A method, device and storage medium for determining scanning electron microscope image period

Abstract

The present application discloses a method, device and storage medium for determining the period of a scanning electron microscope image. The method comprises: expanding the original scanning electron microscope image, obtaining all period values ​​of the expanded original scanning electron microscope image in the target direction by Fourier transform; moving the original scanning electron microscope image along the target direction by one period based on the period value to obtain the corresponding target scanning electron microscope image; obtaining the absolute error and image between the corresponding target scanning electron microscope image and the original scanning electron microscope image, and determining the mean of the absolute error and image; and determining the final period value of the scanning electron microscope image in the target direction according to the mean and the preset mean. The scheme of the present application can improve the accuracy and stability of the period value.

Description

Technical Field

[0001] The present application generally relates to the field of scanning electron microscope image period detection technology. More specifically, the present application relates to a method, device and computer-readable storage medium for determining the period of a scanning electron microscope image. Background Art

[0002] Scanning electron microscopes (SEMs) are high-precision instruments used for characterizing sample morphology and analyzing composition. For example, modern, advanced field-emission SEMs (FESEMs) have a resolution of 1 nanometer, enabling researchers and technicians to explore and uncover the rich microscopic world. In particular, with the rapid development of the semiconductor industry, semiconductor wafer defect detection, a crucial step in semiconductor manufacturing, is facing increasingly stringent requirements. Detecting wafer defects using SEM images of semiconductor wafer defects captured by SEMs is also a widespread practice.

[0003] There are three main methods for detecting semiconductor wafer defects based on charged particle counts: D2D (Die to Die), C2C (Cell to Cell), and D2DB (Die to Database). For inherently periodic patterns, the C2C method can be used to determine the defect location by comparing the pitch differences between adjacent periods. In the actual acquisition of scanning electron microscope images of the same wafer, the period value can reach sub-pixel levels, and the periodic image has rich texture. Therefore, the accuracy of the period value is extremely high. A better method for calculating periodicity helps accurately determine the defect location and reduce the false detection rate. Currently, methods for calculating image period include time domain and frequency domain. In the time domain, the period is determined by calculating the autocorrelation coefficient, but this method is susceptible to noise and the period value calculation is unstable. In the frequency domain, the period is mainly calculated through Fourier transform, but using only Fourier transform does not provide sufficient accuracy.

[0004] In view of this, there is an urgent need to provide a solution for determining the period of a scanning electron microscope image, so as to improve the accuracy and stability of the period value. Summary of the Invention

[0005] In order to at least solve one or more of the above-mentioned technical problems, the present application proposes a solution for determining a scanning electron microscope image period in multiple aspects.

[0006] In a first aspect, the present application provides a method for determining the period of a scanning electron microscope image, comprising: expanding an original scanning electron microscope image, and obtaining all period values ​​of the expanded original scanning electron microscope image in a target direction using Fourier transform; moving the original scanning electron microscope image along the target direction by one period based on the period value to obtain a corresponding target scanning electron microscope image; obtaining an absolute error and an image between the corresponding target scanning electron microscope image and the original scanning electron microscope image, and determining a mean of the absolute error and the image; and determining a final period value of the scanning electron microscope image in the target direction based on the mean and a preset mean.

[0007] In one embodiment, the original scanning electron microscope image is expanded by the following operations: projecting the original scanning electron microscope image along the column direction and calculating the mean to generate a one-dimensional mean matrix; and adding multiple columns of expansion elements to the one-dimensional mean matrix to obtain an expansion matrix to expand the original scanning electron microscope image.

[0008] In another embodiment, using Fourier transform to obtain all periodic values ​​of the expanded original scanning electron microscope image in the target direction includes: performing a one-dimensional Fourier transform on the expanded matrix to obtain a one-dimensional spectrum matrix; determining the position of the target peak point in the one-dimensional spectrum matrix in response to satisfying the minimum peak spacing and the minimum peak value; and obtaining all periodic values ​​of the expanded original scanning electron microscope image in the target direction based on the number of columns of the expanded matrix and the position of the target peak point.

[0009] In another embodiment, performing a one-dimensional Fourier transform on the expanded matrix to obtain a one-dimensional spectrum matrix includes: performing a one-dimensional Fourier transform on the expanded matrix to obtain a spectrum graph; moving the zero-frequency component in the spectrum graph to the center of the spectrum; and selecting the right half of the spectrum center and calculating the absolute value to obtain the one-dimensional spectrum matrix.

[0010] In another embodiment, wherein the target direction includes a horizontal direction and / or a vertical direction, the period value includes a period value of the horizontal direction, a period value of the vertical direction, and a combined period value of the period value of the horizontal direction and the period value of the vertical direction, and moving the original scanning electron microscope image along the target direction by one period based on the period value to obtain a corresponding target scanning electron microscope image includes: moving the original scanning electron microscope image along the horizontal direction or the vertical direction by one period based on the period value of the horizontal direction or the period value of the vertical direction to obtain the corresponding target scanning electron microscope image; or moving the original scanning electron microscope image along the horizontal direction and the vertical direction by one period each based on the combined period value to obtain the corresponding target scanning electron microscope image.

[0011] In another embodiment, obtaining the absolute error and image between the corresponding target SEM image and the original SEM image and determining the mean of the absolute error and image includes: calculating the dot division result between the corresponding target SEM image and the original SEM image to obtain the absolute error and image; and calculating the average value of the pixel grayscale in the absolute error and image to determine the mean of the absolute error and image.

[0012] In another embodiment, calculating the dot division result between the corresponding target SEM image and the original SEM image to obtain the absolute error and image includes: respectively cropping the corresponding target SEM image and the original SEM image according to the target width to obtain corresponding first cropped images and second cropped images; respectively calculating the absolute value result and the sum result of the difference between the first cropped image and the second cropped image; and dividing the absolute value result of the difference by the sum result to obtain the dot division result to obtain the absolute error and image.

[0013] In yet another embodiment, the target width is determined by a corresponding period value and a width of the original SEM image.

[0014] In another embodiment, determining the final periodicity value of the scanning electron microscope image in the target direction based on the mean and the preset mean includes: determining the minimum mean of the absolute error and the image; and comparing the minimum mean with the preset mean to determine the final periodicity value of the scanning electron microscope image in the horizontal direction and / or the vertical direction.

[0015] In yet another embodiment, comparing the minimum mean with the preset mean to determine the final period value of the scanning electron microscope image in the horizontal direction and / or the vertical direction includes: in response to the minimum mean being less than the preset mean, determining the absolute error of the minimum mean and the period value corresponding to the image as the final period value of the scanning electron microscope image in the horizontal direction and / or the vertical direction.

[0016] In a second aspect, the present application provides a device for determining a scanning electron microscope image period, comprising: a processor; and a memory storing program instructions for determining a scanning electron microscope image period, wherein when the program instructions are executed by the processor, the device implements multiple embodiments of the aforementioned first aspect.

[0017] In a third aspect, the present application provides a computer-readable storage medium having stored thereon computer-readable instructions for determining a scanning electron microscope image period, wherein when the computer-readable instructions are executed by one or more processors, multiple embodiments of the aforementioned first aspect are implemented.

[0018] Through the scheme for determining the period of an SEM image provided above, embodiments of the present application expand the original SEM image and calculate the period value using Fourier transform, thereby achieving a higher accuracy in the calculated period value. By calculating the absolute error and image mean of the target SEM image obtained after the original SEM image is moved in the target direction for a corresponding period, and combining the calculated period value with a preset mean, the final period value is determined, thereby facilitating accurate determination of the period value and improving the stability of the calculated period value. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The above and other objects, features and advantages of the exemplary embodiments of the present application will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present application are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0020] Figure 1 is an exemplary flow chart illustrating a method for determining a scanning electron microscope image period according to an embodiment of the present application;

[0021] Figure 2 is an exemplary flowchart illustrating calculation of all periodic values ​​in a target direction according to an embodiment of the present application;

[0022] Figure 3 is an exemplary flow chart illustrating calculation of a SAD image and its mean according to an embodiment of the present application;

[0023] Figure 4 is an exemplary schematic diagram showing an original scanning electron microscope image according to an embodiment of the present application;

[0024] Figure 5 is an exemplary schematic diagram showing a one-dimensional mean matrix according to an embodiment of the present application;

[0025] Figure 6 is an exemplary schematic diagram showing an expansion matrix according to an embodiment of the present application;

[0026] Figure 7 is an exemplary schematic diagram showing a one-dimensional spectrum matrix according to an embodiment of the present application;

[0027] Figure 8 is an exemplary schematic diagram showing an original scanning electron microscope image and a corresponding target scanning electron microscope image according to an embodiment of the present application;

[0028] Figure 9 is a SAD image showing target SEM images and original SEM images corresponding to three periods in the horizontal direction according to an embodiment of the present application;

[0029] Figure 10 This is an exemplary structural block diagram of a device for determining a scanning electron microscope image period according to an embodiment of the present application. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0031] It should be understood that the terms "include" and "comprising" used in the description and claims of this application indicate the presence of described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0032] It should also be understood that the terms used in this specification are for the purpose of describing specific embodiments only and are not intended to limit this application. As used in this specification and claims, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should also be further understood that the term "and / or" as used in this specification and claims refers to any and all possible combinations of one or more of the associated listed items, including and including these combinations.

[0033] As used in this specification and claims, the term "if" can be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.

[0034] The specific implementation of the present application will be described in detail below with reference to the accompanying drawings.

[0035] Figure 1 FIG. 1 is an exemplary flow chart illustrating a method 100 for determining a scanning electron microscope image period according to an embodiment of the present application. Figure 1 As shown in FIG, in step S101, the original SEM image is expanded, and all periodic values ​​of the expanded original SEM image in the target direction are obtained by Fourier transform. In some implementation scenarios, the original SEM image (e.g. Figure 4The image (shown) can be, for example, but not limited to, a scanning electron microscope image of a semiconductor wafer acquired using a scanning electron microscope. In one embodiment, the original scanning electron microscope image can be expanded by projecting the original scanning electron microscope image along the column direction and calculating the mean to generate a one-dimensional mean matrix. Multiple columns of extension elements are then added to the one-dimensional mean matrix to obtain an expansion matrix. The aforementioned extension elements can be, for example, 0.

[0036] In other words, the grayscale mean of all pixels in each column of the original SEM image is calculated to generate a one-dimensional mean matrix (e.g. Figure 5 For example, assuming that the original SEM image is M*N (i.e., M rows and N columns), a 1*N one-dimensional mean matrix is ​​generated by averaging each column. As an example, assuming that N is 1000, when the original SEM image is expanded to 10001, the expanded matrix can be obtained by appending "0" to the 1000th column of the one-dimensional mean matrix (e.g. Figure 6 to expand the original SEM images.

[0037] Based on the expanded original SEM image, in one embodiment, a one-dimensional Fourier transform can first be performed on the expanded matrix to obtain a one-dimensional spectrum matrix. In response to satisfying the minimum peak spacing and the minimum peak value, the location of the target peak point is determined in the one-dimensional spectrum matrix, thereby obtaining all periodic values ​​of the expanded original SEM image in the target direction based on the number of columns in the expanded matrix and the location of the target peak point. More specifically, in one implementation scenario, a spectrum graph can be obtained by performing a one-dimensional Fourier transform on the expanded matrix. The zero-frequency component in the spectrum graph is moved to the center of the spectrum, and then the right half of the spectrum center is selected and the absolute value is calculated to obtain the one-dimensional spectrum matrix.

[0038] It can be understood that the spectrum of the aforementioned expanded matrix after Fourier transformation is bilaterally symmetrical, so the right half of the spectrum center can be selected to calculate the absolute value. In addition, there will be multiple peak points in the one-dimensional spectrum matrix after Fourier transformation. Therefore, according to the obtained one-dimensional spectrum matrix, the position of the target peak point can be determined by setting the minimum peak spacing and the minimum peak value (for example Figure 7 Specifically, peak points that satisfy a value greater than the minimum peak value and a spacing between peak points greater than the minimum peak spacing are selected as target peak points and their positions are determined. For example, in an exemplary scenario, assume that the minimum peak spacing and the minimum peak value are set to 2 and 0.5, respectively. In this scenario, peak points that are greater than 0.5 and have a spacing between peak points greater than 2 are selected as target peak points and their positions are determined.

[0039] Furthermore, based on the position of the determined target peak point and the number of columns of the expansion matrix, all periodic values ​​in the target direction can be determined. In one implementation scenario, the aforementioned target direction may include a horizontal direction and / or a vertical direction. Accordingly, the aforementioned periodic value may include a periodic value in the horizontal direction, a periodic value in the vertical direction, and a combined periodic value of a periodic value in the horizontal direction and a periodic value in the vertical direction. That is, when the target direction is a horizontal direction or a vertical direction, its periodic value corresponds to all periodic values ​​in the horizontal direction or all periodic values ​​in the vertical direction; when the target direction is a horizontal direction and a vertical direction (i.e., considering both the horizontal direction and the vertical direction), its periodic value corresponds to a combination of all periodic values ​​in the horizontal direction and all periodic values ​​in the vertical direction.

[0040] Taking the horizontal direction as an example, assuming the target peak points are 10, 16, and 23, and the number of columns in the expansion matrix is ​​10001, all period values ​​500.50, 312.81, and 217.60 in the horizontal direction can be obtained by 10001 / 2 / 10, 10001 / 2 / 16, and 10001 / 2 / 23. Similarly, all period values ​​can be obtained when the target direction is the vertical direction. In some embodiments, when the target direction is the horizontal direction and the vertical direction, the corresponding period is the combination of all period values ​​in the horizontal direction and all period values ​​in the vertical direction. As an example, assuming that all period values ​​in the horizontal direction are represented as (P1, P2, P3) and all period values ​​in the vertical direction are represented as (p1, p2, p3), the combined period values ​​(P1, p1), (P1, p2), (P1, p3), (P2, p1), (P2, p2), (P2, p3), (P3, p1), (P3, p2) and (P3, p3) can be obtained.

[0041] After obtaining the above-mentioned period value, at step S102, the original scanning electron microscope image is moved by one period in the target direction based on the period value to obtain the corresponding target scanning electron microscope image. In one embodiment, the original scanning electron microscope image is moved by one period in the horizontal direction or the vertical direction based on the period value in the horizontal direction or the period value in the vertical direction to obtain the corresponding target scanning electron microscope image. Alternatively, the original scanning electron microscope image is moved by one period in the horizontal direction and one period in the vertical direction based on the combined period value to obtain the corresponding target scanning electron microscope image. That is, based on each period value, the original scanning electron microscope image is moved by one period in the horizontal direction or the vertical direction; or based on each combined period value, the original scanning electron microscope image is first moved by one period in the horizontal direction and then by one period in the vertical direction to obtain the corresponding target scanning electron microscope image under each period value or each combined period value.

[0042] Then, at step S103, the absolute error sum image between the corresponding target SEM image and the original SEM image is obtained, and the mean of the absolute error sum image is determined. In one embodiment, the absolute error sum image is obtained by calculating the dot division result between the corresponding target SEM image and the original SEM image, and then the mean of the absolute error sum image is determined based on the average value of the pixel grayscale in the absolute error sum image. Specifically, in one implementation scenario, the corresponding target SEM image and the original SEM image are respectively intercepted according to the target width to obtain a corresponding first intercepted image and a second intercepted image, and the absolute value result and the sum result of the difference between the first intercepted image and the second intercepted image are respectively calculated, and the absolute value result of the difference is divided by the sum result to obtain the dot division result to obtain the Sum of Absolute Differences (“SAD”) image.

[0043] That is, the target SEM image and the original SEM image after the corresponding movement of each cycle are first intercepted according to the target width to obtain a first intercepted image (for example, recorded as test) and a second intercepted image (for example, recorded as ref). Then, the absolute value of the difference between the first intercepted image test and the second intercepted image ref, |ref-test|, and the sum ref+test of the first intercepted image test and the second intercepted image ref are calculated respectively. By performing a dot division on |ref-test| and ref+test, the SAD image under the corresponding cycle can be obtained. In some embodiments, the aforementioned target width can be determined by the corresponding cycle value and the width of the original SEM image. For example, the target width can be expressed as ceil(P):N, where P represents the cycle value, ceil(P) represents rounding up P; N represents the width of the original SEM image, that is, the number of columns. The aforementioned dot division can be normalized to the range of 255, that is, 255·|ref-test| / (ref+test).

[0044] Furthermore, based on the obtained SAD image under the corresponding period, the mean value is obtained by calculating the average value of the pixel grayscale in the SAD image. For example, for the horizontal direction or the vertical direction, assuming that three period values ​​are calculated thereon, the SAD image under the three period values ​​and its mean value can be obtained respectively. For the horizontal direction and the vertical direction, assuming that three period values ​​are calculated therein, nine combined periods can be obtained, and the SAD image under the nine period values ​​and its mean value can be obtained accordingly.

[0045] Finally, at step S104, the final period value of the scanning electron microscope image in the target direction is determined based on the mean and the preset mean. In one embodiment, by determining the absolute error and the minimum mean of the image, the minimum mean is compared with the preset mean to determine the final period value of the scanning electron microscope image in the horizontal direction and / or the vertical direction. More specifically, in an implementation scenario, in response to the minimum mean being less than the preset mean, the absolute error of the minimum mean and the period value corresponding to the image are determined as the final period value of the scanning electron microscope image in the horizontal direction and / or the vertical direction. That is, the minimum mean is selected from the mean values ​​of the SAD images obtained corresponding to the multiple periods, and the period value of the SAD image corresponding to the minimum mean is determined, and then the minimum mean is compared with the preset mean (i.e., the preset absolute error and image mean). When the minimum mean is less than the preset mean, the period value of the SAD image corresponding to the minimum mean is determined to be the final period value of the scanning electron microscope image in the horizontal direction and / or the vertical direction.

[0046] In combination with the above description, it can be seen that the embodiment of the present application is expanded by the original scanning electron microscope image, and the cycle value is calculated using Fourier transform, so that the cycle value accuracy of the calculation is higher. Further, by calculating the target scanning electron microscope image obtained after the original scanning electron microscope image moves the corresponding cycle in the target direction (such as horizontal direction and / or vertical direction) and the SAD image and the mean value thereof of the original scanning electron microscope image, and comparing the final cycle value with the preset mean value, it is possible to accurately determine the cycle value and improve the stability of the calculated cycle value. In some embodiments, by collecting the scanning electron microscope image of the semiconductor wafer, the cycle value obtained using the scheme of the present application is used to determine the wafer defect, which can make the position of the wafer defect more accurate, greatly reducing the false detection rate of the wafer defect.

[0047] Figure 2 FIG. 1 is an exemplary flow chart showing the calculation of all periodic values ​​in the target direction according to an embodiment of the present application. Figure 2 As shown in , at step S201, the original scanning electron microscope image is projected along the column direction and the mean is calculated to generate a one-dimensional mean matrix. Specifically, the one-dimensional mean matrix can be generated by calculating the grayscale mean of all pixels on each column in the original scanning electron microscope image. Then, at step S202, a plurality of columns of extension elements are added to the one-dimensional mean matrix to obtain an extended matrix. In an implementation scenario, the aforementioned extension element may be, for example, 0, thereby obtaining an extended matrix by appending "0" to the one-dimensional mean matrix. Further, at step S203, a spectrum diagram is obtained by performing a one-dimensional Fourier transform on the extended matrix, the zero-frequency component in the spectrum diagram is moved to the center of the spectrum, and then the right half of the center of the spectrum is selected and the absolute value is calculated to obtain a one-dimensional spectrum matrix.

[0048] Based on the obtained one-dimensional spectrum matrix, at step S204, the position of the target peak point is determined by setting the peak minimum spacing and the peak minimum value. As mentioned above, there will be multiple peak points in the one-dimensional spectrum matrix after Fourier transform, and thus by setting the peak minimum spacing (e.g., 2) and the peak minimum value (e.g., 0.5), the peak point that satisfies the condition greater than the peak minimum value and the spacing between peak points greater than the peak minimum spacing is used as the target peak point and its position is determined. Further, at step S205, all periodic values ​​of the expanded original scanning electron microscope image in the target direction are obtained according to the number of columns of the expansion matrix and the position of the target peak point. For example, assuming that the target peak points are 10, 16 and 23, and the number of columns of the expansion matrix is ​​10001, all periodic values ​​500.50 in the horizontal direction can be obtained based on 10001 / 2 / 10, 10001 / 2 / 16 and 10001 / 2 / 23 respectively. Subsequently, the target SEM image is obtained by moving the corresponding period in the target direction. The SAD image and its mean between the target SEM image and the original SEM image are calculated and compared with the preset period value to determine the final period.

[0049] Figure 3 FIG. 1 is an exemplary flow chart showing the calculation of the SAD image and its mean according to an embodiment of the present application. Figure 3 As shown in , at step S301, the original SEM image is shifted by one period along a target direction based on a period value to obtain a corresponding target SEM image. As previously mentioned, the target direction may include a horizontal direction and / or a vertical direction. Accordingly, the period value may include a period value in the horizontal direction, a period value in the vertical direction, and a combined period value of the horizontal and vertical directions.

[0050] As an example, assuming that all periodicity values ​​in the horizontal direction are represented as (P1, P2, P3) and all periodicity values ​​in the vertical direction are represented as (p1, p2, p3), by shifting the original SEM image in the horizontal direction by one period according to P1, P2, and P3, three corresponding target SEM images can be obtained. Similarly, by shifting the original SEM image in the vertical direction by one period according to p1, p2, and p3, three corresponding target SEM images can also be obtained. When the target directions are horizontal and vertical, nine combined periodicity values ​​can be obtained: (P1, p1), (P1, p2), (P1, p3), (P2, p1), (P2, p2), (P2, p3), (P3, p1), (P3, p2), and (P3, p3). By shifting the original SEM image by one period in the horizontal direction and then one period in the vertical direction according to each combined periodicity value, nine corresponding target SEM images can be obtained.

[0051] Next, at step S302, the corresponding target SEM image and the original SEM image are respectively clipped according to the target width to obtain the corresponding first clipped image test and second clipped image ref. In one embodiment, the target width can be expressed as ceil(P):N, where P represents the period value, ceil(P) represents P rounded up, and N represents the width of the original SEM image. After obtaining the first clipped image test and the second clipped image ref, at steps S303 and S304, the absolute value of the difference between the first clipped image test and the second clipped image ref, |ref-test|, and the sum of the first clipped image test and the second clipped image ref, ref+test, are calculated. At step S305, a SAD image for the corresponding period is obtained by performing a dot division based on 255·|ref-test| / (ref+test). Based on the obtained SAD image for the corresponding period, at step S306, the average grayscale value of the pixels in the SAD image is calculated to obtain the mean.

[0052] For the horizontal or vertical direction, assuming that each direction contains three periodic values, the target SAD image and the original SAD image at each periodic value are calculated to obtain the average of the three SAD images. For the horizontal and vertical directions, assuming that each direction contains nine combined periodic values, the target SAD image and the original SAD image at each combined periodic value are calculated to obtain the average of the nine SAD images. Under different target directions, by selecting the minimum mean value, comparing it with the preset mean value, and determining the period value corresponding to the SAD image with the minimum mean value smaller than the preset mean value as the final periodic value.

[0053] Figure 4 : is an exemplary schematic diagram showing an original scanning electron microscope image according to an embodiment of the present application. Figure 4 The original scanning electron microscope image is shown in . In some implementation scenarios, the original scanning electron microscope image may be, for example, a scanning electron microscope image of a semiconductor wafer. The original scanning electron microscope image has periodicity in the horizontal and vertical directions. For example, the three dotted boxes in the figure show a period in the horizontal direction, and the period in the vertical direction is the height of the original scanning electron microscope image. In an embodiment of the present application, the period in the horizontal direction or the vertical direction may be calculated separately, or the period in the horizontal and vertical directions may be calculated simultaneously. That is, the target direction may include the horizontal direction and / or the vertical direction.

[0054] Figure 5 : is an exemplary schematic diagram showing a one-dimensional mean matrix according to an embodiment of the present application. Figure 5The figure shows the one-dimensional mean matrix corresponding to the original SEM image. As previously mentioned, a one-dimensional mean matrix can be generated by calculating the grayscale mean of all pixels in each column of the original SEM image, where the horizontal axis represents the corresponding column and the vertical axis represents the grayscale mean. For example, the horizontal axis 0 corresponds to the first column of the original SEM image, and the vertical axis 50 represents the grayscale mean of all pixels in that column. Based on this one-dimensional mean matrix, an expanded matrix is ​​obtained by expanding it for subsequent Fourier transform.

[0055] Figure 6 : is an exemplary schematic diagram showing an expansion matrix according to an embodiment of the present application. Figure 6 The expanded matrix is ​​shown in , where the horizontal axis represents the corresponding column and the vertical axis represents the grayscale mean. In an implementation scenario, the expanded matrix can be obtained by appending "0" to the above-mentioned one-dimensional mean matrix to expand the original scanning electron microscope image. For a 1*N one-dimensional mean matrix, assuming that it is expanded to 10001 columns, the expanded matrix can be obtained by adding 10001-N columns of data after the above-mentioned 1*N one-dimensional mean matrix, whose values ​​correspond to 0. In the implementation scenario, a one-dimensional spectrum matrix is ​​obtained by performing a one-dimensional Fourier transform on the expanded matrix, and then the local peak points are extracted to obtain all periodic values ​​in the target direction.

[0056] Figure 7 FIG. 1 is an exemplary schematic diagram showing a one-dimensional spectrum matrix according to an embodiment of the present application. Figure 7 As shown in , a one-dimensional spectrum matrix is ​​shown, in which the horizontal axis represents frequency and the vertical axis represents amplitude. In one embodiment, a spectrum diagram can be obtained by performing a one-dimensional Fourier transform on the expanded matrix, moving the zero-frequency component in the spectrum diagram to the center of the spectrum, and then selecting the right half of the center of the spectrum and calculating the absolute value to obtain a one-dimensional spectrum matrix. In some embodiments, there will be multiple peak points in the one-dimensional spectrum matrix after Fourier transform, such as a1, a2, a3 and a4 shown as an example in the figure. Therefore, the embodiment of the present application sets the minimum peak spacing and the minimum peak value, and takes the peak point that satisfies the minimum peak value and the spacing between peak points is greater than the minimum peak spacing as the target peak point and determines its position.

[0057] As an example, assume that the minimum peak spacing and the minimum peak value are set to 2 and 0.5 respectively. In this scenario, the peak points with a spacing greater than 0.5 and a spacing greater than 2 among the peak points are taken as target peak points and their positions are determined, for example Figure 7In the figure, a1, a2, and a3 are target peak points, corresponding to positions 10, 16, and 23, respectively. Furthermore, based on 10001 / 2 / 10, 10001 / 2 / 16, and 10001 / 2 / 23, all horizontal period values, for example, 500.50, 312.81, and 217.60, are obtained. Subsequently, the target SEM image is obtained by shifting the original SEM image in the target direction by the corresponding periods according to the three aforementioned period values.

[0058] Figure 8 is an exemplary schematic diagram showing an original SEM image and a corresponding target SEM image according to an embodiment of the present application. Figure 8 The original SEM image is shown in the upper middle. Taking the horizontal direction and the period value 500.50 as an example, the original SEM image is moved horizontally to the right by one period (500.50) to obtain Figure 8 The target SEM image is shown in the lower center. Similarly, a target SEM image corresponding to the corresponding period in the target direction can be obtained. Based on this target SEM image and the original SEM image, the final period value in the target direction can be determined by calculating their SAD image and mean, and combining it with a preset mean.

[0059] Figure 9 : is a SAD image showing the target SEM image and the original SEM image corresponding to three periods in the horizontal direction according to an embodiment of the present application. Figure 9 Shown from top to bottom are the SAD images of the target SEM image and the original SEM image corresponding to the horizontal shift of the original SEM image to the right by a period of 500.50, the SAD images of the target SEM image and the original SEM image corresponding to the horizontal shift of the original SEM image to the right by a period of 312.81, and the SAD images of the target SEM image and the original SEM image corresponding to the horizontal shift of the original SEM image to the right by a period of 217.60. In an implementation scenario, the first cropped image and the second cropped image can be obtained by first cropping the target SEM image and the original SEM image after each cycle according to the target width (e.g., ceil(P):N), respectively calculating the absolute value result and the sum result of the difference between the first cropped image and the second cropped image, and then dividing the absolute value result of the difference by the sum result to obtain the dot division result to obtain the SAD image. For more details about the SAD image calculation, please refer to the above Figure 1 、 Figure 3 The described contents will not be repeated in detail in the embodiments of this application.

[0060] Based on the SAD image obtained under the corresponding period, the mean value is obtained by calculating the average value of the pixel grayscale in the SAD image. Furthermore, the minimum mean value and the period value of the corresponding SAD image are determined from the mean values ​​of the SAD images obtained corresponding to multiple periods, and the period value of the SAD image corresponding to the minimum mean value that is less than the preset mean value is determined as the final period value of the scanning electron microscope image in the target direction. Through the solution of the present application, a period value with higher accuracy and better stability can be obtained. When applied to scenarios such as wafer defect detection, the location of wafer defects can be made more accurate, greatly reducing the false detection rate of wafer defects.

[0061] Figure 10 FIG. 1 is an exemplary structural block diagram of a device 1000 for determining a scanning electron microscope image period according to an embodiment of the present application. Figure 10 As shown in , the device 1000 of the present application may include a processor 1001 and a memory 1002, wherein the processor 1001 and the memory 1002 communicate with each other via a bus. The memory 1002 stores program instructions for determining the period of a scanning electron microscope image. When the program instructions are executed by the processor 1001, the method steps described above in conjunction with the accompanying drawings are implemented: expanding the original scanning electron microscope image, and obtaining all period values ​​of the expanded original scanning electron microscope image in a target direction using Fourier transform; moving the original scanning electron microscope image along the target direction by one period based on the period value to obtain a corresponding target scanning electron microscope image; obtaining an absolute error and an image between the corresponding target scanning electron microscope image and the original scanning electron microscope image, determining the mean of the absolute error and the image, and determining the final period value of the scanning electron microscope image in the target direction based on the mean and a preset mean.

[0062] According to the above description in conjunction with the accompanying drawings, those skilled in the art will also understand that the embodiments of the present application can also be implemented by software programs. Therefore, the present application also provides a computer-readable storage medium. The computer-readable storage medium stores computer-readable instructions for determining the scanning electron microscope image period. When the computer-readable instructions are executed by one or more processors, the present application in conjunction with the accompanying drawings can be implemented. Figure 1 A method for determining the period of a scanning electron microscope image is described.

[0063] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0064] It should be noted that although the operations of the present method are described in a particular order in the accompanying drawings, this does not require or imply that the operations must be performed in that particular order, or that all of the operations shown must be performed to achieve the desired results. Rather, the steps depicted in the flowcharts may be performed in a different order. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into a single step, and / or a single step may be broken down into multiple steps.

[0065] It should be understood that when the terms "first," "second," "third," and "fourth," etc., are used in the claims, specification, and drawings of this application, they are only used to distinguish different objects, rather than to describe a specific order. The terms "comprise" and "comprising" used in the specification and claims of this application indicate the presence of the described features, integers, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or combinations thereof.

[0066] It should also be understood that the terms used in this specification are for the purpose of describing specific embodiments only and are not intended to limit this application. As used in this specification and claims, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should also be further understood that the term "and / or" as used in this specification and claims refers to any and all possible combinations of one or more of the associated listed items, including and including these combinations.

[0067] Although the implementation methods of this application are as described above, the contents are only examples adopted to facilitate understanding of this application and are not intended to limit the scope and application scenarios of this application. Any technician in the technical field described in this application can make any modifications and changes in the form and details of implementation without departing from the spirit and scope disclosed in this application, but the scope of patent protection of this application shall still be based on the scope defined by the attached claims.

Claims

1. A method for determining a scanning electron microscope image period, comprising: Expanding the original SEM image, and using Fourier transform to obtain all periodic values ​​of the expanded original SEM image in the target direction; Based on the period value, the original SEM image is moved along the target direction by one period to obtain a corresponding target SEM image; Obtaining an absolute error and an image between the corresponding target SEM image and the original SEM image, and determining a mean value of the absolute error and the image; as well as Determine the final period value of the scanning electron microscope image in the target direction according to the mean value and the preset mean value, Wherein obtaining the absolute error and image between the corresponding target SEM image and the original SEM image, and determining the mean of the absolute error and image comprises: According to the target width, the corresponding target SEM image and the original SEM image are respectively intercepted to obtain a corresponding first intercepted image and a second intercepted image; respectively calculating an absolute value result and a sum result of a difference between the first captured image and the second captured image; Dividing the absolute value result of the difference by the sum result to obtain a dot division result, so as to obtain the absolute error sum image; and Calculating the absolute error and the average value of the grayscale of pixels in the image to determine the mean value of the absolute error and the image; The target direction includes a horizontal direction and / or a vertical direction, and determining a final period value of the scanning electron microscope image in the target direction according to the mean value and a preset mean value includes: determining the minimum mean of the absolute errors and images; and In response to the minimum mean being less than the preset mean, the absolute error of the minimum mean and the period value corresponding to the image are determined as the final period value of the scanning electron microscope image in the horizontal direction and / or the vertical direction.

2. The method according to claim 1, wherein the original SEM image is expanded by the following operations: Projecting the original SEM image along a column direction and calculating a mean value to generate a one-dimensional mean matrix; and A plurality of columns of extension elements are added to the one-dimensional mean matrix to obtain an extension matrix, so as to expand the original scanning electron microscope image.

3. The method according to claim 2, wherein the method of obtaining all periodic values ​​of the expanded original SEM image in the target direction by Fourier transform comprises: Performing a one-dimensional Fourier transform on the expanded matrix to obtain a one-dimensional frequency spectrum matrix; In response to satisfying the minimum peak spacing and the minimum peak value, determining the position of the target peak point in the one-dimensional spectrum matrix; as well as All periodic values ​​of the expanded original SEM image in the target direction are obtained according to the number of columns of the expanded matrix and the position of the target peak point.

4. The method according to claim 3, wherein performing a one-dimensional Fourier transform on the expanded matrix to obtain a one-dimensional spectrum matrix comprises: Performing a one-dimensional Fourier transform on the expanded matrix to obtain a frequency spectrum; Moving the zero-frequency component in the spectrum diagram to the center of the spectrum; as well as The right half of the center of the spectrum is selected and the absolute value is calculated to obtain the one-dimensional spectrum matrix.

5. The method according to claim 1 or 3, wherein the period value comprises a period value in a horizontal direction, a period value in a vertical direction, and a combined period value of the period value in the horizontal direction and the period value in the vertical direction, and moving the original SEM image along the target direction by one period based on the period value to obtain a corresponding target SEM image comprises: Based on the period value in the horizontal direction or the period value in the vertical direction, the original scanning electron microscope image is moved along the horizontal direction or the vertical direction by one period to obtain a corresponding target scanning electron microscope image; or The original scanning electron microscope image is moved by one period in the horizontal direction and the vertical direction respectively based on the combined period value to obtain a corresponding target scanning electron microscope image.

6. The method of claim 1, wherein the target width is determined by a corresponding period value and a width of the original SEM image.

7. A device for determining the period of a scanning electron microscope image, comprising: processor; as well as A memory storing program instructions for determining a scanning electron microscope image period, wherein when the program instructions are executed by the processor, the device implements the method according to any one of claims 1 to 6.

8. A computer-readable storage medium having stored thereon computer-readable instructions for determining a scanning electron microscope image period, wherein when the computer-readable instructions are executed by one or more processors, the method according to any one of claims 1 to 6 is implemented.

Images