Method for measuring transmission electron microscope tem images and related apparatus

Abstract

The application discloses a kind of transmission electron microscope (TEM) image measurement method and related device, wherein the method comprises: obtaining the TEM image of semiconductor device, TEM image is classified by measurement component, the category of the measurement component of TEM image is obtained, then the target measurement area of TEM image is determined based on the category of measurement component, wherein the target measurement area is the area where the measurement component is located in TEM image.Electronic device measures target measurement point in target measurement area, and obtains the measurement result of TEM image.The embodiments of the present application are beneficial to improve the measurement and analysis efficiency of the TEM image of semiconductor device.

Description

Technical Field

[0001] This application relates to the field of semiconductor process development, and in particular to a method and related apparatus for measuring transmission electron microscope (TEM) images. Background Technology

[0002] Transmission electron microscopy (TEM) is crucial in semiconductor manufacturing processes, and the high-resolution images it acquires are the primary input for detecting and analyzing semiconductor process anomalies. Current TEM image analysis requires researchers to import the images into specialized analysis software (Digital Micrograph, DM). Based on DM, manual measurements are performed, which is inefficient and cannot meet the analytical needs of TEM images used in semiconductor manufacturing processes. Summary of the Invention

[0003] This application provides a method and related apparatus for measuring transmission electron microscope (TEM) images, which helps to improve the measurement and analysis efficiency of TEM images of semiconductor devices.

[0004] In a first aspect, embodiments of this application provide a method for measuring transmission electron microscopy (TEM) images, applicable to electronic devices, chips within electronic devices, or functional modules within electronic devices. Taking an application to an electronic device as an example, the method includes: An electronic device acquires a TEM image of a semiconductor device, classifies the TEM image according to the measurement components, and then determines the target measurement region of the TEM image based on the category of the measurement components. The target measurement region is the area in the TEM image where the measurement components are located. The electronic device then measures the target measurement points within the target measurement region to obtain the measurement results of the TEM image.

[0005] As can be seen, in this embodiment, the electronic device acquires TEM images of semiconductor devices, classifies the TEM images into measurement components, determines the target measurement area where the measurement components are located in the TEM image based on the category of the measurement components, and finally measures the target measurement points in the target measurement area to obtain the measurement results. Compared with manual measurement using DM software, this application can complete the process automatically through electronic devices, which is beneficial to improving the measurement efficiency of TEM images and thus meeting the analysis needs of TEM images in semiconductor manufacturing processes.

[0006] In one possible implementation, in determining the target measurement region of a TEM image based on the category of the measurement component, the electronic device determines an image segmentation model corresponding to the category of the measurement component, then uses the image segmentation model to perform image semantic segmentation on the TEM image to obtain a mask image corresponding to the TEM image, and determines the target measurement region based on the mask image.

[0007] In this implementation, the electronic device can determine the image segmentation model suitable for the current measurement component based on the correspondence between the measurement component category and the image segmentation model. Using this image segmentation model to perform semantic segmentation on the TEM image helps to improve the accuracy of image segmentation, thereby improving the accuracy of determining the target measurement region.

[0008] In one possible implementation, after determining the target measurement area of ​​the TEM image based on the category of the measurement component, the electronic device uses a clustering algorithm to cluster the points within the target measurement area to obtain point clusters. For each point cluster, its boundary contour is identified, candidate measurement points on the boundary contour are extracted, and the target measurement points within the target measurement area are determined based on the candidate measurement points.

[0009] In this implementation, the electronic device obtains continuous boundary contours within the target measurement area through a clustering algorithm, extracts candidate measurement points with obvious features from these boundary contours, and finally determines the target measurement point based on the candidate measurement points. This is beneficial for screening out target measurement points with obvious boundary structure features of the measurement components.

[0010] In one possible implementation, in determining the target measurement point based on candidate measurement points, the electronic device employs the Otsu's method (maximum inter-class variance) to segment the TEM image, aiming to maximize the inter-class variance between the foreground and background, thus obtaining the foreground points of the TEM image. The foreground points are pixels in the region where the measurement component is located in the TEM image. The electronic device then identifies multiple points belonging to the foreground points from the candidate measurement points as the target measurement points.

[0011] In this implementation, the electronic device can segment the foreground points of the measurement component using the OTSU threshold segmentation algorithm, and determine the points that belong to the foreground points among the candidate measurement points as the final target measurement points, which is beneficial for selecting more representative target measurement points.

[0012] In one possible implementation, after measuring the target measurement points within the target measurement area and obtaining the measurement results of the TEM image, the electronic device generates a measurement report of the semiconductor device based on the measurement results of the TEM image, and outputs the measurement report to the TEM device for display.

[0013] In this implementation, the measurement report generated by the electronic device can be transmitted to the TEM device for display, making it convenient for engineers to view.

[0014] In one possible implementation, the measurement results of the TEM image include the linear distance between any two target measurement points on the map and the angle between two vectors formed by any three target measurement points. Regarding generating a measurement report for the semiconductor device based on the measurement results, the electronic device converts the linear distance between any two target measurement points on the map into actual physical distances based on the scale of the original image data, where the TEM image is obtained from the original image data. The electronic device generates the measurement report based on the actual physical distances and angles.

[0015] In this implementation, for the measurement results of straight-line distance on the image, the electronic device can convert the original image data of the TEM image into the actual physical distance based on the scale, and thus generate a measurement report based on the actual physical distance, angle and other physical measurement results.

[0016] In one possible implementation, before acquiring a TEM image of the semiconductor device, the electronic device also receives raw image data and a scale of the raw image data sent by the TEM device.

[0017] In this implementation, the electronic device can receive raw image data and a scale from the TEM device, thereby obtaining a TEM image based on the raw image data, and calling a measurement algorithm to measure the TEM image. The scale in the raw image data helps the electronic device convert linear measurement results such as image distances into actual physical measurement results.

[0018] Secondly, embodiments of this application provide a measuring device for transmission electron microscopy (TEM) images, applied to an electronic device. The device includes an acquisition unit and a processing unit, wherein: Acquisition unit, used to acquire TEM images of semiconductor devices; The processing unit is used to classify the measurement components in the TEM image to obtain the category of the measurement components in the TEM image; determine the target measurement area of ​​the TEM image based on the category of the measurement components, the target measurement area is the area in the TEM image where the measurement components are located; measure the target measurement points in the target measurement area to obtain the measurement result of the TEM image.

[0019] In one possible implementation, the processing unit is specifically used for determining the target measurement region of the TEM image based on the category of the measurement component: The image segmentation model corresponding to the category of the measurement component is determined based on the category of the measurement component; An image segmentation model is used to perform semantic segmentation on the TEM image to obtain the mask image corresponding to the TEM image. The target measurement area is determined based on the mask image.

[0020] In one possible implementation, the processing unit is also used for: Clustering algorithms are used to cluster the points within the target measurement area to obtain point clusters; Identify the boundary contours of point clusters; Extract candidate measurement points on the boundary contour; The target measurement point is determined based on the candidate measurement points.

[0021] In one possible implementation, in determining the target measurement point based on candidate measurement points, the processing unit is specifically used for: The Otsu method is used to segment the TEM image with the goal of maximizing the inter-class variance between the foreground and the background, and the foreground point position of the TEM image is obtained. The foreground point position is the pixel point in the region where the measurement component is located in the TEM image. Multiple points that belong to the foreground points among the candidate measurement points are identified as target measurement points.

[0022] In one possible implementation, the processing unit is also used to generate a measurement report of the semiconductor device based on the measurement results from the TEM image; The acquisition unit is also used to output the measurement report to the TEM device for display.

[0023] In one possible implementation, the measurement results of the TEM image include the linear distance between any two target measurement points and the angle between two vectors formed by any three target measurement points. Specifically, in generating a measurement report for the semiconductor device based on the measurement results, the processing unit is used to: The scale based on the original image data converts the linear distance on the map between any two target measurement points into the actual physical distance. The TEM image is obtained based on the original image data. A measurement report is generated based on the actual physical distance and angle.

[0024] In one possible implementation, the acquisition unit is further configured to: receive raw image data and a scale of the raw image data sent by the TEM device.

[0025] It should be understood that since the method embodiments and the device embodiments are different presentations of the same technical concept, the content of the first aspect of the embodiments of this application should be adapted to the second aspect of the embodiments of this application simultaneously, and can achieve the same or similar beneficial effects, which will not be repeated here.

[0026] Thirdly, embodiments of this application provide an electronic device including a processor, a memory, and one or more programs. The processor is connected to the memory, and the one or more programs are stored in the memory and configured to implement the method described in the first aspect or any implementation thereof when executed by the processor.

[0027] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program for execution by an electronic device, wherein the computer program, when executed, implements the method described in the first aspect or any implementation thereof.

[0028] Fifthly, embodiments of this application provide a computer program product that, when run by a device, causes the device to execute the method described in the first aspect or any implementation thereof. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of the present invention or the background art, the accompanying drawings used in the embodiments of the present invention or the background art will be described below.

[0030] Figure 1 A schematic diagram of a system architecture provided for an embodiment of this application; Figure 2 A schematic flowchart illustrating a method for measuring transmission electron microscope (TEM) images provided in this application embodiment; Figure 3 A schematic flowchart of another method for measuring transmission electron microscopy (TEM) images provided in an embodiment of this application; Figure 4 A schematic diagram illustrating TEM image measurement based on the computing power of an AI measurement platform, as provided in this application embodiment; Figure 5 A schematic diagram of a measurement device for transmission electron microscopy (TEM) images provided in an embodiment of this application; Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0031] The terms "first," "second," "third," and "fourth," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0032] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0033] The terms “component,” “module,” “system,” etc., used in this specification are used to refer to computer-related entities, hardware, firmware, combinations of hardware and software, software, or software in execution. For example, a component can be, but is not limited to, a process running on a processor, a processor, an object, an executable file, an execution thread, a program, and / or a computer. As illustrated, an application running on a terminal device and the terminal device can both be components. One or more components may reside in a process and / or an execution thread, and components may be located on a single computer and / or distributed among two or more computers. Furthermore, these components can be executed from various computer-readable media on which various data structures are stored. Components can communicate, for example, via local and / or remote processes based on signals having one or more data packets (e.g., data from two components interacting with another component between a local system, a distributed system, and / or a network, such as the Internet interacting with other systems via signals).

[0034] To facilitate understanding of the embodiments of this application, further analyze and propose the specific technical problems to be solved by this application, the relevant technical solutions of this application are briefly introduced below.

[0035] 1y nanometer process node: 1y-nm, C1Y; a generation of process technology in a specific naming system for semiconductor device process nodes, representing a technology generation between 14nm and 16nm.

[0036] Digital Micrograph (DM), a TEM-specific software, is used for TEM image acquisition, interplanar spacing measurement (e.g., with an accuracy of 0.3472 nm), diffraction calibration, and electron energy loss spectroscopy (EELS) line analysis.

[0037] FinFET (Fin Field-Effect Transistor)

[0038] High-K Metal Gate (HKMG)

[0039] Gate-All-Around FET (GAAFET).

[0040] TEM (Transformer Electrode Modeling) is crucial in semiconductor C1Y (Computer-In-Process) manufacturing, used to acquire atomic-level characterization data of gate interfaces, FinFET sidewall morphology, photoresist structures, and defects (such as dislocations and voids). The high-resolution (≤0.1nm) images acquired can resolve key structures such as HKMG (High-Performance Magnetized Metal) stacks and GAAFET (Gas-Anaferted Electron-Foil) nanosheet stacks. Currently, TEM devices are limited to acquiring and imaging the internal structure of chips, lacking autonomous measurement and analysis capabilities. Image data must be exported to DM (Digital Measuring) software for manual processing, with an average measurement time of 30 minutes per image, resulting in low overall efficiency. Manual measurement introduces measurement errors, and the lack of standardized management of these errors leads to significant fluctuations in results for the same scenario, hindering fine-tuning of processes. Furthermore, the measurement results are unstructured, with chaotic formats, making it difficult to support subsequent in-depth analysis and resulting in a substantial loss of data value.

[0041] To address the shortcomings of existing technologies, this application provides a method for measuring transmission electron microscopy (TEM) images, which can be applied to... Figure 1 The system architecture is shown below. Figure 1As shown, the system architecture mainly consists of a user terminal layer and a cloud service layer. The user terminal layer is the front end for interaction between the system and chip engineers, and is composed of a TEM device and its associated terminal devices. The TEM device is used to acquire nanoscale images of the internal structure of semiconductor devices manufactured through a series of processes, and can output image data from processes such as diffusion (DIFF), physical vapor deposition (PVD), and chemical vapor deposition (CVD). The terminal devices are used to initiate measurement tasks and present measurement results. Chip engineers access the user interface of the Artificial Intelligence (AI) measurement platform deployed in the cloud service layer, trigger measurement commands through the user interface, and view the visualized measurement results and analysis reports after measurement processing. The cloud service layer is the processing core of the system, consisting of a central server cluster, where the AI ​​measurement platform server is deployed. The platform integrates powerful computing capabilities, dynamically allocating the computing power of the physical graphics processing unit (GPU) and the physical central processing unit (CPU) to provide high-performance computing support for its subordinate classification models, segmentation models, and measurement operators, ensuring the efficiency and stability of complex AI inference and graphics operations.

[0042] The measurement method for transmission electron microscopy (TEM) images provided in this application is described in detail below.

[0043] Please see Figure 2 , Figure 2 This is a schematic flowchart illustrating a method for measuring transmission electron microscopy (TEM) images provided in an embodiment of this application. Exemplarily, this method can be executed by an electronic device, such as a terminal device (e.g., a computer) connected to the TEM apparatus, or... Figure 1 Servers in the cloud service layer, etc. For example... Figure 2 As shown, the method may include steps 201-204: 201: Acquire TEM images of semiconductor devices.

[0044] In this embodiment, the electronic device can receive raw image data of a semiconductor device sent by a TEM device, and obtain a TEM image based on the raw image data. For example, the raw image data may be in a format such as DM3. The electronic device obtains the TEM image by converting the format, and the format conversion preserves metadata such as resolution of the original image data. For example, the electronic device can also receive acquisition parameters of the raw image sent by the TEM device, including scale, magnification, imaging conditions, etc.

[0045] For example, the TEM device can send the above data to the AI ​​measurement platform deployed in the electronic device in response to the triggering of virtual measurement commands from physical measurement buttons or supporting terminal devices (such as touch screens or dedicated industrial tablets).

[0046] 202: Classify the measurement components of the TEM image to obtain the category of the measurement components in the TEM image.

[0047] In this embodiment, the electronic device can call a measurement component classification model to classify the measurement components in the TEM image. Here, a measurement component refers to a component in a semiconductor device that needs to be measured, i.e., various microstructural units on the semiconductor device whose size, morphology, thickness, and other parameters can be measured, such as gate structures, interconnects, and dielectric layers in the semiconductor device.

[0048] For example, the measurement component classification model is pre-trained. Its feature extraction layer (including convolution and pooling operations) can extract the internal structural features of semiconductor devices in the TEM image. Then, the fully connected layer classifies the measurement components in the TEM image based on the internal structural features to obtain the category of the measurement component. For example, the fully connected layer can predict the probability of the TEM image in multiple preset measurement component categories, and take the category with the highest probability as the category of the measurement component in the current TEM image.

[0049] 203: Determine the target measurement area of ​​the TEM image based on the category of the measurement component.

[0050] In this embodiment, the electronic device can obtain the correspondence between the categories of each measurement component and the image segmentation model, thereby determining the image segmentation model corresponding to the category of the measurement component. For example, the gate structure corresponds to image segmentation model 1, the interconnect line corresponds to image segmentation model 2, the dielectric layer corresponds to image segmentation model 3, and so on. For example, the correspondence between the categories of each measurement component and the image segmentation model can be shown in Table 1: Table 1

[0051] It should be noted that Table 1 is only an example and does not limit the correspondence between the categories of measurement components and image segmentation models. For example, this correspondence can also be shown in the documentation.

[0052] After determining the image segmentation model, the electronic device can use the image segmentation model to perform semantic segmentation on the TEM image to obtain a mask map corresponding to the TEM image. Since this mask map includes a mask of the region where the measurement component is located in the TEM image, the target measurement region of the TEM image, i.e., the region where the measurement component is located in the TEM image, can be determined based on this mask map.

[0053] For example, an electronic device can preprocess TEM images by enhancing contrast and suppressing noise, and then call an image segmentation model to extract features and detect boundaries, ultimately generating a semantic segmentation mask (containing pixel-level annotations of the area where the measurement component is located and the background area).

[0054] For example, the image segmentation model can be trained using the deeplabV3 model architecture. The deeplabV3 model architecture uses ResNet-50 as the backbone network. The backbone network extracts features from the TEM image through three convolutional stages to obtain features at multiple scales. In the last residual block, dilated convolutional layers with different dilation rates (e.g., 1, 6, 12, 18) are used to capture information at different scales, thereby expanding the receptive field without sacrificing resolution. After feature extraction, the Atrous Spatial Pyramid Pooling (ASSP) network fuses the multi-scale features. It captures contextual information of different ranges through a combination of 1×1 convolutions and 3×3 dilated convolutions. Then, global average pooling, interpolation, and other operations are used to restore the feature size. Finally, a softmax classifier is used to achieve pixel-level category determination, such as identifying measurement components, substrates, and impurities, thereby obtaining the region where the measurement component is located.

[0055] In this implementation, the electronic device can determine the image segmentation model suitable for the current measurement component based on the correspondence between the measurement component category and the image segmentation model. Using this image segmentation model to perform semantic segmentation on the TEM image helps to improve the accuracy of image segmentation, thereby improving the accuracy of determining the target measurement region.

[0056] For example, after determining the target measurement region of the TEM image based on the category of the measurement component, the method further includes: Clustering algorithms are used to cluster the points within the target measurement area to obtain point clusters; Identify the boundary contours of point clusters; Extract candidate measurement points on the boundary contour; The target measurement point is determined based on the candidate measurement points.

[0057] In this embodiment, the DBSCAN clustering algorithm can be used to cluster the points within the target measurement area into multiple clusters. For example, a normal area may form one cluster, while a defective area may form another. Furthermore, different material layers may form different clusters. Within a cluster, the point features are generally similar, and therefore, their boundary contours are usually continuous. On these continuous boundary contours, the electronic device can use edge detection operators to extract key points on the contour as candidate measurement points. These candidate measurement points can be directly used as the final target measurement points, or these candidate measurement points can be further filtered to obtain the target measurement points.

[0058] For example, the edge detection operator can be the Canny edge detection operator. The Canny edge detection operator has a corresponding first preset threshold and a second preset threshold. The first preset threshold is greater than the second preset threshold. The electronic device can use points on the boundary contour with gray values ​​greater than the first preset threshold as candidate measurement points.

[0059] In this implementation, the electronic device obtains continuous boundary contours within the target measurement area through a clustering algorithm, extracts candidate measurement points with obvious features from these boundary contours, and finally determines the target measurement point based on the candidate measurement points. This is beneficial for screening out target measurement points with obvious boundary structure features of the measurement components.

[0060] For example, when screening candidate measurement points, the electronic device can use the Otsu method to segment the TEM image with the goal of maximizing the inter-class variance between the foreground and the background, and obtain the foreground points of the TEM image. The foreground points are the pixels in the region where the measurement component is located in the TEM image. Then, multiple points belonging to the foreground points among the candidate measurement points are determined as the target measurement points.

[0061] In this implementation, the electronic device can segment the foreground points of the measurement component using the OTSU threshold segmentation algorithm, and determine the points that belong to the foreground points among the candidate measurement points as the final target measurement points, which is beneficial for selecting more representative target measurement points.

[0062] 204: Measure the target measurement points within the target measurement area to obtain the measurement results of the TEM image.

[0063] In this embodiment, for any two target measurement points among multiple target measurement points, the electronic device can calculate their straight-line distance on the image, for example, by calculating the Euclidean distance based on their pixel coordinates. This straight-line distance on the image can be a quantity representing width, thickness, or length. For any three target measurement points among multiple target measurement points, the electronic device can construct two vectors based on their pixel coordinates and calculate their angle using the vector angle formula. That is, the measurement result of the TEM image includes the straight-line distance between any two target measurement points and the angle between the two vectors formed by any three target measurement points.

[0064] As can be seen, in this embodiment, the electronic device acquires TEM images of semiconductor devices, classifies the TEM images into measurement components, determines the target measurement area where the measurement components are located in the TEM image based on the category of the measurement components, and finally measures the target measurement points in the target measurement area to obtain the measurement results. Compared with manual measurement using DM software, this method improves the measurement efficiency of TEM images, thereby meeting the analysis needs of TEM images in semiconductor manufacturing processes. Furthermore, the electronic device can perform measurement processing on TEM images through an AI measurement platform, which can avoid the inconsistencies in standards and measurement errors caused by manual measurement, thus improving measurement accuracy.

[0065] Please see Figure 3 , Figure 3 This is a schematic flowchart illustrating another method for measuring transmission electron microscopy (TEM) images provided in an embodiment of this application. Figure 3 As shown, the method may include steps 301-306: 301: Acquire TEM images of semiconductor devices.

[0066] 302: Classify the measurement components of the TEM image to obtain the category of the measurement components in the TEM image.

[0067] 303: Determine the target measurement area of ​​the TEM image based on the category of the measurement component.

[0068] 304: Measure the target measurement points within the target measurement area to obtain the measurement results of the TEM image.

[0069] Steps 301-304 can be referred to Figure 2 The corresponding descriptions in the illustrated embodiments can achieve the same or similar beneficial effects.

[0070] 305: Generate measurement reports for semiconductor devices based on measurement results from TEM images.

[0071] In this embodiment of the application, the electronic device can convert the linear distance on the map between any two target measurement points into the actual physical distance based on the scale of the original image data, and then generate a measurement report based on the actual physical distance and angle.

[0072] For example, since actual physical distances and angles are actual physical quantities, electronic devices can compare them with standard distances and angles in a pre-defined chip standard sampling database (containing standard values ​​and tolerance ranges for different process nodes). Statistical methods (such as mean deviation and standard deviation) are used to calculate the difference between the measured values ​​and the standard values, marking abnormal parameters that exceed the tolerance range. Based on the abnormal parameters and the associated process knowledge base, a comprehensive analysis report is generated, including measurement data tables, deviation trend charts, and hypothetical causes of the anomalies. Suggestions for adjusting process parameters can also be generated for abnormal data. Even in cases without anomalies, the report can provide comparison data and charts between the measured physical quantities and the standard values.

[0073] In this implementation, for the measurement results of straight-line distance on the image, the electronic device can convert the original image data of the TEM image into the actual physical distance based on the scale, and thus generate a measurement report based on the actual physical distance, angle and other physical measurement results.

[0074] 306: Output the measurement report to the TEM device for display.

[0075] In this embodiment, the electronic device can send measurement reports to the TEM device for display on its control screen, or to a terminal device connected to the TEM device, providing decision-making reference for chip engineers. Chip engineers can receive measurement results and analysis reports in real time, resulting in high timeliness throughout the process.

[0076] The measurement reports output by electronic devices can display multi-dimensional data and support structured data formats such as Excel, XML, YAML, and Lua, eliminating the need for secondary processing and resulting in high data utilization.

[0077] Please see Figure 4Based on the computing power of the AI ​​measurement platform, for multiple raw image data sent from the TEM side, the electronic device can acquire the TEM image corresponding to each raw image data. It then calls a measurement component classification model to classify the measurement components in each TEM image, obtaining the category of the measurement component corresponding to each TEM image. Based on the category of the measurement component corresponding to each TEM image, it calls an image segmentation model to obtain the target measurement region for each TEM image. Then, it uses measurement operators to determine the target measurement points within the target measurement region and performs the measurement. Finally, it generates a measurement report for each TEM image through a result analysis operator and returns it to the TEM device. Therefore, the transmission electron microscope (TEM) image measurement method provided in this application embodiment can meet the growing analytical needs in semiconductor device manufacturing processes.

[0078] The methods of the embodiments of this application have been described in detail above, and the apparatus of the embodiments of this application is provided below.

[0079] Please see Figure 5 , Figure 5 This is a schematic diagram of a measurement device for transmission electron microscopy (TEM) images provided in an embodiment of this application. The device is used in an electronic device and includes an acquisition unit 501 and a processing unit 502; wherein: Acquisition unit 501 is used to acquire TEM images of semiconductor devices; The processing unit 502 is used to classify the measurement components of the TEM image to obtain the category of the measurement components in the TEM image; determine the target measurement area of ​​the TEM image based on the category of the measurement components, the target measurement area is the area where the measurement components are located in the TEM image; measure the target measurement points within the target measurement area to obtain the measurement result of the TEM image.

[0080] In one possible implementation, in determining the target measurement region of the TEM image based on the category of the measurement component, the processing unit 502 is specifically configured to: The image segmentation model corresponding to the category of the measurement component is determined based on the category of the measurement component; An image segmentation model is used to perform semantic segmentation on the TEM image to obtain the mask image corresponding to the TEM image. The target measurement area is determined based on the mask image.

[0081] In one possible implementation, the processing unit 502 is further configured to: Clustering algorithms are used to cluster the points within the target measurement area to obtain point clusters; Identify the boundary contours of point clusters; Extract candidate measurement points on the boundary contour; The target measurement point is determined based on the candidate measurement points.

[0082] In one possible implementation, in determining the target measurement point based on candidate measurement points, the processing unit 502 is specifically used for: The Otsu method is used to segment the TEM image with the goal of maximizing the inter-class variance between the foreground and the background, and the foreground point position of the TEM image is obtained. The foreground point position is the pixel point in the region where the measurement component is located in the TEM image. Multiple points that belong to the foreground points among the candidate measurement points are identified as target measurement points.

[0083] In one possible implementation, the processing unit 502 is also used to generate a measurement report of the semiconductor device based on the measurement results of the TEM image; The acquisition unit 501 is also used to output the measurement report to the TEM device for display.

[0084] In one possible implementation, the measurement results of the TEM image include the linear distance between any two target measurement points and the angle between two vectors formed by any three target measurement points. Specifically, in generating a measurement report for the semiconductor device based on the measurement results, the processing unit 502 is used to: The scale based on the original image data converts the linear distance on the map between any two target measurement points into the actual physical distance. The TEM image is obtained based on the original image data. A measurement report is generated based on the actual physical distance and angle.

[0085] In one possible implementation, the acquisition unit 501 is further configured to: receive raw image data and a scale of the raw image data sent by the TEM device.

[0086] For example, Figure 5 The various units in the illustrated device can be individually or entirely combined into one or more other units, or some of the units can be further divided into multiple functionally smaller units. This achieves the same operation without affecting the technical effects of the embodiments of this application. The above units are based on logical function division. In practical applications, the function of one unit can also be implemented by multiple units, or the function of multiple units can be implemented by one unit. In other examples of this application, Figure 5 The device shown may also include other units. In practical applications, these functions can also be implemented with the assistance of other units, and multiple units can work together to achieve them.

[0087] For example, it is possible to perform such actions by running on a general-purpose electronic device, such as a computer, which includes processing elements and storage elements such as a central processing unit (CPU), random access memory (RAM), and read-only memory (ROM). Figure 2 or Figure 3 The computer program (including program code) for each step involved in the corresponding method shown, to implement as follows Figure 5 The diagram illustrates the functions of each unit in the apparatus and the method for measuring transmission electron microscopy (TEM) images to achieve the embodiments of this application. The computer program can be recorded on, for example, a computer-readable recording medium, loaded onto the aforementioned electronic device via such a medium, and run therein.

[0088] Please see Figure 6 , Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device includes at least a memory 601, a processor 602, a communication interface 603, and a bus 604. For example, the electronic device can be a computer, server, chip, cloud server, terminal device, etc. The memory 601, processor 602, and communication interface 603 are interconnected via the bus 604.

[0089] The memory 601 includes, but is not limited to, RAM, ROM, erasable programmable read-only memory (EPROM), or compact disc read-only memory (CD-ROM), and is used to store related computer programs and data.

[0090] The processor 602 may include one or more processing chips. When the processor 602 includes a single processing chip, the processor 602 may be a single-core processing chip or a multi-core processing chip.

[0091] The processor 602 in the electronic device is used to read one or more programs stored in the memory 601 and perform the following operations: Acquire TEM images of semiconductor devices; The TEM image is classified according to its measurement components to obtain the category of the measurement components in the TEM image; The target measurement region of the TEM image is determined based on the category of the measurement component, and the target measurement region is the area in the TEM image where the measurement component is located; The target measurement points within the target measurement area are measured to obtain the measurement results of the TEM image.

[0092] The communication interface 603 is used in conjunction with the processor 602 to perform operations such as sending and receiving data.

[0093] It should be noted that the implementation of each operation can also be referenced accordingly. Figure 2 or Figure 3 The corresponding description of the method embodiments shown.

[0094] This application also provides a computer-readable storage medium, which is a memory device in an electronic device for storing a computer program executed by the device. When the program is run on the electronic device, Figure 2 or Figure 3 The method and flow shown are thus implemented.

[0095] It is understood that the computer-readable storage medium referred to herein can include both the built-in storage medium in an electronic device and, of course, the extended storage medium supported by the electronic device. The computer-readable storage medium provides storage space containing the operating system of the electronic device. Furthermore, this storage space also contains one or more computer programs suitable for loading and execution by a processor. It should be noted that the computer-readable storage medium referred to herein can be high-speed RAM or non-volatile memory, such as at least one disk storage device; optionally, it can also be at least one computer-readable storage medium located remotely from the aforementioned processor.

[0096] This application also provides a computer program product, which, when run by a device,... Figure 2 or Figure 3 The method and flow shown are thus implemented.

[0097] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0098] It should be understood that the processor mentioned in the embodiments of this application can be a CPU, a BMC, or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0099] It should also be understood that the memory mentioned in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. Non-volatile memory can be ROM, Programmable Read-Only Memory (PROM), EPROM, Electrically Erasable Programmable Read-Only Memory (EEPROM), or flash memory. Volatile memory can be RAM, which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate Synchronous DRAM (DDR SDRAM), Enhanced Synchronous DRAM (ESDRAM), Synchlink DRAM (SLDRAM), and Direct Rambus RAM (DR RAM).

[0100] It should be noted that when the processor is a general-purpose processor, DSP, ASIC, FPGA, or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component, the memory (storage module) is integrated into the processor.

[0101] It should be noted that the memories described herein are intended to include, but are not limited to, these and any other suitable types of memories.

[0102] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0103] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely exemplary. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0104] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0105] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium.

[0106] In this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. In the textual description of this application, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0107] The steps in the method of this application embodiment can be adjusted, combined, or deleted according to actual needs.

[0108] The modules in the device of this application embodiment can be merged, divided, and deleted according to actual needs.

[0109] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for measuring transmission electron microscopy (TEM) images, characterized in that, The method includes: Acquire TEM images of semiconductor devices; The TEM image is classified according to its measurement components to obtain the category of the measurement components in the TEM image; The target measurement region of the TEM image is determined based on the category of the measurement component, and the target measurement region is the area in the TEM image where the measurement component is located; The target measurement points within the target measurement area are measured to obtain the measurement results of the TEM image.

2. The method according to claim 1, characterized in that, Determining the target measurement region of the TEM image based on the category of the measurement component includes: Based on the category of the measurement component, determine the image segmentation model corresponding to the category of the measurement component; The image segmentation model described above is used to perform semantic segmentation on the TEM image to obtain a mask image corresponding to the TEM image. The target measurement area is determined based on the mask image.

3. The method according to claim 1 or 2, characterized in that, After determining the target measurement region of the TEM image based on the category of the measurement component, the method further includes: Clustering algorithms are used to cluster the points within the target measurement area to obtain point clusters; Identify the boundary contours of the point clusters; Extract candidate measurement points on the boundary contour; The target measurement point is determined based on the candidate measurement points.

4. The method according to claim 3, characterized in that, Determining the target measurement point based on the candidate measurement points includes: The Otsu method is used to segment the TEM image with the goal of maximizing the inter-class variance between the foreground and the background, thereby obtaining the foreground point position of the TEM image. The foreground point position is the pixel point in the region where the measurement component is located in the TEM image. Multiple points belonging to the foreground points among the candidate measurement points are determined as the target measurement points.

5. The method according to any one of claims 1-4, characterized in that, After measuring the target measurement points within the target measurement area to obtain the measurement results of the TEM image, the method further includes: A measurement report for the semiconductor device is generated based on the measurement results from the TEM image. The measurement report is output to the TEM device for display.

6. The method according to claim 5, characterized in that, The measurement results of the TEM image include the straight-line distance between any two target measurement points and the angle between two vectors formed by any three target measurement points. The generation of a measurement report for the semiconductor device based on the measurement results includes: The scale based on the original image data converts the linear distance on the map between any two target measurement points into the actual physical distance, and the TEM image is obtained based on the original image data; The measurement report is generated based on the actual physical distance and the angle.

7. The method according to claim 6, characterized in that, Before acquiring a TEM image of the semiconductor device, the method further includes: Receive the raw image data and the scale of the raw image data sent by the TEM device.

8. A measuring device for transmission electron microscopy (TEM) images, characterized in that, It includes an acquisition unit and a processing unit, wherein: The acquisition unit is used to acquire TEM images of semiconductor devices; The processing unit is configured to classify the measurement components of the TEM image to obtain the category of the measurement components in the TEM image; determine the target measurement region of the TEM image based on the category of the measurement components, wherein the target measurement region is the region in the TEM image where the measurement components are located; and measure the target measurement points within the target measurement region to obtain the measurement result of the TEM image.

9. The apparatus according to claim 8, characterized in that, In determining the target measurement region of the TEM image based on the category of the measurement component, the processing unit is specifically configured to: Based on the category of the measurement component, determine the image segmentation model corresponding to the category of the measurement component; The image segmentation model described above is used to perform semantic segmentation on the TEM image to obtain a mask image corresponding to the TEM image. The target measurement area is determined based on the mask image.

10. The apparatus according to claim 8 or 9, characterized in that, The processing unit is also used for: Clustering algorithms are used to cluster the points within the target measurement area to obtain point clusters; Identify the boundary contours of the point clusters; Extract candidate measurement points on the boundary contour; The target measurement point is determined based on the candidate measurement points.

11. The apparatus according to claim 10, characterized in that, In determining the target measurement point based on the candidate measurement points, the processing unit is specifically used for: The Otsu method is used to segment the TEM image with the goal of maximizing the inter-class variance between the foreground and the background, thereby obtaining the foreground point position of the TEM image. The foreground point position is the pixel point in the region where the measurement component is located in the TEM image. Multiple points belonging to the foreground points among the candidate measurement points are determined as the target measurement points.

12. The apparatus according to any one of claims 8-10, characterized in that, The processing unit is also configured to generate a measurement report of the semiconductor device based on the measurement results of the TEM image; The acquisition unit is also used to output the measurement report to the TEM device for display.

13. The apparatus according to claim 12, characterized in that, The measurement results of the TEM image include the linear distance between any two target measurement points and the angle between two vectors formed by any three target measurement points. In generating a measurement report for the semiconductor device based on the measurement results, the processing unit is specifically used for: The scale based on the original image data converts the linear distance on the map between any two target measurement points into the actual physical distance, and the TEM image is obtained based on the original image data; The measurement report is generated based on the actual physical distance and the angle.

14. The apparatus according to claim 13, characterized in that, The acquisition unit is further configured to: receive the original image data and the scale of the original image data sent by the TEM device.

15. An electronic device, characterized in that, It includes a processor, a memory, and one or more programs, the processor being connected to the memory, the one or more programs being stored in the memory, and being configured to implement the method as described in any one of claims 1-7 when executed by the processor.

16. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program for execution by an electronic device, which, when executed, implements the method of any one of claims 1-7.

Images