X-ray measurement method and apparatus

Abstract

The embodiment of the present application provides an X-ray measurement method and device, comprising: measuring a to-be-detected object from different angles through X-rays to obtain a plurality of three-dimensional images, performing dimension reduction processing on the three-dimensional images to obtain a two-dimensional image of the to-be-detected object; extracting spatial electron density information of the to-be-detected object at each layer from the two-dimensional image, and further determining key spatial electron density information associated with a measurement parameter; determining an interlayer correlation corresponding to a current layer according to the key spatial electron density information of the to-be-detected object at a reference layer and the key spatial electron density information of the to-be-detected object at the current layer; the reference layer is at least one layer adjacent to the current layer in the to-be-detected object; and determining measurement data of the to-be-detected object at the current layer for the measurement parameter according to the key spatial electron density information of the to-be-detected object at the current layer, the interlayer correlation corresponding to the current layer, and measurement data of the reference layer for the measurement parameter.

Description

Technical Field

[0001] This invention relates to the field of measurement and analysis technology, and in particular to an X-ray measurement method and apparatus. Background Technology

[0002] X-ray measurement technology uses the interaction of X-rays with matter to obtain information about the internal structure and composition of an object. This technology has been widely used in many fields such as medical imaging, industrial inspection, and materials science.

[0003] Currently, the MCMC (Markov Chain Monte Carlo) method is generally used to measure information about the internal structure and composition of an object. However, this technique relies on empirical modeling and complex mathematical calculations, resulting in slow computation speed and low measurement efficiency.

[0004] In summary, improving the efficiency of X-ray measurement is a pressing technical problem that needs to be solved. Summary of the Invention

[0005] This invention provides an X-ray measurement method to solve the problem of low measurement efficiency of X-rays in the prior art.

[0006] In a first aspect, an X-ray measurement method provided by an embodiment of the present invention includes: measuring an object to be detected from different angles using X-rays to obtain multiple three-dimensional images; performing dimensionality reduction processing on the three-dimensional images to obtain a two-dimensional image of the object to be detected; extracting spatial electron density information of the object to be detected in each layer from the two-dimensional images, and then determining key spatial electron density information associated with measurement parameters; determining the interlayer correlation corresponding to the current layer based on the key spatial electron density information of the object to be detected in a reference layer and the key spatial electron density information in the current layer; the reference layer being at least one layer of the object to be detected adjacent to the current layer; and determining the measurement data of the object to be detected in the current layer for the measurement parameters based on the key spatial electron density information of the object to be detected in the current layer, the interlayer correlation corresponding to the current layer, and the measurement data of the reference layer for the measurement parameters.

[0007] In the above technical solution, for any given measurement parameter, key spatial electron density information of the object under test is extracted from multiple spatial electron density information in each layer of the object under test. This allows for more targeted generation of measurement data for the object under test for the measurement parameter. By leveraging the interlayer correlation between adjacent layers, the measurement data of the reference layer can be used to optimize the measurement data of the object under test for the measurement parameter in the current layer, thereby improving the accuracy of the measurement data for the object under test for the measurement parameter in the current layer.

[0008] Optionally, the method further includes: determining the measurement data for the measurement parameters of the first layer based on the key spatial electron density information of the first layer.

[0009] Optionally, interlayer correlation includes spatial correlation, material correlation, and functional correlation; spatial correlation is used to characterize the spatial smoothness relationship between two adjacent layers; material correlation is used to characterize the similarity of material properties between two adjacent layers; and functional correlation is used to characterize the influence relationship of the functional modules to which two adjacent layers belong in the functional structure.

[0010] In the above technical solution, by taking into account the interlayer correlation, this solution can subsequently optimize the measurement data of the measurement parameters of the object under test by using the measurement data of the measurement parameters of the object under test in adjacent layers.

[0011] Optionally, key spatial electron density information associated with the measurement parameters is determined from multiple spatial electron density information in each layer, including: nonlinear transformation of multiple hidden layers based on a recurrent neural network to determine key spatial electron density information associated with the measurement parameters from multiple spatial electron density information in each layer; and interlayer correlation corresponding to the current layer is determined based on the key spatial electron density information of the object to be detected in the reference layer and the key spatial electron density information in the current layer, including: recurrent connection mechanism based on a recurrent neural network to determine the interlayer correlation corresponding to the current layer based on the key spatial electron density information of the object to be detected in the reference layer and the key spatial electron density information in the current layer.

[0012] In the above technical solution, key spatial electron density information related to the measurement parameter is determined from multiple spatial electron density information of each layer of the object under test. This allows for more targeted generation of measurement data for the object under test. This solution also considers the interlayer correlation between layers of the object under test, and can optimize the measurement data of the object under test through this interlayer correlation.

[0013] Optionally, the recurrent neural network includes multiple channels, each corresponding to a measurement parameter, and each channel has multiple hidden layers and a recurrent connection mechanism.

[0014] In the above technical solution, by setting multiple channels in the recurrent neural network, the channels have a certain degree of isolation, and under the premise of determining the measurement data of multiple measurement parameters in parallel, they will not interfere with each other, thereby improving the accuracy of the measurement data of the measurement parameters.

[0015] Optionally, the three-dimensional images from different angles are dimensionality reduced to obtain a two-dimensional image of the object to be detected, including: for any angle, the three-dimensional image of the angle is dimensionality reduced to obtain two-dimensional bars of the angle; the M bars located in the central region of the two-dimensional bars of the angle are taken as two-dimensional sub-images of the angle; and the two-dimensional sub-images of each angle are stitched together according to the detection order to obtain a two-dimensional image of the object to be detected.

[0016] In the above technical solution, a two-dimensional image is obtained by dimensionality reduction of a three-dimensional image, thereby obtaining the structural information of the object to be detected, which facilitates the improvement of the calculation efficiency of measurement data for subsequent determination of the measurement parameters of the object to be detected.

[0017] Optionally, the measurement parameters include alignment error, feature size, and torsion angle; after determining the measurement data of the object to be detected for the measurement parameters in the current layer, the method further includes: determining the reconstructed contour of the object to be detected based on the measurement data of each layer under each measurement parameter.

[0018] Secondly, embodiments of the present invention provide an X-ray measurement device, comprising: an acquisition unit, configured to measure an object to be detected from different angles using X-rays to obtain multiple three-dimensional images, and to perform dimensionality reduction processing on the three-dimensional images to obtain a two-dimensional image of the object to be detected; and a processing unit, configured to extract spatial electron density information of the object to be detected in each layer from the two-dimensional image, thereby determining key spatial electron density information associated with measurement parameters; determining the interlayer correlation corresponding to the current layer based on the key spatial electron density information of the object to be detected in a reference layer and the key spatial electron density information in the current layer; the reference layer being at least one layer of the object to be detected adjacent to the current layer; and determining the measurement data of the object to be detected in the current layer for the measurement parameters based on the key spatial electron density information of the object to be detected in the current layer, the interlayer correlation corresponding to the current layer, and the measurement data of the reference layer for the measurement parameters.

[0019] Optionally, the processing unit is also used to: determine the measurement data of the first layer for the measurement parameters based on the key spatial electron density information of the first layer.

[0020] Optionally, interlayer correlation includes spatial correlation, material correlation, and functional correlation; spatial correlation is used to characterize the spatial smoothness relationship between two adjacent layers; material correlation is used to characterize the similarity of material properties between two adjacent layers; and functional correlation is used to characterize the influence relationship of the functional modules to which two adjacent layers belong in the functional structure.

[0021] Optionally, the processing unit is specifically used for: nonlinear transformation based on multiple hidden layers of a recurrent neural network to determine key spatial electron density information associated with the measurement parameters from multiple spatial electron density information in each layer. The processing unit is also specifically used for: determining the inter-layer correlation corresponding to the current layer based on the recurrent connection mechanism of the recurrent neural network, according to the key spatial electron density information of the object to be detected in the reference layer and the key spatial electron density information in the current layer.

[0022] Optionally, the recurrent neural network includes multiple channels, each corresponding to a measurement parameter, and each channel has multiple hidden layers and a recurrent connection mechanism.

[0023] Optionally, the acquisition unit is specifically used for: performing dimensionality reduction processing on the three-dimensional image of any angle to obtain two-dimensional bars of the angle; taking the M bars located in the central region of the two-dimensional bars of the angle as two-dimensional sub-images of the angle; and stitching the two-dimensional sub-images of each angle together according to the detection order to obtain a two-dimensional image of the object to be detected.

[0024] Optionally, the measurement parameters include alignment error, feature size, and torsion angle; the processing unit is also used to: determine the reconstructed contour of the object to be detected based on the measurement data of each layer under each measurement parameter.

[0025] Thirdly, embodiments of the present invention also provide an electronic device, including at least one processor and at least one memory, wherein the memory stores a computer program, and when the program is executed by the processor, the processor performs an X-ray measurement method as described in the first aspect.

[0026] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program executable by a computer device, which, when run on the computer device, causes the computer device to perform the X-ray measurement method described in the first aspect. Attached Figure Description

[0027] Figure 1 A flowchart of an X-ray measurement method provided in an embodiment of the present invention;

[0028] Figure 2 A flowchart of a method for determining a two-dimensional image of an object to be detected, provided by an embodiment of the present invention;

[0029] Figure 3 A schematic diagram of a two-dimensional image provided in an embodiment of the present invention;

[0030] Figure 4 This is a schematic diagram of the structure of an X-ray measuring device provided in an embodiment of the present invention;

[0031] Figure 5This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0033] Integrated circuit (IC) nanometer measurement technology is an effective means to ensure the quality and consistency of IC manufacturing. During mass production, rapid, non-destructive, and accurate measurement of IC nanostructure data is required. This measurement data includes not only alignment errors, feature dimensions, twist angles, spacing variations, film thickness, stacking, linewidth, lineheight, and sidewall angles, but also important characteristics such as linewidth roughness (LER) and line edge roughness (LWR). In recent years, to ensure that IC structures maintain excellent performance as they shrink, their structural design has gradually shifted from simple planar structures to complex three-dimensional structures. The critical dimensions of these complex three-dimensional structures are even smaller, requiring more measurement data and placing higher demands on IC nanometer measurement technology.

[0034] This invention provides an X-ray measurement method for measuring measurement data of an object to be tested. The object to be tested can be an integrated circuit (IC), a field-effect transistor, a three-dimensional NAND device, or other nanostructure devices, and is not limited thereto.

[0035] like Figure 1 The diagram shown is a flowchart of an X-ray measurement method provided by an embodiment of the present invention. The method includes the following steps:

[0036] Step 101: Measure the object to be detected from different angles using X-rays to obtain multiple three-dimensional images. Perform dimensionality reduction processing on the three-dimensional images to obtain a two-dimensional image of the object to be detected.

[0037] In this embodiment of the invention, X-rays are used to measure the object to be detected from different angles to obtain three-dimensional images from different angles. This makes it easier to accurately determine the structural information of the object to be detected based on the three-dimensional images from different angles. The object to be detected has multiple bright spot structures and dark detail structures.

[0038] Because 3D images are larger than 2D images and typically have higher resolution, loading and analyzing 3D images takes a longer time. To improve the efficiency of determining measurement data for the target object with respect to measurement parameters, 3D images from different angles can be dimensionality-reduced to obtain 2D images of the target object. This improves the efficiency of determining measurement data for the target object with respect to measurement parameters while preserving the structural information of the target object.

[0039] Step 102: Extract the spatial electron density information of the object to be detected in each layer from the two-dimensional image, and then determine the key spatial electron density information associated with the measurement parameters.

[0040] In this embodiment of the invention, since the object to be detected has a multi-layer structure, multiple spatial electron density information of the object to be detected in each layer can be extracted from the two-dimensional image of the object to be detected.

[0041] Because there are various types of measurement parameters, including alignment error, feature size, torsion angle, spacing variation, film thickness, stacking, linewidth, lineheight, sidewall angle and linewidth roughness, and line edge roughness, the spatial electron density information associated with different measurement parameters is different. To improve the accuracy of determining the measurement data of the object under test for each measurement parameter, for any given measurement parameter, key spatial electron density information associated with the measurement parameter is determined from multiple spatial electron density information sources in each layer. This allows for more targeted generation of measurement data for the object under test for the measurement parameter.

[0042] Step 103: Determine the interlayer correlation corresponding to the current layer based on the key spatial electron density information of the object to be detected in the reference layer and the key spatial electron density information in the current layer.

[0043] In this embodiment of the invention, since the object to be detected has a multi-layered structure, and the layers are correlated, a high inter-layer correlation indicates a high similarity between the measurement data of the two layers for the measurement parameters. Therefore, the inter-layer correlation of the current layer is first determined based on the key spatial electron density information of the object to be detected in the reference layer and the key spatial electron density information in the current layer. This facilitates subsequent optimization of the measurement data of the current layer for the measurement parameters based on the inter-layer correlation of the current layer. The reference layer is at least one layer in the object to be detected that is adjacent to the current layer; the inter-layer correlation is used to characterize the similarity between adjacent layers.

[0044] Step 104: Based on the key spatial electron density information of the object to be detected in the current layer, the interlayer correlation of the current layer, and the measurement data of the reference layer for the measurement parameters, determine the measurement data of the object to be detected for the measurement parameters in the current layer.

[0045] In this embodiment of the invention, the degree of influence of the measurement data of the object to be detected on the measurement parameters in the reference layer and the measurement data of the object to be detected on the measurement parameters in the current layer is determined by considering the level of inter-layer correlation in the current layer. The higher the inter-layer correlation, the greater the influence. Thus, the accuracy of determining the measurement data of the object to be detected on the measurement parameters in the current layer can be improved based on the inter-layer correlation.

[0046] As can be seen from steps 101 to 104 above, by selectively extracting key spatial electron density information of the target object for any given measurement parameter from multiple spatial electron density information in each layer of the target object, more targeted measurement data of the target object for the measurement parameter can be generated subsequently. Through the interlayer correlation between adjacent layers, the measurement data of the target object for the measurement parameter in the current layer can be optimized using the measurement data of the reference layer of the current layer, thereby improving the accuracy of the measurement data of the target object for the measurement parameter in the current layer.

[0047] To accurately determine the measurement data of the object under inspection at each layer for the measurement parameters, it is first necessary to obtain three-dimensional images from different angles using X-rays. Then, these three-dimensional images are dimensionality-reduced to obtain two-dimensional images of the object under inspection. To facilitate understanding of this method, the following describes how to perform dimensionality reduction on three-dimensional images from different angles to obtain two-dimensional images of the object under inspection.

[0048] like Figure 2 The diagram shown is a flowchart of a method for determining a two-dimensional image of an object to be detected according to an embodiment of the present invention. The method includes the following steps:

[0049] Step 201: For any three-dimensional image at any angle, perform dimensionality reduction processing on the three-dimensional image at the angle to obtain two-dimensional bars of the angle, and take the M bars located in the central region of the two-dimensional bars of the angle as two-dimensional sub-images of the angle.

[0050] In this embodiment of the invention, for any angle of the 3D image to be detected, dimensionality reduction processing is performed on the 3D image to be detected to obtain two-dimensional bars for that angle. Then, the M bars located in the central region of the two-dimensional bars for that angle are taken as the two-dimensional sub-image of that angle. For example, if there are 100 angles of the 3D image to be detected, and M is 5, the dimensionality reduction processing of the 3D image to be detected at each angle yields 250 two-dimensional bars. For each angle, 5 two-dimensional bars are selected. The principle for selecting the two-dimensional bars is to take 2 bars forward and 2 bars backward based on the central axis of the two-dimensional bars. These 5 two-dimensional bars are the two-dimensional sub-image of that angle.

[0051] Step 202: According to the detection order, the two-dimensional sub-images from each angle are stitched together to obtain a two-dimensional image of the object to be detected.

[0052] In this embodiment of the invention, according to step 201, a two-dimensional sub-image for each angle can be determined. Then, according to a set detection order, the two-dimensional sub-images for each angle are stitched together to obtain a two-dimensional image. See also... Figure 3 .

[0053] As can be seen from steps 201 to 202 above, by reducing the dimensionality of the three-dimensional image to obtain a two-dimensional image, the structural information of the object to be detected can be obtained, which facilitates the improvement of the calculation efficiency of measurement data for determining the measurement parameters of the object to be detected.

[0054] In this embodiment of the invention, after obtaining a two-dimensional image of the object to be detected, it is necessary to extract multiple spatial electron density information of the object in each layer from the two-dimensional image. This facilitates the subsequent determination of key spatial electron density information associated with measurement parameters from the multiple spatial electron density information of each layer, thereby making the measurement data determined based on the key spatial electron density information more targeted, and thus improving the efficiency and accuracy of the measurement data. To facilitate understanding of this solution, the following describes how to extract multiple spatial electron density information of the object in each layer from the two-dimensional image of the object to be detected.

[0055] Optionally, a two-dimensional image of the object to be detected is input into the first module, which outputs multiple spatial electron density information of the object in each layer. The first module consists of at least one of FCN, Unet, ResNet, dilated convolution, Enet, DenseNet, and R-CNN.

[0056] In this embodiment of the invention, the measurement parameters include alignment error, feature size, torsion angle, spacing variation, film thickness, stacking, linewidth, line height, sidewall angle, linewidth roughness, and line edge roughness, etc., which are not limited here. Since there are many types of measurement parameters, in order to generate more targeted measurement data for the object under test, it is necessary to target the corresponding measurement parameters. The following describes in detail how to determine the key spatial electron density information related to the measurement parameter from multiple spatial electron density information of each layer of the object under test.

[0057] Optionally, based on the nonlinear transformation of multiple hidden layers of the second module, key spatial electron density information associated with the measurement parameters is determined from multiple spatial electron density information in each layer. The second module is a recurrent neural network; specifically, it can be a long short-term memory network, a gated recurrent network, a bidirectional long short-term memory network, a bidirectional gated recurrent network, or a variant of the above models, without limitation.

[0058] In this embodiment of the invention, since the object to be tested has a multi-layered structure, the key spatial electron density information associated with the measurement parameter in each layer of the object to be tested can be determined for any measurement parameter, based on the above description. This scheme takes into account the inter-layer correlation between the layers of the object to be tested, and the measurement data of the object to be tested can be optimized through this inter-layer correlation. The following describes how to determine the inter-layer correlation corresponding to the current layer based on the key spatial electron density information of the object to be tested in the reference layer and the key spatial electron density information in the current layer.

[0059] Optionally, a recurrent connection mechanism based on a recurrent neural network is used to determine the interlayer correlation corresponding to the current layer based on the key spatial electron density information of the object to be detected in the reference layer and the key spatial electron density information in the current layer.

[0060] Optionally, the measurement data for the measurement parameters of the first layer can be determined based on the key spatial electron density information of the first layer.

[0061] For example, consider a three-layer structure. The measurement data for the measurement parameters in the first layer is determined based on the key spatial electron density information of the first layer. Then, using a recurrent neural network's loop connection mechanism, the interlayer correlation of the second layer is determined based on the key spatial electron density information of the object in the first and second layers. The first layer serves as the reference layer for the second layer. There are two methods to determine the interlayer correlation of the third layer. The first method is based on a recurrent neural network's loop connection mechanism, determining the interlayer correlation of the third layer based on the key spatial electron density information of the object in the first, second, and third layers. The third layer is the current layer of the object, and the first and second layers serve as reference layers for the third layer.

[0062] The second approach is based on a recurrent neural network-based recurrent connection mechanism. It determines the interlayer correlation of the third layer based on the key spatial electron density information of the second and third layers. The second layer serves as the reference layer for the third layer.

[0063] Optionally, interlayer correlation includes spatial correlation, material correlation, and functional correlation. Spatial correlation is used to characterize the spatial smoothness relationship between two adjacent layers. Material correlation is used to characterize the similarity of material properties between two adjacent layers. Functional correlation is used to characterize the influence relationship of functional modules belonging to two adjacent layers on the functional structure.

[0064] In this embodiment of the invention, interlayer correlation includes spatial correlation, material correlation, and functional correlation. The key spatial electron density information of each layer is used to characterize the spatial smoothness information within that layer, which is determined by the contrast relationship between bright and dark structures within that layer. The key spatial electron density information of each layer also characterizes the material properties of that layer and the functional structure information of the functional modules to which that layer belongs. Specifically, if the interlayer correlation of the second layer is determined based on the key spatial electron density information of the object under test in the first layer and the key spatial electron density information of the object under test in the second layer, a high interlayer correlation indicates that the spatial smoothness trends between the first and second layers are similar, the material properties between the first and second layers are highly similar, and the functional modules to which the first and second layers belong have a high degree of mutual influence and dependence on each other in their functional structures.

[0065] Therefore, determining the preliminary measurement data for the measurement parameters in the second layer solely based on the key spatial electron density information of the second layer would result in low accuracy. This solution utilizes the interlayer correlation corresponding to the second layer. A high interlayer correlation indicates a high similarity between the measurement data of the target object in the first layer and its measurement data in the second layer. This allows for the optimization of the preliminary measurement data of the target object in the second layer using the first-layer measurement data, thereby improving the accuracy of the measurement data in the second layer.

[0066] The following describes how to determine the measurement data of the object to be tested for the measurement parameters in the current layer.

[0067] Optionally, the interlayer correlation corresponding to the current layer is determined based on the key spatial electron density information of the object to be detected in the reference layer and the key spatial electron density information in the current layer; the reference layer is at least one layer of the object to be detected that is adjacent to the current layer; the interlayer correlation is used to characterize the similarity between adjacent layers; i is a positive integer.

[0068] Optionally, after determining the measurement data of the object to be inspected for the measurement parameters in the current layer, the reconstructed contour of the object to be inspected is determined based on the measurement data of each layer of the object to be inspected under the three measurement parameters of alignment error, feature size, and torsion angle.

[0069] Optionally, this solution involves a reconstruction model, which includes a first module and a second module. By inputting a two-dimensional image of the object to be detected into the reconstruction model, the first module extracts multiple spatial electron density information of the object at each layer from the two-dimensional image. Then, this spatial electron density information is input into the second module. For any given measurement parameter, the second module determines the key spatial electron density information associated with the measurement parameter from the multiple spatial electron density information at each layer. Based on the key spatial electron density information of the object at the reference layer and the key spatial electron density information at the current layer, the interlayer correlation corresponding to the current layer is determined. Based on the key spatial electron density information of the object at the current layer, the interlayer correlation corresponding to the (i+1)th layer, and the measurement data of the object at the reference layer for the measurement parameter, the measurement data of the object at the current layer for the measurement parameter is determined. The second module in this solution is a correlation learning module, which can accelerate the computation speed and improve the efficiency of determining measurement data compared to the modeling process that requires multiple iterations in existing technologies.

[0070] Optionally, the recurrent neural network includes multiple channels, each corresponding to a measurement parameter, and each channel has multiple hidden layers and a recurrent connection mechanism. For example, if the measurement parameters include alignment error, feature size, and twist angle, then the recurrent neural network includes three channels, each corresponding to a measurement parameter, and each channel has multiple hidden layers and a recurrent connection mechanism. This allows for the parallel determination of measurement data for multiple measurement parameters. Furthermore, due to the isolation between channels, they do not interfere with each other while determining the measurement data for multiple measurement parameters in parallel, thereby improving the accuracy of the measurement data.

[0071] Based on the same technical concept, embodiments of the present invention provide an X-ray measuring device, such as... Figure 4 As shown, the device 400 includes: an acquisition unit 401, used to measure the object to be detected from different angles using X-rays to obtain multiple three-dimensional images, and to perform dimensionality reduction processing on the three-dimensional images to obtain a two-dimensional image of the object to be detected. A processing unit 402 is used to extract the spatial electron density information of the object to be detected in each layer from the two-dimensional image, and then determine the key spatial electron density information associated with the measurement parameters; based on the key spatial electron density information of the object to be detected in the reference layer and the key spatial electron density information in the current layer, determine the interlayer correlation corresponding to the current layer; the reference layer is at least one layer of the object to be detected adjacent to the current layer; based on the key spatial electron density information of the object to be detected in the current layer, the interlayer correlation corresponding to the current layer, and the measurement data of the reference layer for the measurement parameters, determine the measurement data of the object to be detected in the current layer for the measurement parameters.

[0072] Optionally, the processing unit 402 is also configured to: determine the measurement data of the first layer for the measurement parameters based on the key spatial electron density information of the first layer.

[0073] Optionally, interlayer correlation includes spatial correlation, material correlation, and functional correlation; spatial correlation is used to characterize the spatial smoothness relationship between two adjacent layers; material correlation is used to characterize the similarity of material properties between two adjacent layers; and functional correlation is used to characterize the influence relationship of the functional modules to which two adjacent layers belong in the functional structure.

[0074] Optionally, processing unit 402 is specifically used for: nonlinear transformation based on multiple hidden layers of a recurrent neural network to determine key spatial electron density information associated with the measurement parameters from multiple spatial electron density information in each layer. Processing unit 402 is specifically used for: based on the recurrent connection mechanism of a recurrent neural network, determining the interlayer correlation corresponding to the current layer according to the key spatial electron density information of the object to be detected in the reference layer and the key spatial electron density information in the current layer.

[0075] Optionally, the recurrent neural network includes multiple channels, each corresponding to a measurement parameter, and each channel has multiple hidden layers and a recurrent connection mechanism.

[0076] Optionally, the acquisition unit 401 is specifically used for: performing dimensionality reduction processing on the three-dimensional image of any angle to obtain two-dimensional bars of the angle; taking the M bars located in the central region of the two-dimensional bars of the angle as two-dimensional sub-images of the angle; and stitching the two-dimensional sub-images of each angle together according to the detection order to obtain a two-dimensional image of the object to be detected.

[0077] Optionally, the measurement parameters include alignment error, feature size, and torsion angle; the processing unit 402 is also used to: determine the reconstructed contour of the object to be detected based on the measurement data of each layer under each measurement parameter.

[0078] Based on the same technical concept, embodiments of this application also provide an electronic device, such as... Figure 5 As shown, the electronic device 500 includes at least one processor 501 and a memory 502 connected to the at least one processor. In this embodiment, the specific connection medium between the processor 501 and the memory 502 is not limited. Figure 5 Taking the connection between processor 501 and memory 502 via a bus as an example, the bus can be divided into address bus, data bus, control bus, etc.

[0079] In this embodiment of the application, the memory 502 stores instructions that can be executed by at least one processor 501. By executing the instructions stored in the memory 502, at least one processor 501 can perform the steps included in the aforementioned X-ray measurement method.

[0080] The processor 501 is the control center of the computing device. It can connect to various parts of the computing device using various interfaces and lines. It performs data processing by running or executing instructions stored in the memory 502 and calling data stored in the memory 502. The processor 501 can be a general-purpose processor, such as a central processing unit (CPU), digital signal processor, application-specific integrated circuit (ASIC), field-programmable gate array or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component. It can implement or execute the methods, steps and logic block diagrams disclosed in the embodiments of this application.

[0081] Memory 502, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules. Memory 502 may include at least one type of storage medium, such as flash memory, hard disk, multimedia card, card-type memory, random access memory (RAM), static random access memory (SRAM), programmable read-only memory (PROM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), magnetic storage, magnetic disk, optical disk, etc.

[0082] Based on the same technical concept, embodiments of this application also provide a computer-readable storage medium storing a computer program executable by a computing device, which, when run on an electronic device, causes the electronic device to perform the steps of the above-described X-ray measurement method.

[0083] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0084] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0085] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0086] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0087] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0088] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. An X-ray measurement method, characterized in that, include: Multiple three-dimensional images are obtained by measuring the object to be detected from different angles using X-rays. The three-dimensional images are then subjected to dimensionality reduction processing to obtain a two-dimensional image of the object to be detected. The spatial electron density information of the object to be detected in each layer is extracted from the two-dimensional image, thereby determining the key spatial electron density information associated with the measurement parameters; Based on the key spatial electron density information of the object to be detected in the reference layer and the key spatial electron density information in the current layer, the interlayer correlation corresponding to the current layer is determined. The reference layer is at least one layer in the object to be detected that is adjacent to the current layer; Based on the key spatial electron density information of the object under test in the current layer, the interlayer correlation of the current layer, and the measurement data of the reference layer for the measurement parameters, the measurement data of the object under test for the measurement parameters in the current layer are determined.

2. The method as described in claim 1, characterized in that, The method further includes: determining the measurement data of the first layer for the measurement parameters based on the key spatial electron density information of the first layer.

3. The method as described in claim 1, characterized in that, The interlayer correlation includes spatial correlation, material correlation, and functional correlation; The spatial correlation is used to characterize the spatial smoothness relationship between two adjacent layers; The material correlation is used to characterize the similarity of material properties between two adjacent layers; The functional correlation is used to characterize the influence relationship between functional modules belonging to two adjacent layers in the functional structure.

4. The method as described in claim 1, characterized in that, Key space electron density information associated with the measurement parameters is determined from multiple space electron density information in each layer, including: Based on the nonlinear transformation of multiple hidden layers of a recurrent neural network, key spatial electron density information associated with the measurement parameters is determined from multiple spatial electron density information in each layer. Based on the key spatial electron density information of the object to be detected in the reference layer and the key spatial electron density information in the current layer, the interlayer correlation corresponding to the current layer is determined, including: Based on the recurrent connection mechanism of the recurrent neural network, the interlayer correlation corresponding to the current layer is determined according to the key spatial electron density information of the object to be detected in the reference layer and the key spatial electron density information in the current layer.

5. The method as described in claim 4, characterized in that, The recurrent neural network includes multiple channels, each corresponding to a measurement parameter, and each channel has multiple hidden layers and a recurrent connection mechanism.

6. The method according to any one of claims 1 to 5, characterized in that, The three-dimensional images from different angles are subjected to dimensionality reduction processing to obtain a two-dimensional image of the object to be detected, including: For any three-dimensional image at any angle, the three-dimensional image at that angle is subjected to dimensionality reduction processing to obtain a two-dimensional bar chart of that angle; The M bars located in the central region of the two-dimensional bar chart of the angle are taken as the two-dimensional sub-charts of the angle; Following the detection sequence, the two-dimensional sub-images from each angle are stitched together to obtain a two-dimensional image of the object to be detected.

7. The method according to any one of claims 1 to 5, characterized in that, The measurement parameters include alignment error, feature size, and torsion angle; After determining the measurement data of the object to be detected for the measurement parameters in the current layer, the method further includes: The reconstructed contour of the object to be detected is determined based on the measurement data of each layer under each measurement parameter.

8. An X-ray measuring device, characterized in that, include: The acquisition unit is used to measure the object to be detected from different angles using X-rays to obtain multiple three-dimensional images, and to perform dimensionality reduction processing on the three-dimensional images to obtain a two-dimensional image of the object to be detected. The processing unit is configured to extract the spatial electron density information of the object to be detected in each layer from the two-dimensional image, and then determine the key spatial electron density information associated with the measurement parameters; and determine the interlayer correlation corresponding to the current layer based on the key spatial electron density information of the object to be detected in the reference layer and the key spatial electron density information in the current layer. The reference layer is at least one layer of the object to be detected that is adjacent to the current layer; based on the key spatial electron density information of the object to be detected in the current layer, the interlayer correlation corresponding to the current layer, and the measurement data of the reference layer for the measurement parameters, the measurement data of the object to be detected for the measurement parameters in the current layer are determined.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the method according to any one of claims 1 to 7.

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

Images