Optimization method for autofocus or aberration correction and charged particle beam device

Abstract

The present invention provides an optimization method for automatic focusing or aberration correction and a charged particle beam device. The method includes: establishing an evaluation function regarding device parameters to evaluate the image quality during automatic focusing or aberration correction; establishing a determination condition for evaluating the anti-interference ability of a first algorithm, using the first algorithm to optimize the evaluation function, and switching to using a second algorithm to optimize the evaluation function when the determination condition is met, wherein the anti-interference ability of the second algorithm is higher than that of the first algorithm; obtaining the device parameters when the evaluation function forms an extreme value such that the image quality is optimal. In the present invention, when the first algorithm meets the determination condition, it switches to using the second algorithm to optimize the evaluation function, which can improve the anti-interference ability of the solution process and the accuracy of the calculation result.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and particularly to an optimization method for automatic focusing or aberration correction and a charged particle beam device. Background Art

[0002] Charged particle beam devices include scanning electron microscopes (SEM), focused ion beam devices, transmission electron microscopes, etc. Charged particle beam devices achieve scanning imaging of samples by controlling the focusing, deflection, etc. of charged particle beams. Due to their high resolution and other characteristics, they have wide applications in fields such as semiconductor front-end detection and measurement.

[0003] In order to obtain high-quality images, it is usually necessary to separately obtain the optimal device parameters of the charged particle beam device for automatic focusing and aberration correction. Taking the aberration correction specifically as the astigmatism correction as an example, generally, the astigmatism correction process of the charged particle beam is roughly described as follows: By adjusting the device parameters regarding astigmatism (the current or voltage value of the astigmatism corrector), the astigmatism of the charged particle beam in the horizontal direction (x-direction) or the vertical direction (y-direction) is adjusted to make the elliptical beam spot of the charged particle beam become a circular beam spot, searching for the device parameters regarding astigmatism to make the image clear, and making the evaluation function regarding the image quality reach the maximum value to achieve the purpose of correcting aberration. The same applies to the process of automatic focusing. Search for the device parameters regarding automatic focusing to obtain the device parameters when the evaluation function reaches the maximum value to make the image clear, which will not be elaborated here. Currently, the calculation method of the maximum value of the evaluation function has the following disadvantages: It is greatly affected by noise interference, has poor anti-interference ability, and affects the calculation result.

[0004] Therefore, the present invention proposes an optimization method for automatic focusing or aberration correction and a charged particle beam device to improve the accuracy of the calculation result of the extreme value of the image evaluation function. Summary of the Invention

[0005] Embodiments of the present invention provide an optimization method for automatic focusing or aberration correction and a charged particle beam device to improve the efficiency of the calculation process of the extreme value of the image evaluation function and the accuracy of the calculation result.

[0006] In a first aspect, the present invention provides an optimization method for automatic focusing or aberration correction of a charged particle beam device, including: establishing an evaluation function regarding device parameters to evaluate the image quality during automatic focusing or aberration correction; establishing a determination condition for evaluating the anti-interference ability of a first algorithm, using the first algorithm to optimize the evaluation function, and switching to using a second algorithm to optimize the evaluation function when the determination condition is met, where the anti-interference ability of the second algorithm is higher than that of the first algorithm; and obtaining the device parameters when the evaluation function forms an extreme value such that the image quality is optimal.

[0007] The beneficial effect is that: during the solution process of the extreme value of the image quality evaluation function, switching to using the second algorithm to optimize the evaluation function when the first algorithm meets the determination condition can improve the anti-interference ability of the solution process, and further improve the accuracy of the solution result of the extreme value of the image quality evaluation function.

[0008] Optionally, establishing the determination condition for evaluating the anti-interference ability of the first algorithm includes: the first algorithm is the inverse quadratic interpolation method, obtaining three input parameters a, b, and c of the inverse quadratic interpolation method, where a < b < c; establishing at least one distance threshold formula regarding two of the input parameters and / or a slope difference threshold formula regarding the three input parameters to form the determination condition. The beneficial effect is that: the inverse quadratic interpolation algorithm has a relatively fast convergence speed, which can ensure a high calculation efficiency and shorten the calculation time. However, when the values of the input parameters are too close during the solution process, it will affect the accuracy of the calculation result. Therefore, it is necessary to set a determination condition to limit the input parameters, and when the input parameters meet the determination condition, switch to another algorithm with higher anti-interference ability.

[0009] Further optionally, the distance threshold formula is one or more of the following formulas: |b - a| ≤ ε1, |c - b| ≤ ε2, where ε1 and ε2 are preset first and second distance thresholds. The beneficial effect is that: in this aspect, the determination condition is set by judging the difference between the input parameters, which is simple and easy to implement and operate.

[0010] Further optionally, the slope threshold formula is: where f is the evaluation function and δ is a preset slope threshold. The beneficial effect is that: when the slope of the line connecting the points corresponding to the input parameters on the evaluation function is relatively close, it can also indicate that the corresponding input parameters are relatively close.

[0011] Optionally, the second algorithm is the golden section method. The iteration termination conditions for both the first algorithm and the second algorithm include: |x - b| ≤ ε3 or both include reaching the maximum number of iterations, where x is the current device parameter and ε3 is a preset third distance threshold. The beneficial effect is that when |x - b| ≤ ε3, it can be shown that the current device parameter x is very close to the current input parameter b. Therefore, at this time, the value of the evaluation function corresponding to x or b is the extreme value of the evaluation function.

[0012] Further optionally, obtaining the device parameter when the evaluation function forms an extreme value to optimize the image quality includes: obtaining the device parameter x corresponding to when the evaluation function forms an extreme value t and the corresponding input parameter b t ; taking the mean value (x t + b t ) / 2 of the device parameter x t and the input parameter b t as the device parameter for optimizing the image quality. The beneficial effect is that when the evaluation function forms an extreme value, the device parameter x t and the input parameter b t are very close. Therefore, the mean value of the two can be taken as the device parameter for optimizing the image quality.

[0013] Optionally, the second algorithm is the golden section method or the bisection method. The beneficial effect is that the golden section method or the bisection method has high anti-interference ability, which can improve the accuracy of the extreme value solution of the image quality evaluation function.

[0014] Optionally, the device parameter is at least one of the current parameter for focusing a charged particle beam, the voltage parameter for focusing a charged particle beam, the horizontal aberration correction parameter, and the vertical aberration correction parameter. The beneficial effect is that different parameters can be characterized by the device parameter, and the influence of different parameters on the extreme value solution process of the image quality evaluation function can be studied.

[0015] Optionally, the method further includes: obtaining an image of the charged particle beam device using the obtained device parameter. The beneficial effect is that the image of the charged particle beam device obtained by the device parameter has higher quality. For example, the clarity of the image is the highest.

[0016] In a second aspect, the present invention provides a charged particle beam device, which includes a module / unit that executes the method according to any possible design in the first aspect above. These modules / units can be implemented by hardware or by hardware executing corresponding software.

[0017] For the beneficial effects of the above second aspect, reference may be made to the description in the above first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 FIG. is a flowchart of an optimization method for automatic focusing or aberration correction of a charged particle beam device provided by an embodiment of the present application;

[0019] Figure 2 FIG. is a flowchart of an optimization method for an image quality evaluation function provided by an embodiment of the present application;

[0020] Figure 3 FIG. is a schematic diagram of an inverse quadratic interpolation method provided by an embodiment of the present application;

[0021] Figure 4 FIG. is a schematic diagram of a golden section method provided by an embodiment of the present application;

[0022] Figure 5 FIG. is a schematic diagram of a charged particle beam device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] Next, the technical solutions in the embodiments of the present application will be described with reference to the accompanying drawings in the embodiments of the present application. Among them, in the description of the embodiments of the present application, the terms used in the following embodiments are only for the purpose of describing specific embodiments, and are not intended to limit the present application. As used in the specification and appended claims of the present application, the singular forms "a", "the", "above", "the" and "this" are also intended to include, for example, the expression "one or more", unless the context clearly indicates otherwise. It should also be understood that in the following embodiments of the present application, "at least one" and "one or more" mean one or more than two (including two). The term "and / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist; for example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone, where A and B may be singular or plural. The character " / " generally represents an "or" relationship between the associated objects before and after.

[0024] References to "one embodiment" or "some embodiments" or the like described in this specification mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of the present application. Thus, statements such as "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc., which appear in different places in this specification, do not necessarily all refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized. The terms "comprising", "including", "having" and their variants mean "including but not limited to", unless otherwise specifically emphasized. The term "connection" includes direct connection and indirect connection, unless otherwise stated. "First" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features.

[0025] In the embodiments of the present application, words such as "exemplarily" or "for example" are used to represent examples, illustrations or explanations. Any embodiment or design solution described as "exemplarily" or "for example" in the embodiments of the present application should not be construed as being more preferred or more advantageous than other embodiments or design solutions. Rather, the use of words such as "exemplarily" or "for example" is intended to present the relevant concepts in a specific manner.

[0026] An embodiment of the present invention provides an optimization method for automatic focusing or aberration correction of a charged particle beam device, and its process is as Figure 1 shown, and the specific steps are as follows:

[0027] S101, establish an evaluation function for device parameters to evaluate the image quality during automatic focusing or aberration correction.

[0028] In this step, the device parameters are related to automatic focusing or aberration correction and can evaluate the quality of the image. When it is for automatic focusing or some types of aberrations, it is used to characterize the quality of the image, specifically sharpness. When it is for aberration of the distortion type, it is used to characterize the quality of the image but is not related to the sharpness of the image.

[0029] S102, establish a determination condition for evaluating the anti-interference ability of the first algorithm, use the first algorithm to optimize the evaluation function, and switch to using the second algorithm to optimize the evaluation function when the determination condition is met, where the anti-interference ability of the second algorithm is higher than that of the first algorithm.

[0030] In this step, when the determination condition is met (for example, under this determination condition, the optimization effect of the evaluation function obtained by optimizing the evaluation function through the first algorithm is not good), the evaluation function can be further optimized by switching to the second algorithm. This step optimizes the evaluation function through at least two algorithms, which can comprehensively consider the characteristics of the evaluation function and combine the advantages of different algorithms to optimize the evaluation function.

[0031] S103. Obtain the device parameters when the evaluation function forms an extreme value to optimize the image quality.

[0032] In this step, it is not limited to solving the maximum value of the evaluation function to optimize the image quality, and the minimum value of the evaluation function can also be solved. For example, the problem of solving the maximum value of the image quality evaluation function can be transformed into the problem of solving the minimum value of the negative evaluation function. The following examples are all given by solving the maximum value of the evaluation function.

[0033] In this embodiment, during the process of solving the maximum value of the evaluation function, when the first algorithm meets the determination condition, it is switched to the second algorithm to optimize the evaluation function, which can improve the anti-interference ability of the solution process, and then improve the accuracy of the solution result of the extreme value of the image quality evaluation function. Moreover, when the first algorithm is the inverse quadratic interpolation method, a relatively accurate calculation method is also used to shorten the calculation time and improve the calculation efficiency.

[0034] Exemplarily, when the first algorithm is the inverse quadratic interpolation method and the second algorithm is the golden section method, the process of solving the extreme value of the evaluation function in the above embodiment is as Figure 2 shown, including the following steps:

[0035] S201. Set the search range [a, c] of the device parameters, that is, the value of the device parameters ranges from a to c, and a < c, and establish an evaluation function.

[0036] S202. If the conditions of the iteration termination module are met, the iteration terminates, and S206 is executed; otherwise, S203 is executed. The conditions for iteration termination include: |x - b| ≤ ε3 or both include reaching the maximum number of iterations.

[0037] S203. Calculate the value of the device parameters according to the inverse quadratic interpolation method.

[0038] S204. When the determination condition is met, S205 is executed; otherwise, S202 is executed.

[0039] S205. Calculate the value of the device parameters according to the golden section method.

[0040] S206. Output the device parameters corresponding to the extreme value of the evaluation function as the optimal device parameters.

[0041] In a possible embodiment, the establishment of the determination condition for evaluating the anti-interference ability of the first algorithm includes: the first algorithm is the inverse quadratic interpolation method, and three input parameters a, b, and c of the inverse quadratic interpolation method are obtained. Among them, all three are variables, and the values of the output parameters a, b, and c can change during the iteration process, but all satisfy a < b < c. After initial values are assigned to a and c, the initial value of b can be calculated according to the formula described later; at least one distance threshold formula for two of the input parameters and / or a slope difference threshold formula for the three input parameters are established to form the determination condition.

[0042] Exemplarily, the specific content of step S203 is as Figure 3 shown, and specifically includes:

[0043] S301, calculate the initial value of the input parameter b and the initial value of the device parameter x according to the inverse quadratic interpolation method, specifically as follows:

[0044] where the value range of the device parameter is from a to c, so that the initial value of the input parameter b can be directly calculated through the value range of the device parameter and the above algorithm.

[0045] S302, judge the magnitudes of x and b. When x < b, execute S303, and when x > b, execute S304.

[0046] S303, assign b to c to form a new change range [a, c] of the device parameter, then assign x to b to update the value of b, and continue to calculate the value of the new x according to the formula

[0047] S304, assign b to a to form a new change range [a, c] of the device parameter, then assign x to b to update the value of b, and continue to calculate the value of the new x according to the formula

[0048] In this embodiment, the convergence rate of the inverse quadratic interpolation method is relatively fast, but there are certain defects. Therefore, it is necessary to set determination conditions to limit the input parameters, and when the input parameters meet the determination conditions, switch to other algorithms.

[0049] In another possible embodiment, the distance threshold formula is one or more of the following formulas: |b - a| ≤ ε1, |c - b| ≤ ε2, where ε1 and ε2 are the preset first distance threshold and second distance threshold. In this embodiment, the determination condition is set by judging the difference between the input parameters, which is simple and easy to operate. ​​

[0050] In still another possible embodiment, the slope threshold formula is as follows: where f is the evaluation function and δ is a preset slope threshold. In this embodiment, when the slopes of the lines connecting the points corresponding to the input parameters on the evaluation function are relatively close, it can also indicate that the corresponding input parameters are relatively close.

[0051] In a possible embodiment, the second algorithm is the golden section method, and the iteration termination conditions of both the first algorithm and the second algorithm include: |x - b| ≤ ε3 or both include reaching the maximum number of iterations, where x is the current device parameter and ε3 is a preset third distance threshold.

[0052] Although the input parameters are continuously updated during the solution process of the extreme value of the evaluation function, the value of the corresponding device parameter can be obtained according to the value of the updated input parameter and the value of its corresponding evaluation function. The iteration process terminates when the device parameter x and the input parameter satisfy certain conditions or reach the maximum number of iterations. In this embodiment, when |x - b| ≤ ε3, it can be shown that the current device parameter x is very close to the input parameter b, so at this time, the value of the evaluation function corresponding to x or b is the extreme value of the evaluation function.

[0053] In yet another possible embodiment, obtaining the device parameter when the evaluation function forms an extreme value to optimize the image quality includes: obtaining the device parameter x corresponding to when the evaluation function forms an extreme value t and the corresponding input parameter b t ; taking the mean value (x t +b t ) / 2 of the device parameter x t and the input parameter b t as the device parameter when the image quality is optimal. In this embodiment, because when the evaluation function forms an extreme value, the device parameter x t and the input parameter b t are very close, the mean value of the two can be taken as the device parameter when the image quality is optimal.

[0054] Exemplarily, the relationship between the maximum value of the evaluation function and the device parameter is as follows:

[0055] p * = argmax f(I(x)), where x represents the current device parameter, I represents the gray function of the image, f represents the evaluation function, and p * represents the value of the device parameter when the evaluation function takes the maximum value, that is, x as described in the embodiment t .

[0056] Exemplarily, taking astigmatism correction as an example, the evaluation function is the variance function of astigmatism normalization. Specifically, the evaluation function is as follows:

[0057] where M represents the number of rows of pixel points in the image, N represents the number of columns of pixel points in the image, μ represents the mean value of the image gray level, and the variance function can be used to represent the degree of astigmatism.

[0058] In still another possible embodiment, the second algorithm is the golden section method or the bisection method. The golden section method or the bisection method has high anti-interference ability, which can improve the accuracy of the extreme value solution result of the image quality evaluation function.

[0059] Among them, the inverse quadratic interpolation method has high calculation efficiency but poor anti-interference ability, and the golden section method or the bisection method has low calculation efficiency but good anti-interference ability. In this embodiment, the optimization is based on the inverse quadratic interpolation method, which can make the optimization method have high calculation efficiency as a whole, and switch to the golden section method or the bisection method with higher anti-interference ability than the inverse quadratic interpolation method when meeting the judgment conditions, so it has good anti-interference ability. To sum up, the optimization method provided in this embodiment comprehensively utilizes the advantages of the inverse quadratic interpolation method and the golden section method or the bisection method, and can improve the anti-interference ability on the basis of maintaining high calculation efficiency.

[0060] Exemplarily, when the second algorithm is the golden section method, the specific content of S205 in the above embodiment includes:

[0061] S401, calculating the initial value of the input parameter b and the initial value of the device parameter x according to the golden section method, specifically as follows:

[0062] S402, judging the magnitudes of f(b) and f(x), and when f(x) < f(b), execute S403; otherwise execute S404.

[0063] S403, assign b to a, and keep c unchanged to form a new change range [a, c] of the device parameter, and recalculate b and x according to the new change range and the formula mentioned in S401.

[0064] S404, assign x to c, and keep a unchanged to form a new change range [a, c] of the device parameter, and recalculate b and x according to the new change range and the formula mentioned in S401.

[0065] In yet another possible embodiment, the device parameters are at least one of the current parameter for focusing a charged particle beam, the voltage parameter for focusing a charged particle beam, the horizontal aberration correction parameter, and the vertical aberration correction parameter. In this embodiment, when the device parameter is characterized as the current parameter for focusing a charged particle beam or the device parameter is characterized as the voltage parameter for focusing a charged particle beam, the corresponding evaluation function created can be expressed as an index function of the charged particle beam focusing (which can also be said to be the image sharpness); taking astigmatism correction as an example, when the device parameter is characterized as the horizontal astigmatism correction parameter, the evaluation function can be expressed as the degree of astigmatism of the charged particle in the horizontal direction; when the device parameter is characterized as the vertical astigmatism correction parameter, the evaluation function can be expressed as the degree of astigmatism of the charged particle in the vertical direction. In this embodiment, different parameters are characterized by the device parameters, and the influence of different parameters on the extreme value solution process of the evaluation function can be studied. Correspondingly, the evaluation function can be expressed as an index function of the image quality sharpness or the degree of astigmatism in different directions.

[0066] In a possible embodiment, the method further includes: obtaining an image of the charged particle beam device using the obtained device parameters. In this embodiment, the quality of the image of the charged particle beam device obtained through the device parameters is higher.

[0067] An embodiment of the present application provides a charged particle beam device, as Figure 5 shown, the device 500 includes: a creation module 501, a determination module 502, and an optimal parameter acquisition module 503.

[0068] The creation module 501 establishes an evaluation function regarding the device parameters to evaluate the image quality during autofocus or aberration correction.

[0069] Among them, the device parameters are related to autofocus or aberration correction and can evaluate the quality of the image. When it is autofocus or some types of aberration, it is used to characterize the quality of the image, specifically sharpness. When it is aberration of the distortion type, it is used to characterize the quality of the image but has nothing to do with the sharpness of the image.

[0070] The determination module 502 establishes a determination condition for evaluating the anti-interference ability of the first algorithm, optimizes the evaluation function using the first algorithm, and switches to using the second algorithm to optimize the evaluation function when the determination condition is met, where the anti-interference ability of the second algorithm is higher than that of the first algorithm.

[0071] Exemplarily, when the determination condition is met (for example, under this determination condition, the optimization effect of the evaluation function obtained by optimizing the evaluation function through the first algorithm is not good), the second algorithm can be switched to further optimize the evaluation function. This step optimizes the evaluation function through at least two algorithms, which can comprehensively consider the characteristics of the evaluation function and combine the advantages of different algorithms to optimize the evaluation function.

[0072] The optimal parameter acquisition module 503 obtains the device parameters when the evaluation function forms an extreme value to make the image quality optimal.

[0073] It is not limited to solving the maximum value of the evaluation function to make the image quality optimal, and the minimum value of the evaluation function can also be solved. For example, the problem of solving the maximum value of the evaluation function can be transformed into the problem of solving the minimum value of the negative evaluation function.

[0074] In this embodiment, during the process of solving the maximum value of the evaluation function, when the first algorithm meets the determination condition, it is switched to the second algorithm to optimize the evaluation function, which can improve the anti-interference ability of the solving process, and further improve the accuracy of the extreme value solving result of the image quality evaluation function. Moreover, when the first algorithm is the inverse quadratic interpolation method, a relatively accurate calculation method is also used to shorten the calculation time and improve the calculation efficiency.

[0075] In a possible embodiment, the establishment of the determination condition for evaluating the anti-interference ability of the first algorithm includes: the first algorithm is the inverse quadratic interpolation method, and three input parameters a, b, and c of the inverse quadratic interpolation method are obtained, where all three are variables and their values can change during the iteration process, but all satisfy a < b < c. After initial values are assigned to a and c, the initial value of b can be calculated according to the formula described later; at least one distance threshold formula for two of the input parameters and / or a slope difference threshold formula for the three input parameters are established to form the determination condition.

[0076] Exemplarily, the specific content of updating the search range of the device parameters by the inverse quadratic interpolation method is as follows:

[0077] Calculate the initial value of the input parameter b and the initial value of the device parameter x according to the inverse quadratic interpolation method, specifically as follows:

[0078] Where the value range of the device parameter is from a to c, so the initial value of the input parameter b can be directly calculated through the value range of the device parameter and the above algorithm. Determine the size of x and b. When x < b, assign b to c to form a new change range [a, c] of the device parameter, then assign x to b to update the value of b, and according to the formula Continue to calculate the new value of x. When x > b, assign b to a to form a new range of variation [a, c] of the device parameters, then assign x to b to update the value of b, and according to the formula Continue to calculate the new value of x.

[0079] In this embodiment, the convergence rate of the inverse quadratic interpolation method is relatively fast, but there are certain defects. Therefore, it is necessary to set a judgment condition to limit the input parameter, and when the input parameter meets the judgment condition, switch to other algorithms.

[0080] In another possible embodiment, the distance threshold formula is one or more of the following formulas: |b - a| ≤ ε1, |c - b| ≤ ε2, where ε1 and ε2 are preset first and second distance thresholds. In this embodiment, by judging the difference of the input parameters to set the judgment condition, it is simple and easy to operate.

[0081] In still another possible embodiment, the slope threshold formula is: where f is the evaluation function and δ is a preset slope threshold. In this embodiment, when the slopes of the lines connecting the points corresponding to the input parameters on the evaluation function are relatively close, it can also indicate that the corresponding input parameters are relatively close.

[0082] In a possible embodiment, the second algorithm is the golden section method, and the iteration termination conditions of the first algorithm and the second algorithm both include: |x - b| ≤ ε3 or both include reaching the maximum number of iterations, where x is the current device parameter and ε3 is a preset third distance threshold.

[0083] Although the input parameter is continuously updated during the solution process of the extreme value of the evaluation function, the value of the corresponding device parameter can be obtained according to the value of the updated input parameter and the value of its corresponding evaluation function. The iteration process terminates when the device parameter x and the input parameter meet certain conditions or reach the maximum number of iterations. In this embodiment, when |x - b| ≤ ε3, it can be shown that the current device parameter x is very close to the input parameter b. Therefore, at this time, the value of the evaluation function corresponding to x or b is the extreme value of the evaluation function.

[0084] In another possible embodiment, obtaining the device parameter when the evaluation function forms an extreme value to optimize the image quality includes: obtaining the device parameter x corresponding to when the evaluation function forms an extreme value t and the corresponding input parameter b t ; taking the mean value of the device parameter x t and the input parameter b t (xt +b t ) / 2 as the device parameter when the image quality is optimal. In this embodiment, since when the evaluation function forms an extreme value, the device parameter x t and the input parameter b t are very close, the average value of the two can be taken as the device parameter when the image quality is optimal.

[0085] Exemplarily, the relationship between the maximum value of the evaluation function and the device parameter is as follows:

[0086] p * = argmax f(I(x)), where x represents the device parameter, I represents the grayscale function of the image, f represents the evaluation function, and p * represents the value of the device parameter when the evaluation function takes the maximum value, that is, x in the embodiment. t .

[0087] Exemplarily, taking astigmatism correction as an example, the evaluation function is the variance function of astigmatism normalization. The evaluation function is specifically:

[0088] Among them, and M represents the number of rows of pixel points in the image, N represents the number of columns of pixel points in the image, μ represents the average value of the image grayscale, and the variance function can be used to represent the degree of astigmatism.

[0089] In still another possible embodiment, the second algorithm is the golden section method or the bisection method. In this embodiment, the golden section method or the bisection method can quickly shorten the calculation range for solving the device parameter and improve the calculation speed.

[0090] Exemplarily, when the second algorithm is the golden section method, the content for updating the search range is specifically:

[0091] Calculate the initial value of the input parameter b and the initial value of the device parameter x according to the golden section method, specifically as follows: Judge the magnitudes of f(b) and f(x). When f(x) < f(b), assign b to a, and keep c unchanged to form a new change range [a, c] of the device parameter, and recalculate b and x according to the new change range and the above formula. When f(x) > f(b), assign x to c, and keep a unchanged to form a new change range [a, c] of the device parameter, and recalculate b and x according to the new change range and the above formula.

[0092] In yet another possible embodiment, the device parameter is at least one of a current parameter for focusing a charged particle beam, a voltage parameter for focusing a charged particle beam, an aberration correction parameter in the horizontal direction, and an aberration correction parameter in the vertical direction. In this embodiment, when the device parameter is characterized as the current parameter for focusing the charged particle beam or the device parameter is characterized as the voltage parameter for focusing the charged particle beam, the corresponding evaluation function created can be expressed as an index function of the charged particle beam focusing (which can also be said to be the image sharpness); taking astigmatism correction as an example, when the device parameter is characterized as the astigmatism correction parameter in the horizontal direction, the evaluation function can be expressed as the degree of astigmatism of the charged particle in the horizontal direction; when the device parameter is characterized as the astigmatism correction parameter in the vertical direction, the evaluation function can be expressed as the degree of astigmatism of the charged particle in the vertical direction. In this embodiment, by characterizing different parameters with the device parameter, the influence of different parameters on the extreme value solution process of the evaluation function can be studied. Correspondingly, the evaluation function can be expressed as an index function of the image quality sharpness or the degree of astigmatism in different directions.

[0093] In one possible embodiment, the method further includes: obtaining an image of the charged particle beam device using the obtained device parameter. In this embodiment, the image of the charged particle beam device obtained through the device parameter has a higher resolution.

[0094] As described above, the above is only the specific implementation manner of the embodiments of the present application, but the protection scope of the embodiments of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the embodiments of the present application should be covered within the protection scope of the embodiments of the present application. Therefore, the protection scope of the embodiments of the present application should be subject to the protection scope of the claims.

Claims

1. An optimization method for automatic focusing or aberration correction of a charged particle beam device, characterized in that, Including: Establish an evaluation function for device parameters to evaluate the image quality during autofocus or aberration correction; Establish a determination condition for evaluating the anti-interference ability of the first algorithm, use the first algorithm to optimize the evaluation function, and switch to using the second algorithm to optimize the evaluation function when the determination condition is met, where the anti-interference ability of the second algorithm is higher than that of the first algorithm, and the second algorithm is the golden section method or the bisection method; Obtain the device parameters when the evaluation function forms an extreme value to make the image quality optimal; Wherein, the establishment of the determination condition for evaluating the anti-interference ability of the first algorithm includes: The first algorithm is the inverse quadratic interpolation method, and three input parameters a, b, and c of the inverse quadratic interpolation method are obtained, where a < b < c; Establish at least one distance threshold formula for two of the input parameters and / or a slope difference threshold formula for the three input parameters to form the determination condition.

2. The method according to claim 1, wherein The distance threshold formula is one or more of the following formulas: |b - a| ≤ ε1, |c - b| ≤ ε2, where ε1 and ε2 are the preset first distance threshold and second distance threshold.

3. The method according to claim 1, wherein The slope difference threshold formula is: , where f is the evaluation function and δ is the preset slope threshold.

4. The method according to claim 1, wherein The second algorithm is the golden section method, and the iteration termination conditions of the first algorithm and the second algorithm both include: |x - b| ≤ ε3 or both include reaching the maximum number of iterations, where x is the current device parameter and ε3 is the preset third distance threshold.

5. The method according to claim 4, characterized in that The obtaining of the device parameters when the evaluation function forms an extreme value to make the image quality optimal includes: Obtain the device parameter xt corresponding to when the evaluation function forms an extreme value and the corresponding input parameter bt; Take the mean value (xt + bt) / 2 of the device parameter xt and the input parameter bt as the device parameter when the image quality is optimal.

6. The method according to claim 1, wherein The device parameter is at least one of the current parameter for charged particle beam focusing, the voltage parameter for charged particle beam focusing, the horizontal direction aberration correction parameter, and the vertical direction aberration correction parameter.

7. The method according to claim 1, wherein The method further includes: obtaining an image of the charged particle beam device using the obtained device parameters.

8. A charged particle beam device, characterized in that, Configured to execute the method according to any one of claims 1 to 7, the device includes: a creation module, a determination module, and an optimal parameter acquisition module, the creation module is used to establish an evaluation function for device parameters to evaluate the image quality during autofocus or aberration correction; The determination module establishes a determination condition for evaluating the anti-interference ability of the first algorithm, uses the first algorithm to optimize the evaluation function, and switches to using the second algorithm to optimize the evaluation function when the determination condition is met; The optimal parameter acquisition module is used to obtain the device parameters when the evaluation function forms an extreme value to make the image quality optimal.

Images