Patent Application
20250067688
The present invention relates to a particle beam analyzer and a particle beam analysis method, and more particularly to a particle beam analyzer and a particle beam analysis method for analyzing a measurement result obtained using a particle beam.
Measurement using a particle beam is widely used, mainly in the field of materials science, as a method for observing a microstructure of a substance. The particle beam herein refers to a proton beam (α ray), an electron beam (β ray), a muon beam, a photon beam (that is, an electromagnetic wave, γ-ray, X-ray, visible light, infrared ray, and the like), a neutron beam, or the like. A sample such as a metal is irradiated with such particle beams, and outputs of reflected, transmitted, and scattered particle beams (particles different from the injected particles may be outputted) are measured by a chemical or mechanical detection unit. Then, a nanometer-scale microstructure of the sample is estimated from the distribution shape of the intensity of the particle beams (the number of measured particles). Note that similar measurement can be performed as long as measurement is possible using the frequency of a sound wave or the like, or may be treated as a particle. Hereinafter, the reflected, scattered, and transmitted particle beam is referred to as an output particle beam.
A process until particles injected into a sample in such measurement interact with a minute structure (hereinafter, referred to as a microstructure) inside the sample and are scattered is formulated as a dynamic of a quantum mechanical wave function. Therefore, the distribution of the particle beam intensity can be calculated as a function that depends on the vector change of the wave count (½π of the frequency) of the wave function of the output particle beam and on the microstructure. Hence, attempts have been made to use this function to reconstruct information on a microstructure from the intensity distribution of an output particle beam. However, the above-described microstructure-dependent function cannot be formulated by an inverse function, and information on the microstructure cannot be easily calculated from the intensity distribution of the output particle beam. For this reason, a solution for solving this inverse problem by means of estimation is required.
For example, N-PTL 1 discloses the Monte Carlo method as a method for calculating information on a scatterer from the results of a scattering experiment. The Monte Carlo method is a method for calculating a scattering pattern while randomly changing a quantity related to a spatial structure and searching for a condition that a difference from a measurement result becomes small.
In addition, PTL 1 discloses a method for obtaining a scatterer size distribution from a two-dimensional scattering pattern by means of function fitting.
As another similar technique, a super-resolution technique for estimating a higher-resolution image from a plurality of photographic images obtained by photographing the same object is known. In addition, there is also known a tomography technique in which an object is photographed from various directions and a three-dimensional structure is restored from the images.
PTL 1: JP 2017-116330 A
N-PTL 1: ISO 17867:2015 “Particle size analysis—Small-angle X-ray scattering”
In the Monte Carlo method disclosed in N-PTL 1, because the parameters are randomly changed, it takes a lot of calculation time to reach a correct result. Furthermore, because the parameters are changed randomly, the result obtained is not always correct.
In addition, in the method of PTL 1, a simple distribution function (basis function) with which the distribution of scatterer size can be easily calculated is assumed. Further, the basis function is expressed by multiplying the basis function by a coefficient and then performing addition so that the difference from the measurement result becomes small. As the basis function, a simplified rectangular function that has a constant value of “1” in a specific section and a constant value of “0” in other sections is often used. The estimation calculation at that time is called an indirect Fourier transform. However, in this method, there are also cases where the distribution function of the scatterer size cannot be accurately expressed by adding together simplified basis functions. In that case, a problem arises where an appropriate result of the microstructure cannot be obtained. Therefore, a method for expressing a distribution function of the size of the scatterer using a larger number of basis functions is also conceivable. For example, in the case of the indirect Fourier transform, a method for reducing the width of the rectangular function is used. However, in this case, the number of coefficients to be determined also increases, and the coefficients cannot be determined. In this case, there is also a method in which the coefficient can be determined by providing some kind of constraint. However, this constraint needs to be determined as needed from a condition obtained as prior knowledge at the time of analysis, for example, smoothness assumed for the distribution of the sizes of the scatterer. As described above, in PTL 1, analysis work by an expert having sufficient prerequisite knowledge regarding the scatterer is essential, and it is difficult to automate this analysis.
In addition, another similar technique described above is a technique for reducing deterioration of a signal in a measurement process by using a plurality of pieces of information in combination. These techniques cannot be applied to a particle beam analyzer because the conditions are different from the conditions for measurement using a particle beam.
In view of the above problems, an object of the present invention is to provide a particle beam analyzer and a particle beam analysis method capable of estimating an objective and accurate density distribution of a spatial structure while reducing a processing load.
In order to achieve the foregoing object, one representative example of the particle beam analyzer of the present invention is a particle beam analyzer that, when an input profile is provided which is obtained by a particle beam being injected into a sample and a reflected beam or a transmitted beam thereof k calculates a spatial density distribution of the sample, the particle beam analyzer including: a profile database that holds a plurality of pieces of data of a spatial density distribution and a profile corresponding to the spatial density distribution; a profile difference evaluation unit that calculates a difference between a profile held in the profile database and the input profile; a profile selection unit that selects a plurality of profiles from the data held in the profile database, on the basis of the difference calculated by the profile difference evaluation unit; an uncertainty level evaluation unit that calculates an optimization target position constituting a spatial position specified on the basis of a variation in spatial density distribution for each spatial position with respect to a plurality of spatial density distributions corresponding to the profile selected by the profile selection unit; a differential regression analysis unit that calculates, using regression analysis, a function for obtaining, from the spatial density distribution, a difference from the input profile by using the data held in the profile database and the difference calculated by the profile difference evaluation unit; and a spatial density distribution optimization unit that calculates a spatial density distribution for which the difference of the function calculated by the differential regression analysis unit is minimized, by using, as a variable, only the spatial density in the optimization target position calculated by the uncertainty level evaluation unit.
According to the present invention, in the particle beam analyzer and the particle beam analysis method, it is possible to estimate an objective and accurate density distribution of a spatial structure while reducing a processing load.
Problems, configurations, and advantageous effects other than the above will be clarified by the following embodiments.
A mode for carrying out the present invention will be described.
The computer system 1 may include one or more processing devices 2A and 2B collectively referred to as the processors 2. Each processor 2 executes instructions stored in the memory 4 and may include an on-board cache. In a certain embodiment, the computer system 1 may include a plurality of processors, and in another embodiment, the computer system 1 may be a system which includes a single processing device. As the processing device, a central processing unit (CPU), a field-programmable gate array (FPGA), a graphics processing unit (GPU), a digital signal processor (DSP), or the like, can be applied.
In a certain embodiment, the memory 4 may include a random access semiconductor memory, a storage device, or a storage medium (either volatile or non-volatile) for storing data and programs. In a certain embodiment, the memory 4 represents the entire virtual memory of the computer system 1, and may include a virtual memory of another computer system connected to the computer system 1 via a network. Although the memory 4 may be conceptually considered to be a single component, in other embodiments this memory 4 may be more complex in configuration, such as a hierarchy of caches and other memory devices. For example, the memory may exist as caches of a plurality of levels, and these caches may be divided up for each function. As a result, one cache may hold instructions, and the other cache may be configured to hold non-instruction data used by the processors. The memory may be distributed and associated with a variety of different processing devices as per a so-called NUMA (Non-Uniform Memory Access) computer architecture.
The memory 4 may store all or a portion of the programs, modules, and data structures for carrying out the functions described herein. For example, the memory 4 may store a potential factor specific application 50. In a certain embodiment, the latent factor specifying application 50 may include instructions or descriptions for performing the functions described below on the processors 2, or may include instructions or descriptions that are interpreted by another instruction or description. In a certain embodiment, the latent factor specifying application 50 may be implemented using hardware via a semiconductor device, a chip, logic gates, a circuit, a circuit card, and/or other physical hardware devices instead of a processor-based system or in addition to a processor-based system. In a certain embodiment, the latent factor specifying application 50 may include data other than instructions or descriptions. In a certain embodiment, a camera, sensor, or another data input device (not shown) may be provided to communicate directly with the bus interface unit 9, the processors 2, or other hardware of the computer system 1. Such a configuration may reduce the need for the processors 2 to access the memory 4 and the latent factor specifying application.
The computer system 1 may include a bus interface unit 9 that performs communications between the processors 2, the memory 4, the display system 24, and the I/O bus interface unit 10. The I/O bus interface unit 10 may be coupled to the I/O bus 8, which serves to transfer data to and from various I/O units. The I/O bus interface unit 10 may communicate with a plurality of I/O interface units 12, 14, 16, and 18, also known as I/O processors (IOP) or I/O adapters (IOA), via the I/O bus 8. The display system 24 may include a display controller, a display memory, or both. The display controller may provide video data, audio data, or both video and audio data to a display device 26. The computer system 1 may also include devices, such as one or more sensors, configured to collect data and provide the data to the processors 2. For example, the computer system 1 may include an environment sensor that collects data such as humidity data, temperature data, and pressure data, a motion sensor that collects data such as acceleration data or motion data, and the like. Other types of sensors can also be used. The display memory may be a dedicated memory for buffering video data. The display system 24 may be connected to a display device 26 such as a single display screen, a television, a tablet, or a portable device. In a certain embodiment, the display device 26 may include a speaker to render audio. Alternatively, a speaker for rendering audio may be connected to the I/O interface unit. In other embodiments, the functions provided by the display system 24 may be implemented by an integrated circuit that includes the processors 2. Similarly, the functions provided by the bus interface unit 9 may be implemented by an integrated circuit including the processors 2.
The I/O interface unit has a function for communicating with various storage or I/O devices. For example, the terminal interface unit 12 enables attachment of a user I/O device 20 such as a user output device, namely, a video display device or a speaker television, for example, or a user input device, namely, a keyboard, a mouse, a keypad, a touchpad, a trackball, a button, a light pen, or another pointing device, for example. The user may use the user interface to operate a user input device to input input data and instructions to the user I/O device 20 and the computer system 1 and to receive output data from the computer system 1. The user interface may, via the user I/O device 20, be displayed on a display device, reproduced by a loudspeaker, or printed via a printer, for example.
One or more disk drives or direct-access storage devices 22 can be attached to the storage interface 14 (typically, the devices are magnetic disk drive storage devices, but may also be a disk drive array configured to appear as a single disk drive, or may be other storage devices). In a certain embodiment, the storage device 22 may be implemented as any secondary storage device. The content of the memory 4 may be stored in the storage device 22 and read, as necessary, from the storage device 22. The network interface 18 may provide a communication path to enable the computer system 1 and other devices to communicate with each other. This communication path may be the network 30, for example.
Although the computer system 1 shown in
In a certain embodiment, the computer system 1 may be a device that receives requests from other computer systems (clients) that do not have a direct user interface, such as a multi-user mainframe computer system, a single-user system, or a server computer. In other embodiments, the computer system 1 may be a desktop computer, a portable computer, a notebook computer, a tablet computer, a pocket computer, a telephone, a smartphone, or any other suitable electronic device.
The particle beam analyzer 100 has functions of a count distribution data reception unit 101, a spatial parameter estimation unit 102, a spatial density distribution optimization unit 103, a particle behavior simulator 104, a differential regression analysis unit 105, an uncertainty level evaluation unit 106, a profile difference evaluation unit 107, and a microspatial distribution data output unit 108. These functions will be described in detail with reference to
The particle beam analyzer 100 inputs the particle beam measurement data 150, performs analysis, and outputs an estimated density distribution 160. Here, examples of particle beams include a proton beam (α ray), an electron beam (β ray), a muon beam, a photon beam (that is, an electromagnetic wave, γ-ray, X-ray, visible light, infrared ray, and the like), and a neutron beam.
As a hardware configuration example, the particle beam analyzer 100 includes a processor 201, a DRAM202, a storage device 203, an input device 204, a monitor 205, and a communications interface 206.
The processor 201 is a processor possessing computation performance, and is a processing unit that performs processing for executing each function illustrated in
The DRAM 202 is a volatile temporary storage area that can be read and written at high speeds. The storage device 203 is a permanent storage area utilizing a hard disk drive (HDD), a flash memory, or the like, and can store the profile database 110 illustrated in
The input device 204 is a device for performing an input such as a mouse, a keyboard, and a control panel for performing an operation. The input device 204 may be provided integrally with the particle beam analyzer 100 or may be connected separately. The input device 204 can correspond to the user I/O device 20 of
The monitor 205 is a display device for showing a user a result of the estimated density distribution 160 and the like illustrated in
The communications interface 206 is an interface such as a serial port for performing communication to exchange information with the outside. The communications interface 206 may correspond to the network interface 18 in
The particle beam reflection measurement device 300 includes a particle beam output unit 301 and a particle beam detection unit 305. The particle beam output unit 301 irradiates the sample 303 with the particle beam 302. The particle beam 302 is then reflected by the sample 303. The reflected particle beam 304 is detected by the particle beam detection unit 305. The information detected by the particle beam detection unit 305 is the brightness of the reflected wave.
An injection angle θ into the sample 303 relative to the sample 303 has a value equal to a reflection angle θ of the particle beam 304 reflected from the sample 303. Therefore, when the position of the injection angle θ is determined, the position of the particle beam output unit 301 is determined, and the position of the corresponding particle beam detection unit 305 is also determined. It has been clarified theoretically that the degrees θ between the injection angle and the reflection angle correspond to the wave count q. Therefore, the angle θ and the wave count q have a proportional relationship (for example, a constant multiple). Further, when the measurement is performed while changing the injection angle θ, it is possible to acquire information on the wave count and the brightness corresponding to the information on the angle and the brightness.
As illustrated in
Next, calculation of the density distribution in the spatial structure will be described.
The relationship between the reflectance of the particle beam and the density at the position of the space can be expressed by the following well-known Equation 1. Examples of the particle beam include an X-ray.
Here, R represents the brightness of the reflected wave. In addition, Q=4 π sin (θ)/λ, where λ represents the wavelength. Therefore, Q represents the wave count, and R is a function of the wave count Q. z represents the depth, in particular the depth from the surface of the sample. ρ is the density, and is a function of z as a density distribution. This equation uses a complex function, but takes an absolute value, and hence, ultimately, a real value is obtained. In practice, calculation may be performed in each of the real part and the imaginary part, and the sum of squares may be finally obtained.
The integration on the right side of Equation 1 above can be converted into the form of addition as in Equation 2 below. This conversion enables calculation by a computer. Here, the differential form of dp/dz is calculated through substitution with a value δ. δ is obtained by taking the difference in density ρ before and after a depth z. As a result, if the density distribution relative to the depth z is known, a brightness R relative to the wave count Q can be calculated.
As described above, the feature of Equation 1 above enables calculation of the distribution of the brightness R relative to the wave count Q as long as the density distribution (spatial density distribution) relative to the depth z (spatial position) is known. However, even if the distribution of the brightness R relative to the wave count Q is known, the structure of the equation is such that the density distribution relative to the depth z cannot be calculated (back-calculated). Therefore, the density distribution relative to the depth z cannot be obtained directly from the measurement result (profile data) as illustrated in
In
In
Therefore, as illustrated in
When the processing is started in the particle beam analyzer 100, in the initial step S1, the count distribution data reception unit 101 receives the particle beam measurement data 150 and calculates the wave count distribution. The data here is profile data of the particle beam measurement data as exemplified in
Next, in step S2, the spatial parameter estimation unit 102 records the wave count distribution as a profile to be estimated (an input profile). The wave count distribution here is the wave count distribution calculated by the count distribution data reception unit 101 in step S1.
Next, in step S3, the spatial density distribution optimization unit 103 performs simulation condition specifying processing. The processing here is performed using the differential regression analysis unit 105, the uncertainty level evaluation unit 106, the profile difference evaluation unit 107, and the profile database 110 in addition to the spatial density distribution optimization unit 103. The simulation condition specifying processing is processing to specify a simulation condition. Specifically, the optimized microspatial distribution is calculated as a simulation condition. The microspatial distribution is a distribution of the density ρ relative to the depth z as described above. Specific details of the processing will be described with reference to
Next, in step S4, the particle behavior simulator 104 executes a simulation as per the simulation condition. The simulation condition here is the simulation condition specified in step S3. In the simulation here, a profile is calculated for an optimized microspatial distribution (spatial density distribution) constituting a simulation condition. The profile is the relationship of the wave count Q with respect to the luminance R, and can be calculated using Equations 1 and 2 above.
Next, in step S5, the profile difference evaluation unit 107 evaluates and stores the difference between the simulation result and the profile to be estimated. The simulation result here is the profile calculated in step S4. The profile to be estimated is the profile to be estimated recorded in step S2. The evaluation of the difference between the profiles can be performed, for example, by computing a difference such as the sum of squares of the differences between the profile values or the sum of absolute values of the profile values. In addition, by using the average of “absolute value of profile value difference÷profile value”, it is possible to calculate what % difference exists. Accordingly, it is possible to equally evaluate a portion having a small absolute value and a portion having a large absolute value. For example, even in a case where the wave count q increases and the profile value decreases, it is possible to uniformly evaluate the difference. The data is stored in the profile database 110. Here, the data to be stored is the content of the optimized microspatial distribution as the simulation condition, the simulation result, and the evaluation of the difference between the profiles, and is stored as a set. Therefore, the profile difference evaluation unit 107 here also functions as a database update unit.
An example of data in the profile database 110 is illustrated in
Next, in step S6, the spatial parameter estimation unit 102 selects a profile which has a small difference from the profile to be estimated. The difference is the same as that in S5, and for example, a difference such as a sum of squares of differences between profile values, a sum of absolute values of profile values, or an average of “absolute value of difference between profile values: profile value” can be used. Among profiles having a small difference, the profile having the smallest difference is selected. The selection here is made from the profile database 110. Therefore, in addition to the profile stored in step S5, a profile already present in the profile database 110 also constitutes a target. Among the profiles, the profile having the smallest difference is selected.
Next, in step S7, the spatial parameter estimation unit 102 evaluates whether the end condition is satisfied. When the end condition is satisfied, the processing advances to step S8. When the end condition is not satisfied, the processing returns to step S3. Various conditions can be applied as the end condition. For example, a case where the difference is within a predetermined threshold value, a case where the number of times the processing is performed is a predetermined number of times or more, and so forth.
In step S8, the microspatial distribution data output unit 108 outputs the estimation result. The estimation result here is the profile data selected in step S6 and the density distribution (microspatial distribution) corresponding thereto. In addition, an index based on the difference may be outputted as described below with reference to
First, in step S301, the spatial density distribution optimization unit 103 sends, to the differential regression analysis unit 105, a request for the function to be optimized. The function here is a function for performing optimization processing, and is expressed by Equation 3 described below.
Next, in step S302, the differential regression analysis unit 105 that has received the request for the function in step S301 issues a request for the difference in profile data to the profile difference evaluation unit 107. The difference here is the difference between the profile to be estimated recorded in step S2 of
Next, in step S303, the profile difference evaluation unit 107 that has received the difference request in step S302 issues a request for profile data to the profile database 110. The profile data to be requested is all the profile data in the profile database 110. Alternatively, the profile data may be profile data in a predetermined range. In the next step S304, the profile difference evaluation unit 107 acquires profile data from the profile database 110. The profile data acquired here is the profile data as exemplified in
Next, in step S305, the profile difference evaluation unit 107 that has acquired the profile data from the profile database 110 in step S304 calculates the difference between the profile data. The difference here is the difference between the profile to be estimated recorded in step S2 of
Next, in step S306, the differential regression analysis unit 105 acquires the profile data differences from the profile difference evaluation unit 107 for which the difference was calculated in step S305. In addition, the acquisition here includes profile data obtained by calculating the difference, and density distribution information corresponding thereto.
Next, in step S307, the differential regression analysis unit 105 that has acquired the profile data differences in step S306 selects a predetermined number of pieces of data in ascending order of the profile data differences. The predetermined number can be determined in advance. For example, the number of items is 10 or more, 30 or more, 50 or more, 100 or less, or 200 or less. In addition, the predetermined number of pieces of data may be determined using a threshold value related to a difference or a variation. That is, profile data having a difference or variation within a predetermined range is selected. As described above, the differential regression analysis unit 105 in step S307 has a function as a profile selection unit.
Next, in step S308, the differential regression analysis unit 105 issues, to the uncertainty level evaluation unit 106, a request for the optimization target depths for the profile data selected in step S307. Here, the optimization target depth is the value of the depth (z) to be optimized.
Next, in step S309, the uncertainty level evaluation unit 106 that has received the optimization target depth request selects the optimization target depth. Here, a predetermined number of depths having large variations in the density value of the microspatial distribution (density distribution) are selected and chosen as optimization target depths. The profile data obtained by performing the selection in step S307 also includes density distribution information. Here, on the basis of the density distribution information, the variation in density value for each depth is examined among the selected data. Then, a predetermined number of depths selected in descending order of variation in density value are selected as the optimization target depths. Alternatively, depths at which the variation in density value is equal to or greater than a predetermined value may be selected as the optimization target depths. The variation here can be evaluated using, for example, a variance value or the like. Therefore, the number of optimization target depths is not limited to one, rather, values may be of a plurality of depths (z) corresponding to variations in density value. The predetermined number of pieces of data is smaller than the total number of depth data for which density was examined, and is, for example, within 30%, particularly within 20%, and more particularly within 10%, or the like, of the number of depth data.
Next, in step S310, the differential regression analysis unit 105 acquires the optimization target depths selected in step S309 from the uncertainty level evaluation unit 106.
Next, in step S311, the differential regression analysis unit 105 calculates the function to be optimized after acquiring the optimization target depth in step S310. The function to be optimized can be estimated using well-known regression analysis. For example, Gaussian process regression can be used, or another known regression analysis may be used. In addition, the calculation may be performed while learning using artificial intelligence (AI).
Here, the function to be optimized can be expressed by the following equation.
Here, y is an objective variable and is the profile data difference. Further, explanatory variables are ρ1, ρ2, ρ3, . . . , and ρn. Here, ρ1, ρ2, ρ3, . . . , and on are values of the density ρ for each depth. Here, an example is illustrated in which the number of explanatory variables is n (n is a natural number), that is, an example in which the density ρ for the depth at n points exists. The differences between the profile data are the differences between the profile to be estimated recorded in step S2 of
The function f in Equation 3 can be calculated using the predetermined number of pieces of profile data selected in step S307, the density distribution information corresponding thereto, and the data of the profile data difference calculated in step S305. The calculation is performed using well-known regression analysis as described above.
Next, in step S312, the spatial density distribution optimization unit 103 acquires the function to be optimized and the optimization target depths from the differential regression analysis unit 105 that calculated the function to be optimized in step S311. The function to be optimized is Equation 3 described above. The optimization target depth is the optimization target depth selected in step S309.
Next, in step S313, the optimization calculation is executed using, as a variable, only the density corresponding to the optimization target depth. In the optimization calculation, processing to find the value of each explanatory variable is performed such that the objective variable y expressed in Equation 3 is minimized or approaches 0. That is, the value of the density ρ (values of ρ1 to ρn) of each depth is found so as to minimize the difference y. The function f in Equation 3 has n variables, but performing the processing by using all the n variables as variables involves a load and time that correspond to the processing. For example, if n is 100, there are 100 variables. Therefore, these values are searched using only the value of the density with respect to the optimization target depth as a variable. The density for other depths is calculated as a fixed value. For example, when the densities related to the optimization target depth are ρ1, ρ5, and ρ10, only these three are set as variables, and the others are set as fixed values. Here, the value of the density ρ which is a fixed value originally has a small variation. Therefore, for example, the value of the density of each depth corresponding to the profile having the smallest difference calculated in step S307 may be used as the fixed value. In addition, the average of the densities at the respective depths corresponding to the profile selected in S307 may be taken as a fixed value.
Any well-known method can be used for the optimization calculation. For example, the method is a gradient descent method, a genetic algorithm, or the like. Parameters such as an initial value of the optimization calculation may be executed in a plurality of settings. In this case, there are a plurality of simulation conditions.
The density values of pl to pn obtained in this manner are microspatial distributions indicating the distribution of the density p relative to the depth z. This is outputted as a simulation condition.
In the analysis result display 400, a measurement data display 410 and a density distribution data display 420 are displayed. In the example of
The measurement data display 410 displays graphs of a estimation target profile 411 and an estimation result profile 412. The estimation target profile 411 is a graph of the profile data recorded in step S2 of
In the density distribution data display 420, a density distribution graph 421 is displayed, and a matching degree 422 is displayed as a numerical value above the density distribution graph. The density distribution graph 421 is a graph display of the density distribution corresponding to the estimation result profile outputted in step S8 of
When the user selects this button, the learning data add button display 430 is switched to the next screen in
By estimating the density distribution data display 420 in
In the data addition screen display 500, a profile display 510 and a density distribution display 520 are displayed. In the example of
The user inputs parameters such as numerical values to the parameter input display section 530. The parameter is a parameter such as a coefficient or a range for a predetermined function or the like. Although
Next, the density distribution display 520 displays the shape of the density distribution graph 521 determined by the parameters inputted to the parameter input display section 530. In the density distribution graph 521 here, the shape is determined by a function determined by the parameters thus inputted. This processing can be performed by the processor 201.
Next, a profile graph 511 corresponding to the density distribution graph 521 is calculated. This calculation can be performed using Equations 1 and 2 described above. This processing can be performed by the processor 201.
Next, when the user selects the transmission button display 540, data related to the profile graph 511 and the density distribution graph 521 is transmitted to the profile database 110. The data configuration is a data value configuration as exemplified in the profile database of
Thus, the user can create a desired density distribution and accumulate, in the profile database 110, profile data corresponding to the density distribution. It is thus possible to increase the accuracy of the processing illustrated in
According to the embodiment as described above, it is possible to more accurately estimate the density distribution by estimating the shape of the density distribution (spatial density distribution) of the spatial structure by means of a method that does not limit the shape. Furthermore, at that time, an optimization calculation is performed. By limiting the variables of the optimization calculation to a depth density which has a large density variation, the load of calculation processing is reduced, and the calculation can be performed in a shorter time. It is also possible to reduce the cost of the processing device performing this processing. These processing steps are performed automatically, and an objective density distribution can be calculated simply by inputting data without requiring special knowledge for analysis. Convenience regarding spatial structure analysis is thus improved.
Note that the present invention is not limited to or by the above-described embodiment and includes various modifications. For example, the above-described embodiments have been described in detail to facilitate understanding of the present invention, and are not necessarily limited to or by embodiments having all the configurations described. In addition, part of the configuration of one embodiment can be replaced with the configuration of another embodiment, and the configuration of the other embodiment can also be added to the configuration of the one embodiment. Moreover, it is possible to add other configurations to part of the configuration of each embodiment, and to delete or substitute part of the configuration of the embodiments.
For example, in the above embodiment, the reflected wave has been described, but the present invention can be applied to transmission other than reflection.
Furthermore, X-ray is mainly given as an example of a particle beam, but other particle beams such as a neutron beam are also applicable. In addition to the particle beam, the present embodiment can also be applied to non-destructive inspection in which an ultrasonic wave is injected in a measurement target and reverberation thereof is subjected to frequency analysis, seismic source estimation based on frequency analysis of earthquake motion, and the like. As described above, the present embodiment can also be applied to an object that requires inverse estimation analysis in a case where it is difficult to directly measure a measurement target but can be measured as a frequency signal.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/JP2022/000971 | 1/13/2022 | WO |
