A scanning electron microscope focusing method, apparatus, device and medium

Abstract

The application relates to the technical field of electron microscopes, and discloses a scanning electron microscope focusing method, device, equipment and medium. The method comprises the following steps: collecting a sample image of a sample to be measured by a scanning electron microscope; calculating the average energy of secondary electrons of the sample image according to the gray value of the sample image; determining the voltage value of an electrostatic accelerating tube based on the average energy of the secondary electrons and the electronic charge quantity of the electrostatic accelerating tube; and configuring the electrostatic accelerating tube of the scanning electron microscope based on the voltage value of the electrostatic accelerating tube, so that the scanning electron microscope outputs a first focused image. The application improves the focusing accuracy and focusing efficiency of the scanning electron microscope.

Description

Technical Field

[0001] This invention relates to the field of electron microscopy, and more specifically to a focusing method, apparatus, device, and medium for scanning electron microscopy. Background Technology

[0002] Scanning electron microscopy (SEM) plays a crucial role in numerous fields such as materials science, biology, and nanotechnology. It provides information on the surface morphology and composition of samples at the microscopic scale, helping researchers gain a deeper understanding of the structure-property relationship of materials. It is indispensable in new material development, biological tissue research, and the fabrication of micro- and nano-devices. In traditional field emission scanning electron microscopy (FETS), image sharpness is related to the voltage value of the electrostatic accelerator tube (ETU). Setting an appropriate ETU voltage value is one of the key factors in achieving proper focusing. Currently, the conventional method for setting the ETU voltage value is for the operator to pre-set a relatively fixed value based on experience. During imaging, if the image quality is poor, the operator gradually adjusts the ETU voltage value based on experience, observing the image changes until a relatively satisfactory image is obtained. However, setting the ETU voltage value based on experience often leads to large errors and poor focusing, requiring multiple fine-tuning steps by the user to achieve proper focusing, thus reducing the focusing efficiency of the scanning electron microscope. Summary of the Invention

[0003] In view of this, the present invention provides a focusing method, apparatus, device and medium for scanning electron microscopes to solve the problem of slow focusing speed caused by poor setting of electrostatic accelerating tube voltage in scanning electron microscopes.

[0004] In a first aspect, the present invention provides a focusing method for a scanning electron microscope, the method comprising: acquiring a sample image of a sample to be tested using a scanning electron microscope; calculating the average secondary electron energy of the sample image based on the grayscale value of the sample image; determining the voltage value of an electrostatic accelerator tube based on the average secondary electron energy and the electron charge of the electrostatic accelerator tube; and configuring the electrostatic accelerator tube of the scanning electron microscope based on the voltage value of the electrostatic accelerator tube, so that the scanning electron microscope outputs a first focused image.

[0005] In some optional embodiments, acquiring the sample image of the sample to be tested using a scanning electron microscope includes: acquiring a secondary electron image and a backscattered electron image of the sample to be tested using the scanning electron microscope; preprocessing and fusing the secondary electron image and the backscattered electron image to obtain the sample image.

[0006] In some optional implementations, calculating the average secondary electron energy of the sample image based on the grayscale value of the sample image includes: determining a region of interest in the sample image; extracting key features of the region of interest; calculating the average grayscale value of the region of interest; obtaining an energy prediction model, which is obtained by training a neural network based on known secondary electron energy labels, key feature samples, and grayscale value samples; and inputting the key features and the average grayscale value into the energy prediction model to obtain the average secondary electron energy of the region of interest.

[0007] In some optional implementations, the average gray value of the region of interest is calculated, including: determining a seed position for image segmentation based on the key features; performing a region growing algorithm on the region of interest according to the seed position to subdivide the region of interest into several sub-regions of interest; and calculating the average gray value corresponding to each sub-region of interest.

[0008] In some optional implementations, inputting the key features and the average gray value into the energy prediction model to obtain the average secondary electron energy of the region of interest includes: inputting the key features and average gray value included in each sub-region of interest into the corresponding decision tree to obtain the secondary electron energy predicted by each decision tree, wherein the decision tree is the energy prediction model trained for each sub-region of interest; calculating the average value of the secondary electron energy predicted by each decision tree to obtain the average secondary electron energy.

[0009] or,

[0010] The key features and average gray values ​​of each sub-region of interest are input into the corresponding decision tree to obtain the secondary electron energy predicted by each decision tree; the weight parameters corresponding to each decision tree are obtained, wherein the sum of the weight parameters corresponding to each decision tree is 1; the secondary electron energy predicted by each decision tree is weighted and summed based on the weight parameters corresponding to each decision tree to obtain the average secondary electron energy.

[0011] In some optional embodiments, the method further includes: acquiring the secondary electron brightness value and backscattered electron brightness of the region of interest in the first focused image; determining the brightness difference between the secondary electron brightness value and the backscattered electron brightness; and adjusting the electrostatic accelerator tube voltage value of the scanning electron microscope according to the brightness difference, so that the scanning electron microscope outputs a second focused image.

[0012] In some optional embodiments, the method further includes: acquiring a backscattered electron image after processing the first focused image; extracting the surface roughness and grain size of the region of interest in the processed backscattered electron image; if the surface roughness and the grain size are not within the corresponding preset threshold range, adjusting the electrostatic accelerator tube voltage value of the scanning electron microscope according to the preset threshold corresponding to the surface roughness and the grain size, so that the surface roughness and the grain size fall within the corresponding preset threshold range, and outputting a third focused image.

[0013] Secondly, the present invention provides a focusing device for a scanning electron microscope, the device comprising: an initial image acquisition module for acquiring a sample image of a sample to be tested using a scanning electron microscope; an electron energy prediction module for calculating the average secondary electron energy of the sample image based on the grayscale value of the sample image; a voltage calibration module for determining the voltage value of the electrostatic accelerator tube based on the average secondary electron energy and the electron charge of the electrostatic accelerator tube; and a microscope adjustment module for configuring the electrostatic accelerator tube of the scanning electron microscope based on the voltage value of the electrostatic accelerator tube, so that the scanning electron microscope outputs a first focused image.

[0014] Thirdly, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the method described in the first aspect or any corresponding embodiment thereof.

[0015] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to perform the method described in the first aspect or any corresponding embodiment thereof.

[0016] Fifthly, the present invention provides a computer program product, including computer instructions for causing a computer to perform the method described in the first aspect or any corresponding embodiment thereof.

[0017] The technical solution provided by this invention has the following advantages:

[0018] Before focusing, this invention first acquires a sample image of the sample to be tested using a scanning electron microscope. Whether the sample image is focused or not is not required. In this embodiment, the average secondary electron energy of the sample image is calculated based on the grayscale value of the initial sample image, thereby determining the magnitude and distribution of the secondary electron energy of the sample image. Then, the voltage value of the electrostatic accelerator tube is calculated and determined based on the average secondary electron energy and the electron charge of the electrostatic accelerator tube, providing a reliable basis for setting the electrostatic accelerator tube voltage for the electron microscope scanning image. The electrostatic accelerator tube of the scanning electron microscope is configured according to the electrostatic accelerator tube voltage value so that the scanning electron microscope outputs the first focused image, which can significantly improve the focusing effect of the scanning electron microscope. Attached Figure Description

[0019] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0020] Figure 1 This is a schematic flowchart of a scanning electron microscope focusing method according to an embodiment of the present invention;

[0021] Figure 2 This is a schematic diagram of the process for acquiring sample images of the sample to be tested and calculating the average energy of secondary electrons according to an embodiment of the present invention;

[0022] Figure 3 This is a schematic diagram of the process for subdividing the region of interest according to an embodiment of the present invention;

[0023] Figure 4 This is a flowchart illustrating the further optimization of the focused image according to an embodiment of the present invention;

[0024] Figure 5 This is another schematic flowchart of a scanning electron microscope focusing method according to an embodiment of the present invention;

[0025] Figure 6 This is a schematic diagram of the focusing device of a scanning electron microscope according to an embodiment of the present invention;

[0026] Figure 7 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] According to an embodiment of the present invention, a scanning electron microscope focusing method embodiment is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0029] This embodiment provides a focusing method for scanning electron microscopy. Figure 1 This is a flowchart of a scanning electron microscope focusing method according to an embodiment of the present invention, the process including the following steps:

[0030] Step S101: Acquire sample images of the sample to be tested using a scanning electron microscope;

[0031] Step S102: Calculate the average secondary electron energy of the sample image based on the grayscale value of the sample image.

[0032] Step S103: Determine the voltage value of the electrostatic accelerator tube based on the average energy of the secondary electrons and the electron charge of the electrostatic accelerator tube.

[0033] Step S104: Configure the electrostatic accelerator tube of the scanning electron microscope based on the voltage value of the electrostatic accelerator tube, so that the scanning electron microscope outputs a first focused image.

[0034] Specifically, parameters affecting the focusing of field emission scanning electron microscope (FET) images include condenser voltage, objective voltage, and electrostatic accelerator tube (EAC) voltage adjustment. In this embodiment, it is assumed that other parameters such as condenser voltage and objective voltage have been adjusted. Regarding the influence of EAC voltage on focusing effect, the configuration and adjustment method of EAC voltage are improved to enhance the focusing accuracy and efficiency of FET images. Specifically, in this embodiment, the energy E gained by secondary electrons in the EAC tube is determined to have the following relationship with the EAC voltage U: E = eU, where e is the electron charge. This indicates that the energy of secondary electrons is directly proportional to the voltage of the EAC tube, which is the theoretical basis for setting and adjusting the EAC voltage using average energy. This embodiment further considers that if the energy of secondary electrons in the scanned image of the sample can be accurately predicted, a more reliable and accurate basis for adjusting the EAC voltage can be provided.

[0035] Because image brightness is closely related to the energy of secondary electrons, higher secondary electron energy usually produces a brighter image, and grayscale values ​​can reflect image brightness. Therefore, there is a close relationship between grayscale values, brightness, and secondary electron energy. Based on this relationship, embodiments of the present invention can transform the problem of adjusting the voltage of the electrostatic accelerator tube in a microscope into a problem of predicting the secondary electron energy based on grayscale values.

[0036] The first step involves acquiring an initial image of the sample to be tested using a scanning electron microscope (SEM). This initial image is used solely for calculating the secondary electron energy and does not require focusing. Next, the grayscale values ​​of the sample image are statistically analyzed. Then, an algorithm is used to predict the average energy of the secondary electrons in the sample image based on these grayscale values. For example, a machine learning algorithm, such as a neural network model, can be used to determine the relationship between grayscale values ​​and secondary electron energy. By inputting the grayscale values ​​into the trained model, the prediction of the secondary electron energy can be achieved. In this embodiment of the invention, considering that the brightness and grayscale values ​​differ at different locations in the sample image, resulting in different secondary electron energies, the average secondary electron energy of the sample image is calculated based on the average secondary electron energy at different locations. This provides a more accurate and reliable basis for subsequent voltage adjustment of the electrostatic accelerator tube, enabling the SEM to output a first focused image with better focusing effect, thus significantly improving the focusing performance of the SEM.

[0037] In some optional implementations, step S101 above includes:

[0038] Step a1: Acquire secondary electron images and backscattered electron images of the sample under test using a scanning electron microscope;

[0039] Step a2 involves preprocessing and fusing the secondary electron image and the backscattered electron image to obtain the sample image.

[0040] Specifically, this embodiment of the invention acquires secondary electron (SE) images and backscattered electron (BSE) images of the sample under test using a scanning electron microscope (SEM). The secondary electron image is generated by the SEM using secondary electrons produced by the interaction between the sample surface and the electron beam scanning the sample surface. The secondary electron image primarily reflects the morphological features of the sample surface within approximately 10 nm. The image contrast is morphological contrast, which is mainly determined by the tilt angle of the sample surface relative to the incident electron beam. The backscattered electron image is generated in the SEM by electrons generated by the electron gun, which, after passing through an accelerating electric field and a deflecting magnetic field, irradiate the sample surface. The sample reflects some of these electrons, and the backscattered electron probe in the working cavity of the SEM detects these reflected electrons, forming an image on the detector. The backscattered electron image uses the signal generated by the interaction between backscattered electrons and the sample to reflect the composition and structural information of the sample.

[0041] like Figure 2 As shown, this embodiment of the invention removes noise from the acquired image using an image preprocessing algorithm, ensuring that background noise in the original secondary electron image is minimized, thus guaranteeing the accuracy of grayscale data analysis. Then, the secondary electron image and the backscattered electron image are fused, followed by further denoising, and image enhancement techniques are used to improve image visibility. The secondary electron image and the backscattered electron image are merged into a single multi-channel image, which is then input into a subsequent deep learning network for learning, enabling the prediction of secondary electron energy. This embodiment combines morphological and compositional information by fusing the secondary electron image and the backscattered electron image, allowing for the simultaneous acquisition of both morphological and compositional information of the sample surface during image analysis. This facilitates a more comprehensive and accurate understanding of the sample's characteristics. Because complex samples may exhibit both morphological differences and compositional variations, the fused sample image offers significant advantages for studying complex samples.

[0042] In some optional implementations, step S102 above includes:

[0043] Step b1: Determine the region of interest in the sample image;

[0044] Step b2: Extract key features of the region of interest;

[0045] Step b3: Calculate the average gray value of the region of interest;

[0046] Step b4: Obtain the energy prediction model. The energy prediction model is obtained by training a neural network based on known secondary electron energy labels, key feature samples, and gray value samples.

[0047] Step b5: Input the key features and average gray value into the energy prediction model to obtain the average secondary electron energy of the region of interest.

[0048] Specifically, in determining the average secondary electron energy of a sample image, the embodiments of the present invention first determine the region of interest (ROI) in the sample image. The ROI is used to select the specific location of the sample to be tested in the image, reducing interference from other irrelevant locations; on the other hand, it improves the accuracy of analysis and facilitates feature extraction and model training.

[0049] Subsequently, this embodiment extracts key features of the region of interest using a feature extraction model. In some optional implementations, widely used image feature extraction models such as convolutional networks and backbone networks can be used to extract key features of the region of interest. The extracted key features include, but are not limited to, features related to shape, features related to composition, features related to the overall structure of the image, and features related to image grayscale. This embodiment of the invention is merely an example and is not intended to be limiting.

[0050] Simultaneously, the average gray value of the region of interest is calculated. Then, both the average gray value and the key features of the region of interest are used as prediction input data and fed into the energy prediction model. The trained energy prediction model predicts the average secondary electron energy of the region of interest based on the input data.

[0051] The technical solution provided by this invention not only considers the influence of grayscale value on secondary electron energy, but also introduces the influence of key features such as texture and morphology on secondary electron energy into the prediction model. This fully utilizes key features such as grayscale value, texture, and morphology to comprehensively predict the average secondary electron energy of the region of interest, further improving the prediction accuracy of the average secondary electron energy of the region of interest.

[0052] In some alternative implementations, step b3 above includes:

[0053] Step c1: Determine the seed location for image segmentation based on key features;

[0054] Step c2: Based on the seed position, the region of interest is subdivided into several sub-regions of interest by performing a region growing algorithm.

[0055] Step c3: Calculate the average gray value corresponding to each sub-region of interest.

[0056] Specifically, such as Figure 3 As shown, this embodiment of the invention uses an image segmentation network trained with a deep learning model and employs a region growing algorithm to determine seed locations for image segmentation based on key features in the image. For example, key features such as morphology, composition, structure, and grayscale reflect image structure and semantic information, determining which locations within the region of interest (ROI) can serve as initial seeds. For instance, suitable points are selected as initial seeds based on key feature information such as the boundaries of different regions in the image and the center of specific shapes. These selected seed locations become the starting points for subsequent region growing algorithms, used for further image segmentation and other processing, thereby subdividing and expanding the ROI, dividing a large ROI into several sub-ROIs. After subdividing the ROI, key features and average grayscale values ​​within each sub-ROI can be identified and extracted individually, making local features clearer and average grayscale value calculation more accurate, thus improving the accuracy of subsequent predictions of the average energy of secondary electrons.

[0057] Furthermore, in this embodiment of the invention, the process of subdividing the region of interest (ROI) based on seed points using a region growing algorithm can dynamically adjust the growth threshold according to the grayscale value of each pixel and other features of the multi-channel image. Based on this, the region growing algorithm determines whether to continue expanding and subdividing according to the local features of the image, thereby improving the accuracy of the division of each ROI sub-region. In this embodiment, the growth threshold is dynamically adjusted according to the grayscale value distribution of the grown region. For example, if the grayscale value of the grown sub-region changes little, i.e., the grayscale distribution is relatively uniform, it indicates that the image features in that sub-region are relatively simple. In this case, the growth threshold can be lowered to more precisely include pixels with similar grayscale values ​​in the grown ROI, allowing the grown sub-region to more completely contain this uniform feature. Conversely, if the grayscale value of the grown sub-region changes significantly, it indicates that the image may have multiple features or edges in that sub-region. To avoid over-growing into other unrelated sub-regions, the growth threshold of the region growing algorithm needs to be increased.

[0058] In some alternative implementations, step b5 above includes:

[0059] Step d1: Input the key features and average gray value of each sub-region of interest into the corresponding decision tree to obtain the secondary electron energy predicted by each decision tree. The decision tree is an energy prediction model trained for each sub-region of interest.

[0060] Step d2: Calculate the average value of the secondary electron energy predicted by each decision tree to obtain the average secondary electron energy.

[0061] Specifically, in this embodiment of the invention, after subdividing the region of interest, the grayscale data of each sub-region of interest is extracted, and the average energy of secondary electrons in each sub-region of interest is calculated. When extracting the grayscale data of each sub-region of interest, the average grayscale value of each sub-region of interest is optimal. In this embodiment of the invention, the average grayscale value of the corresponding sub-region of interest can be obtained by summing and then averaging the grayscale values ​​of each pixel. For example, the calculation formula is as follows:

[0062]

[0063] In the formula, ROI represents a sub-region of interest. I(x,y) represents the corresponding average gray value, (x,y) represents the pixel coordinates, I(x,y) represents the gray value of the pixel, and N represents the number of pixels.

[0064] Subsequently, in this embodiment of the invention, a decision tree is trained for each sub-region of interest based on the random forest algorithm, serving as the energy prediction model for each sub-region of interest. The average gray value and key features (texture, composition, etc.) are the input data for the decision tree, and the output is the secondary electron energy.

[0065] By establishing a correlation between grayscale values ​​and electron energies using an empirical formula, and combining this with regression analysis or machine learning methods, a more accurate mapping relationship between grayscale values ​​and secondary electron energies can be established, as shown in the following formula. This allows the model to not only consider a single grayscale value but also incorporate additional features such as texture and morphology for energy estimation.

[0066]

[0067] In the above formula, f i (x) is the prediction result of the i-th decision tree, used to represent the secondary electron energy of the i-th sub-region of interest. The input variables are the extracted features (grayscale value, texture, morphology, etc.); N is the number of decision trees, used to correspond to the N sub-regions of interest; E represents the predicted average secondary electron energy.

[0068] Based on the above formula, after predicting the secondary electron energy of each sub-region of interest using the random forest algorithm, the secondary electron energy of the sub-region of interest is summed and averaged using a majority voting method. By combining the prediction results of multiple decision trees, the final average secondary electron energy of the region of interest is obtained, which can significantly improve the prediction accuracy of the average secondary electron energy.

[0069] In some optional implementations, step b5 above further includes:

[0070] Step e1: Input the key features and average gray value of each sub-region of interest into the corresponding decision tree to obtain the secondary electron energy predicted by each decision tree;

[0071] Step e2: Obtain the weight parameters corresponding to each decision tree, where the sum of the weight parameters corresponding to each decision tree is 1;

[0072] Step e3: Based on the weight parameters corresponding to each decision tree, the predicted secondary electron energies of each decision tree are weighted and summed to obtain the average secondary electron energy.

[0073] Specifically, in this embodiment of the invention, another method for calculating the average energy of secondary electrons within a region of interest is provided. Similarly, decision trees are trained based on the key features and average gray values ​​of each sub-region of interest. During training, different weight parameters are assigned to different decision trees based on the training data. Then, the secondary electron energies predicted by each decision tree are weighted and summed using the weight parameters corresponding to each decision tree to obtain the average energy of the secondary electrons, as shown in the following formula:

[0074]

[0075] In the formula, E represents the predicted average energy of secondary electrons; X represents the input features, including the average gray value, texture features, and morphological features of the t-th sub-region of interest; w t h represents the weight coefficients of the t-th decision tree, obtained through training on the training data. t (X) is the secondary electron energy of the t-th sub-region of interest obtained by the t-th decision tree predicting the input feature X, and N is the number of decision trees, used to correspond to the N sub-regions of interest.

[0076] The above two approaches enable accurate prediction of the average secondary electron energy within the region of interest. Furthermore, the first approach (steps d1-d2) and the second approach (steps e1-e3) can complement each other to solve complex problems, providing different perspectives and computational support. For example, the first and second approaches can be used to predict the average secondary electron energy within the region of interest, and then the results of the two approaches can be combined to calculate the mean, thereby further improving the prediction accuracy of the average secondary electron energy within the region of interest.

[0077] In some optional embodiments, the scanning electron microscope focusing method provided by the present invention further includes:

[0078] Step f1: Calculate the energy statistics based on the secondary electron energies of the region of interest;

[0079] Step f2: Adjust the voltage value of the electrostatic accelerator tube of the scanning electron microscope according to the energy statistics index.

[0080] Specifically, this embodiment of the invention performs global and local feature analysis on the secondary electron energy within the region of interest (ROI) to examine the energy distribution throughout the ROI. By calculating energy statistical indicators such as the mean, standard deviation, and skewness of the energy distribution, the understanding of electron energy can be further refined, thereby revealing the overall energy level and distribution pattern of secondary electrons within the ROI. Local feature analysis and weighted analysis are performed based on the overall energy level and distribution pattern of secondary electrons within the ROI. Different weights are assigned to different ROIs according to their varying sensitivities to energy changes. These statistical indicators help understand the relationship between the intensity of the secondary electron signal and the sample surface characteristics, providing crucial reference for subsequent adjustments to the electrostatic accelerator tube voltage, analysis of image brightness, and assessment of sample surface roughness and graininess. This enables high-quality ROI analysis and image quality optimization, provides auxiliary data for setting the electrostatic accelerator tube voltage, and further improves the accuracy of voltage adjustment.

[0081] For example, in embodiments of the present invention, the average energy can be calculated using the following formula:

[0082]

[0083] In the formula, μ is the average energy, and E i Let be the secondary electron energy of the i-th pixel, and N be the total number of pixels in the region of interest.

[0084] The energy standard deviation can also be calculated using the following formula to measure the dispersion of secondary electron energies within the region of interest, i.e., the fluctuation of energy values:

[0085]

[0086] In the formula, N is the total number of pixels in the region of interest, and E i Let be the secondary electron energy of the i-th pixel, μ be the mean energy, and σ be the standard deviation of energy.

[0087] The energy bias can also be calculated using the following formula to represent the symmetry of the energy distribution. A positive energy bias indicates a right-skewed distribution (higher energy pixels are in the majority), while a negative energy bias indicates a left-skewed distribution (lower energy pixels are in the majority):

[0088]

[0089] In the formula, N is the total number of pixels in the region of interest, and E i Let be the secondary electron energy of the i-th pixel, μ be the mean energy, σ be the standard deviation of energy, and Skewness be the energy deviation.

[0090] The mean energy value can be used to determine the central level of the energy distribution, the standard deviation energy value can be used to determine the dispersion of the energy distribution, and the energy deviation energy value can be used to further understand the symmetry or skewness of the energy distribution. These features provide the necessary basic data for subsequent local feature analysis and weighted analysis.

[0091] In some optional embodiments, the scanning electron microscope focusing method provided by this invention further incorporates a real-time dynamic analysis module. During image acquisition, based on feedback from ROI grayscale changes and energy calculations, the module dynamically adjusts the parameters of the analysis algorithm, optimizing the secondary electron energy calculation process in real time and improving the prediction accuracy of the secondary electron average energy. For example, it determines the grayscale threshold of the region of interest, the sensitivity parameter for grayscale changes, etc., and determines whether to trigger algorithm adjustment when the threshold is exceeded, otherwise no adjustment is made. The adjusted parameters include, but are not limited to, time interval parameters during image acquisition, weighting coefficients involved in energy calculation, and constant terms in the energy calculation formula.

[0092] In some optional embodiments, the scanning electron microscope focusing method provided by the present invention also performs statistical analysis based on multiple acquisitions of grayscale data, key features, etc., to identify and correct possible error sources. By addressing the negative impacts caused by error sources, the energy value is standardized in the final secondary electron average energy prediction result through statistical learning methods or calibration curves to ensure the accuracy of the final result.

[0093] In some optional embodiments, the scanning electron microscope focusing method provided by the present invention further includes:

[0094] Step g1: Obtain the secondary electron brightness value and backscattered electron brightness of the region of interest in the first focused image;

[0095] Step g2: Determine the brightness difference between the secondary electron brightness value and the backscattered electron brightness value;

[0096] Step g3: Adjust the voltage value of the electrostatic accelerating tube of the scanning electron microscope according to the brightness difference so that the scanning electron microscope outputs a second focused image.

[0097] Specifically, in this embodiment of the invention, a further optimization method for image focusing is also provided. By acquiring the secondary electron brightness value and backscattered electron brightness value of the region of interest in the first focused image, and comparing the brightness difference between the two, the brightness difference can reflect characteristics such as the roughness and grain size of the sample surface. Based on the analysis results of the brightness difference, the voltage of the electrostatic accelerator tube is adjusted to optimize the ratio of secondary electron and backscattered electron signals, thereby improving image quality.

[0098] In some optional embodiments, the scanning electron microscope focusing method provided by the present invention further includes:

[0099] Step h1: Obtain the backscattered electron image after processing the first focused image;

[0100] Step h2: Extract the surface roughness and graininess of the region of interest in the processed backscattered electron image;

[0101] Step h3: If the surface roughness and particle size are not within the corresponding preset threshold range, adjust the electrostatic accelerating tube voltage value of the scanning electron microscope according to the preset threshold corresponding to the surface roughness and particle size, so that the surface roughness and particle size fall within the corresponding preset threshold range, and output the third focused image.

[0102] Specifically, in embodiments of the present invention, such as Figure 4 As shown, a further optimization method for image focusing is also provided. The surface roughness and grain size of the sample are determined by analyzing the backscattered electron image of the region of interest using image processing techniques. Roughness analysis can be obtained by calculating the distribution of image grayscale values ​​(such as standard deviation or variance), while grain size can be estimated by analyzing the distribution, size, and morphology of particles in the image.

[0103] Next, it is determined whether the surface roughness and grain size values ​​are within the corresponding preset threshold range. Based on the preset threshold range, it is determined whether the surface roughness and grain size of the region of interest meet the standard. The preset threshold can be set based on historical data or standard test results, which will not be elaborated in this embodiment. If the surface roughness and grain size are not within the corresponding preset threshold range, the electrostatic accelerator tube voltage value of the scanning electron microscope is adjusted according to the preset threshold corresponding to the surface roughness and grain size until the surface roughness and grain size values ​​are within the preset threshold range. Finally, a third focused image is output, further improving the focusing effect of the image.

[0104] In some optional embodiments, when the scanning electron microscope finishes focusing, provided that the surface roughness and grain size meet the standards, embodiments of the present invention can also record the current voltage value of the electrostatic accelerator tube as a voltage adjustment reference for the electrostatic accelerator tube of the sample under test. This voltage value will be directly used as a voltage adjustment reference for future tests of the sample, ensuring that subsequent experiments are conducted under the same or similar voltage conditions, facilitating subsequent analysis and experiment reproduction, and improving the scanning efficiency of the microscope.

[0105] In a specific application scenario embodiment, such as Figure 5 As shown, a complete technical solution provided by the present invention is implemented through the following steps:

[0106] 1. Place the sample to be tested under a scanning electron microscope;

[0107] 2. Acquire sample images using a scanning electron microscope, and then determine the region of interest;

[0108] 3. Calculate the average secondary electron energy of the sample image based on the grayscale values ​​of the region of interest in the sample image;

[0109] 4. The voltage control module calculates the voltage value of the electrostatic accelerator tube based on the average energy of the secondary electrons and the electron charge of the electrostatic accelerator tube;

[0110] 5. The voltage control module sets the voltage of the electrostatic accelerator tube of the scanning electron microscope based on the voltage value of the electrostatic accelerator tube, so that the scanning electron microscope outputs the first focused image;

[0111] 6. The scanning electron microscope continues to acquire the first focused image. On the one hand, it determines the brightness difference between the secondary electron brightness value and the backscattered electron brightness. On the other hand, it extracts the surface roughness and graininess of the region of interest in the processed backscattered electron image from the first focused image.

[0112] 7. Adjust the electrostatic accelerator tube voltage value according to the brightness difference, and / or surface roughness and graininess, until the surface roughness and graininess are within the corresponding preset threshold range, and then end the adjustment.

[0113] 8. Output the final focused image and save the electrostatic accelerator tube voltage value.

[0114] The technical solution provided by the embodiments of the present invention has the following advantages:

[0115] 1. By precisely analyzing the grayscale data of the secondary electron image, its average energy is determined, and the control voltage of the electrostatic accelerator tube is then accurately set. This energy-matching-based voltage adjustment method enables the electron beam to be focused more accurately on the sample surface, thereby significantly improving image resolution and allowing the microstructure and surface details of the sample to be presented more clearly. For example, when observing the micromorphology of nanomaterials, smaller particles and finer textures can be clearly distinguished, providing more accurate data support for materials research.

[0116] 2. During the analysis, the brightness values ​​of secondary electrons and backscattered electrons are comprehensively considered. By comparing and optimizing their proportional relationship, the image contrast is further enhanced. This helps to highlight the differences between different component or structural regions on the sample surface, making various features in the image more distinct and facilitating researchers to more accurately identify and analyze the characteristics of the sample.

[0117] 3. For complex samples, their surfaces often exhibit diverse regional characteristics, such as varying compositions, roughness, and particle size. This approach subdivides these samples into multiple regions of interest and performs grayscale analysis, energy estimation, brightness comparison, and surface feature extraction on each region separately. This allows for precise focusing and analysis of the characteristics of different regions. For instance, when studying biological tissue sections, optimized imaging can be performed on cellular and extracellular matrix regions separately, providing a comprehensive and in-depth understanding of the overall characteristics of the sample.

[0118] 4. By training an image segmentation network with the help of a deep learning model, key features in the image can be automatically extracted. For example, the region growing algorithm can be used to expand based on seed points, and the growth threshold can be dynamically adjusted according to pixel gray values ​​and other features of multi-channel images, thereby extracting sample surface features more accurately and providing a more powerful tool for the analysis of complex samples.

[0119] 5. During the extraction of grayscale values ​​from the region of interest, the secondary electron image and the backscattered electron image are preprocessed and fused. The influence of various features such as grayscale, texture, and morphology on the secondary electron energy is analyzed to improve the accuracy of secondary electron energy calculation.

[0120] This embodiment also provides a scanning electron microscope focusing device for implementing the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0121] This embodiment provides a focusing device for a scanning electron microscope, such as... Figure 6 As shown, it includes:

[0122] The initial image acquisition module 601 is used to acquire sample images of the sample to be tested using a scanning electron microscope;

[0123] The electron energy prediction module 602 is used to calculate the average secondary electron energy of the sample image based on the grayscale value of the sample image.

[0124] Voltage calibration module 603 is used to determine the voltage value of the electrostatic accelerator tube based on the average energy of the secondary electrons and the electron charge of the electrostatic accelerator tube.

[0125] The microscope adjustment module 604 is used to configure the electrostatic accelerator tube of the scanning electron microscope based on the voltage value of the electrostatic accelerator tube, so that the scanning electron microscope outputs a first focused image.

[0126] Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0127] In this embodiment, the focusing device of the scanning electron microscope is presented in the form of a functional unit. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0128] This invention also provides a computer device; please refer to [link / reference]. Figure 7 , Figure 7 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 7 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 7 Take a processor 10 as an example.

[0129] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0130] The memory 20 stores instructions executable by at least one processor 10 to cause at least one processor 10 to perform the method shown in the above embodiments.

[0131] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0132] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0133] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0134] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0135] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0136] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A focusing method for scanning electron microscopy, characterized in that, The method includes: Sample images of the sample to be tested are acquired using a scanning electron microscope; The calculation of the average secondary electron energy of the sample image based on its grayscale values ​​includes: determining a region of interest (ROI) in the sample image; extracting key features of the ROI; calculating the average grayscale value of the ROI; obtaining an energy prediction model, which is trained using a neural network based on known secondary electron energy labels, key feature samples, and grayscale value samples; and inputting the key features and the average grayscale value into the energy prediction model to obtain the average secondary electron energy of the ROI. The voltage value of the electrostatic accelerator tube is determined based on the average energy of the secondary electrons and the electron charge of the electrostatic accelerator tube. The electrostatic accelerator tube of the scanning electron microscope is configured based on the voltage value of the electrostatic accelerator tube so that the scanning electron microscope outputs a first focused image.

2. The method according to claim 1, characterized in that, The process of acquiring sample images of the sample to be tested using a scanning electron microscope includes: Secondary electron images and backscattered electron images of the sample under test are acquired using the scanning electron microscope. The secondary electron image and the backscattered electron image are preprocessed and fused to obtain the sample image.

3. The method according to claim 1 or 2, characterized in that, The average grayscale value of the region of interest is calculated, including: The seed location for image segmentation is determined based on the key features; Based on the seed position, a region growing algorithm is performed on the region of interest to subdivide the region of interest into several sub-regions of interest; Calculate the average gray value corresponding to each of the sub-regions of interest.

4. The method according to claim 3, characterized in that, The key features and the average gray value are input into the energy prediction model to obtain the average secondary electron energy of the region of interest, including: The key features and average gray values ​​of each sub-region of interest are input into the corresponding decision tree to obtain the secondary electron energy predicted by each decision tree. The decision tree is the energy prediction model trained for each sub-region of interest. The average secondary electron energy predicted by each decision tree is calculated to obtain the average secondary electron energy. or, Input the key features and average gray value of each sub-region of interest into the corresponding decision tree to obtain the secondary electron energy predicted by each decision tree; Obtain the weight parameters corresponding to each decision tree, where the sum of the weight parameters corresponding to each decision tree is 1; The average energy of the secondary electron is obtained by weighting and summing the secondary electron energies predicted by each decision tree based on the weight parameters corresponding to each decision tree.

5. The method according to claim 1, characterized in that, The method further includes: Obtain the secondary electron brightness value and backscattered electron brightness value of the region of interest in the first focused image; Determine the brightness difference between the secondary electron brightness value and the backscattered electron brightness; The voltage value of the electrostatic accelerator tube of the scanning electron microscope is adjusted according to the brightness difference so that the scanning electron microscope outputs a second focused image.

6. The method according to claim 1, characterized in that, The method further includes: Obtain the backscattered electron image after processing the first focused image; Extract the surface roughness and graininess of the region of interest in the processed backscattered electron image; If the surface roughness and the particle size are not within the corresponding preset threshold range, the electrostatic acceleration tube voltage value of the scanning electron microscope is adjusted according to the preset threshold corresponding to the surface roughness and the particle size, so that the surface roughness and the particle size fall within the corresponding preset threshold range, and a third focused image is output.

7. A focusing device for a scanning electron microscope, characterized in that, The device includes: The initial image acquisition module is used to acquire sample images of the sample to be tested using a scanning electron microscope; An electron energy prediction module is used to calculate the average secondary electron energy of a sample image based on its grayscale values. The calculation includes: determining a region of interest (ROI) in the sample image; extracting key features from the ROI; calculating the average grayscale value of the ROI; obtaining an energy prediction model, which is trained using a neural network based on known secondary electron energy labels, key feature samples, and grayscale value samples; and inputting the key features and the average grayscale value into the energy prediction model to obtain the average secondary electron energy of the ROI. A voltage calibration module is used to determine the voltage value of the electrostatic accelerator tube based on the average energy of the secondary electrons and the electron charge of the electrostatic accelerator tube. The microscope adjustment module is used to configure the electrostatic accelerator tube of the scanning electron microscope based on the voltage value of the electrostatic accelerator tube, so that the scanning electron microscope outputs a first focused image.

8. A computer device, characterized in that, include: A memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, the processor executing the computer instructions to perform the method of any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a computer to perform the method of any one of claims 1 to 6.

Images