the program may be stored in a computer readable storage medium, and where the program, when executed, includes one or a combination of the steps of the method embodiments.
In addition, each functional unit in each embodiment of the present application may be integrated in one processing module, or each unit may exist alone physically, or two or more units may be integrated in one module. The integrated modules may be implemented in hardware or in software functional modules. The integrated modules may also be stored in a computer readable storage medium if implemented in the form of software functional modules and sold or used as a stand-alone product.
The above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, or the like. Although embodiments of the present application have been shown and described above, it will be understood that the above embodiments are illustrative and not to be construed as limiting the application, and that variations, modifications, alternatives, and variations may be made to the above embodiments by one of ordinary skill in the art within the scope of the application.
Claims (10)
1. A method of stack imaging comprising optimization of sample shape, comprising the steps of:
scanning a sample to acquire diffraction information of a corresponding scanning point, and generating a diffraction data set according to the diffraction information;
when a forward propagation process is constructed, taking the shape parameter of the sample as a variable parameter, including the variable parameter into the forward propagation process, and calculating a loss function of a forward propagation model;
solving the gradient of the loss function with respect to the parameter to be optimized, and optimizing the parameter to be optimized according to the gradient; and
and recalculating forward propagation based on the parameters to be optimized, calculating the loss function until a preset iteration termination condition is met, and outputting final parameters to be optimized, wherein the sample shape parameters according to the final parameters to be optimized are taken as parameters for reflecting the shape of the sample in a three-dimensional space.
2. The method of claim 1, wherein the parameters to be optimized comprise at least one of an objective function, a light source function, and the sample shape parameter.
3. The method of claim 2, further comprising, prior to taking the sample shape parameter as the variable parameter:
initializing the object function, the light source function, and the sample shape parameter.
4. The method of claim 1, wherein the outgoing wave function of the forward propagation model is:
wherein Δz represents the thickness of each layer function, j represents the number of scan positions, r represents the coordinates in real space, P (r-r) j ) For an incident beam scanned to the j-th position,distance pre-propagation for an incident beam scanned to the jth position，O N (r) is an object function of the nth layer.
5. The method of claim 1, wherein the predetermined iteration termination condition is the loss function converging or reaching a predetermined number of iterations.
6. A stacked imaging apparatus comprising sample shape optimization, comprising:
the generation module is used for scanning the sample to acquire diffraction information of a corresponding scanning point and generating a diffraction data set according to the diffraction information;
the calculation module is used for taking the shape parameter of the sample as a variable parameter when constructing a forward propagation process, and calculating a loss function of a forward propagation model in the forward propagation process;
the solving module is used for solving the gradient of the loss function on the parameter to be optimized and optimizing the parameter to be optimized according to the gradient; and
and the iteration module is used for recalculating forward propagation based on the parameters to be optimized, calculating the loss function until a preset iteration termination condition is met, and outputting final parameters to be optimized, wherein the sample shape parameters according to the final parameters to be optimized are taken as parameters for reflecting the shape of the sample in a three-dimensional space.
7. The apparatus of claim 6, wherein the parameters to be optimized comprise at least one of an objective function, a light source function, and the sample shape parameter.
8. The apparatus as recited in claim 7, further comprising:
an initialization module for initializing the object function, the light source function and the sample shape parameter before taking the sample shape parameter as the variable parameter.
9. An electronic device, comprising: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor executing the program to implement the stack imaging method comprising sample shape optimization of any one of claims 1-5.
10. A computer-readable storage medium, on which a computer program is stored, characterized in that the program is executed by a processor for implementing a stack imaging method comprising sample shape optimization according to any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311414552.2A CN117451626A (en) | 2023-10-27 | 2023-10-27 | Stacked imaging method and apparatus including sample shape optimization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311414552.2A CN117451626A (en) | 2023-10-27 | 2023-10-27 | Stacked imaging method and apparatus including sample shape optimization |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117451626A true CN117451626A (en) | 2024-01-26 |
Family
ID=89588643
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311414552.2A Pending CN117451626A (en) | 2023-10-27 | 2023-10-27 | Stacked imaging method and apparatus including sample shape optimization |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117451626A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112784508A (en) * | 2021-02-12 | 2021-05-11 | 西北工业大学 | Deep learning-based airfoil flow field rapid prediction method |
| US20210207956A1 (en) * | 2020-01-07 | 2021-07-08 | Kla Corporation | Methods And Systems For Overlay Measurement Based On Soft X-Ray Scatterometry |
| CN113720865A (en) * | 2021-08-06 | 2021-11-30 | 清华大学 | Electronic lamination imaging method and device for automatically correcting tape axis deviation of sample |
| CN114241072A (en) * | 2021-12-17 | 2022-03-25 | 中国科学院大学 | Laminated imaging reconstruction method and system |
| US20230280290A1 (en) * | 2022-03-02 | 2023-09-07 | Rigaku Corporation | Device and method for analyzing diffraction pattern of mixture, and information storage medium |
-
2023
- 2023-10-27 CN CN202311414552.2A patent/CN117451626A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210207956A1 (en) * | 2020-01-07 | 2021-07-08 | Kla Corporation | Methods And Systems For Overlay Measurement Based On Soft X-Ray Scatterometry |
| CN112784508A (en) * | 2021-02-12 | 2021-05-11 | 西北工业大学 | Deep learning-based airfoil flow field rapid prediction method |
| CN113720865A (en) * | 2021-08-06 | 2021-11-30 | 清华大学 | Electronic lamination imaging method and device for automatically correcting tape axis deviation of sample |
| CN114241072A (en) * | 2021-12-17 | 2022-03-25 | 中国科学院大学 | Laminated imaging reconstruction method and system |
| US20230280290A1 (en) * | 2022-03-02 | 2023-09-07 | Rigaku Corporation | Device and method for analyzing diffraction pattern of mixture, and information storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210090244A1 (en) | Method and system for optimizing optical inspection of patterned structures | |
| TWI322330B (en) | Pattern generation method and charged particle beam writing apparatus | |
| JP3636438B2 (en) | Method and apparatus for high speed aerial image simulation | |
| JP5057554B2 (en) | System and method for minimizing print line shape distortion by illumination and reticle optimization | |
| TWI305299B (en) | Method, computer program product, and appararus for generating models for simulating the imaging performance of a plurality of exposure tools | |
| JP5575776B2 (en) | Improvements in the field of imaging | |
| JP2006200999A (en) | Image processor and refractive index distribution measuring instrument | |
| US7246343B2 (en) | Method for correcting position-dependent distortions in patterning of integrated circuits | |
| EP2618738A2 (en) | Three dimensional imaging | |
| Finckh et al. | Geometry construction from caustic images | |
| WO2023142174A1 (en) | Method and apparatus for reconstructing electron orbit spatial distribution and electron beam function | |
| CN113888444A (en) | Image reconstruction method and system based on laminated self-focusing experiment | |
| US7379582B2 (en) | Three-dimensional structure verification supporting apparatus, three-dimensional structure verification method, recording medium, and program therefor | |
| TWI603070B (en) | Method and system for use in measuring in complex patterned structures | |
| JP4671473B2 (en) | Mask data correction apparatus, method for manufacturing transfer mask, and method for manufacturing apparatus having pattern structure | |
| TW202234144A (en) | Multi-component kernels for vector optical image simulation | |
| CN117451626A (en) | Stacked imaging method and apparatus including sample shape optimization | |
| TW202235997A (en) | Litho-aware source sampling and resampling | |
| CN108428245A (en) | Sliding method for registering images based on self-adapting regular item | |
| Villarrubia et al. | Virtual rough samples to test 3D nanometer-scale scanning electron microscopy stereo photogrammetry | |
| Lobato et al. | Deconvolution with neural grid compression: A method to accurately and quickly process beamforming results | |
| Xie et al. | Deep learning for estimation of Kirkpatrick–Baez mirror alignment errors | |
| JP6450561B2 (en) | Imaging pattern simulation system, imaging pattern simulation method, imaging pattern simulation program, and recording medium recording this program | |
| US11143499B2 (en) | Three-dimensional information generating device and method capable of self-calibration | |
| CN110007581B (en) | Method and device for generating orthogonal scanning multi-view projection calculation hologram |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |