Metrology verification method, computer device and computer readable storage medium

Abstract

This disclosure provides a measurement verification method, computer device, and computer-readable storage medium, belonging to the field of semiconductor measurement technology. The method includes: obtaining a measurement image and a standard image of a semiconductor structure, the measurement image including measurement feature size boundaries; processing the measurement image and the standard image respectively to obtain a first measurement signal boundary in the measurement image and a second measurement signal boundary in the standard image; obtaining first contrast information of the first measurement signal boundary and second contrast information of the second measurement signal boundary; and obtaining a matching index of the first measurement signal boundary and the second measurement signal boundary based on the first contrast information and the second contrast information, the matching index indicating the degree of matching between the measurement feature size boundary and the first measurement signal boundary.

Description

Technical Field

[0001] This disclosure relates to the field of semiconductor measurement technology, and more specifically, to a measurement verification method, a computer device, and a computer-readable storage medium. Background Technology

[0002] Critical Dimension (CD) refers to a specific line pattern designed to reflect the width of integrated circuit feature lines in photomask manufacturing and lithography processes, used to evaluate and control the patterning accuracy of the process. In addition, the critical dimension can also be defined as the width of photoresist trenches or lines obtained at a specific exposure intensity threshold. Similarly, since a lithographic spatial image provides relative intensity, the critical dimension of the lithographic spatial image can be defined as the width of the pattern obtained at a specific relative intensity threshold. Currently, there is no verification method in related technologies to verify the accuracy of the measured CD. Summary of the Invention

[0003] The purpose of this disclosure is to provide a measurement verification method, computer device, and computer-readable storage medium capable of verifying the accuracy of measurements.

[0004] This disclosure provides a measurement verification method, which includes: obtaining a measurement image of a semiconductor structure and a standard image, the measurement image including measurement feature size boundaries; processing the measurement image and the standard image respectively to obtain a first measurement signal boundary in the measurement image and a second measurement signal boundary in the standard image; obtaining first contrast information of the first measurement signal boundary and second contrast information of the second measurement signal boundary; and obtaining a matching index of the first measurement signal boundary and the second measurement signal boundary based on the first contrast information and the second contrast information, the matching index indicating the degree of matching between the measurement feature size boundary and the first measurement signal boundary.

[0005] According to another aspect of this disclosure, a measurement verification apparatus is provided, the apparatus comprising: an acquisition unit for acquiring a measurement image of a semiconductor structure and a standard image, the measurement image including measurement feature size boundaries; a processing unit for processing the measurement image and the standard image respectively to acquire a first measurement signal boundary in the measurement image and a second measurement signal boundary in the standard image; the acquisition unit further for acquiring first contrast information of the first measurement signal boundary and second contrast information of the second measurement signal boundary; and the acquisition unit further for acquiring a matching index of the first measurement signal boundary and the second measurement signal boundary based on the first contrast information and the second contrast information, the matching index indicating the degree of matching between the measurement feature size boundary and the first measurement signal boundary.

[0006] According to another aspect of this disclosure, a computer device is provided, including one or more processors; and a memory configured to store one or more programs, which, when executed by the one or more processors, cause the computer device to implement the measurement verification method of any embodiment of this disclosure.

[0007] According to another aspect of this disclosure, a computer-readable storage medium is provided that stores a computer program adapted to be loaded and executed by a processor, such that a computer device having the processor performs the measurement verification method of any embodiment of this disclosure.

[0008] According to another aspect of this disclosure, a computer program product is provided that, when executed by a processor, implements the measurement verification method in any embodiment of this disclosure. Attached Figure Description

[0009] Figure 1 A flowchart of a measurement verification method provided in an embodiment of this disclosure is shown.

[0010] Figure 2 A schematic diagram showing a standard image provided in an embodiment of this disclosure is illustrated.

[0011] Figure 3 A schematic diagram of a measurement image provided in an embodiment of this disclosure is shown.

[0012] Figure 4 A flowchart of another measurement verification method provided by an embodiment of this disclosure is shown.

[0013] Figure 5 It shows Figure 4 A flowchart of S440 in an exemplary embodiment.

[0014] Figure 6 A schematic diagram of another measurement image provided by an embodiment of this disclosure is shown.

[0015] Figure 7 It shows the Figure 6 The diagram shows a process of processing the measurement image to obtain the boundary of the first measurement signal.

[0016] Figure 8 It shows the basis Figure 7 A schematic diagram of obtaining the first contrast information by the boundary of the first measurement signal.

[0017] Figure 9 This illustration shows a schematic diagram of obtaining the degree of matching based on first contrast information and second contrast information, according to an embodiment of the present disclosure.

[0018] Figure 10 It shows Figure 4 A flowchart of S440 in another exemplary embodiment.

[0019] Figure 11 A schematic diagram of obtaining the boundary of a first measurement signal based on a gradient image provided in an embodiment of this disclosure is shown.

[0020] Figure 12 It shows the basis Figure 11 A schematic diagram of obtaining the first contrast information from the mapped image 800.

[0021] Figure 13 A schematic diagram of the second contrast information provided in an embodiment of this disclosure is shown.

[0022] Figure 14 This illustration shows another schematic diagram of obtaining the degree of matching based on first contrast information and second contrast information, provided by an embodiment of this disclosure.

[0023] Figure 15 A schematic diagram illustrating the acquisition of a first boundary matching index and a second boundary matching index provided in an embodiment of this disclosure is shown.

[0024] Figure 16 A schematic diagram of an image pixel provided in an embodiment of this disclosure is shown.

[0025] Figure 17 A schematic diagram of the structure of a measurement and verification device provided in an embodiment of this disclosure is shown.

[0026] Figure 18 A schematic diagram of the structure of a computer device provided in an embodiment of this disclosure is shown. Detailed Implementation

[0027] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that this disclosure will be more comprehensive and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0028] Figure 1 A flowchart of a measurement verification method provided by an embodiment of this disclosure is shown. Figure 1 As shown, the method provided in this disclosure embodiment may include the following steps.

[0029] In S110, a measurement image and a standard image of the semiconductor structure are obtained, the measurement image including the boundary of the measurement feature size.

[0030] The semiconductor structure in the embodiments of this disclosure can be any semiconductor device or part or all of the structure used to manufacture a semiconductor device. For example, the semiconductor device can be DRAM (Dynamic Random Access Memory), but this disclosure is not limited thereto. In the following embodiments, the semiconductor structure is illustrated as a wafer, but this disclosure is not limited thereto. The measurement image of the semiconductor structure refers to the raw image obtained by scanning the semiconductor structure using a measurement instrument. For example, the measurement instrument is illustrated as a CD-SEM (scanning electron microscope), but this disclosure is not limited thereto. The CD-SEM instrument forms the measurement image by using a high-energy electron beam to interact with the surface of the semiconductor structure to excite secondary electrons, and then collecting the secondary electron signals to form the measurement image.

[0031] In this embodiment of the disclosure, the measurement image includes the boundary of the measurement feature size. The boundary of the measurement feature size refers to the actual or true boundary of the feature size (CD) to be measured in the measurement image obtained by scanning with a measurement instrument. For example, assuming there is a groove in the semiconductor structure, and the feature size of the groove needs to be measured, when the CD-SEM instrument scans the semiconductor structure, it will display the edge of the groove on the measurement image. The width of the two sides of the groove is the actual feature size in the measurement image, that is, the opening size of the groove.

[0032] In this embodiment of the disclosure, a standard image (golden image) of the semiconductor structure is pre-stored in the recipe. The golden image refers to the designed image of the semiconductor structure that is desired to be achieved during its fabrication. This standard image contains standard feature size boundaries. Standard feature size boundaries refer to the actual or true boundaries of the feature dimensions to be measured in the standard image.

[0033] In S120, the measurement image and the standard image are processed respectively to obtain the first measurement signal boundary in the measurement image and the second measurement signal boundary in the standard image.

[0034] In this embodiment of the disclosure, the pixel values ​​of the measurement feature size boundary in the measurement image differ from the pixel values ​​of other areas in the measurement image. For example, the pixel values ​​of the measurement feature size boundary are relatively large, appearing as white or near-white areas; while the pixel values ​​of other areas are relatively small, appearing as black or gray areas. Therefore, by utilizing this feature of the measurement feature size boundary in the measurement image, a first measurement signal boundary can be obtained by performing image processing on the measurement image. The first measurement signal boundary refers to the boundary of the feature size to be detected in the measurement image through image processing; it may coincide with the measurement feature size boundary or may deviate to some extent.

[0035] In this embodiment of the disclosure, the pixel values ​​of the standard feature size boundary in the standard image are different from the pixel values ​​of other areas in the standard image. Therefore, by using this feature of the standard feature size boundary in the standard image, a second measurement signal boundary can be obtained by performing image processing on the standard image. The second measurement signal boundary refers to the boundary of the feature size to be detected in the standard image through image processing, and it generally coincides with the standard feature size boundary.

[0036] In S130, the first contrast information of the boundary of the first measurement signal and the second contrast information of the boundary of the second measurement signal are obtained.

[0037] In this embodiment of the disclosure, the first contrast information may include the contrast value of the pixel at the boundary position of the first measurement signal, which will be referred to as the first contrast value in the following embodiments for distinction. Based on the first contrast value of each pixel at the boundary position of the first measurement signal, the number of pixels with each first contrast value on the boundary of the first measurement signal can be counted, that is, the number of pixels with the same first contrast value. In some embodiments, the first contrast information may further include the correspondence between the first contrast value and its corresponding number of pixels.

[0038] In this embodiment of the disclosure, the second contrast information may include the contrast values ​​of pixels at the boundary positions of the second measurement signal, which will be referred to as second contrast values ​​in the following embodiments for distinction. Based on the second contrast values ​​of each pixel at the boundary positions of the second measurement signal, the number of pixels with each second contrast value on the boundary of the second measurement signal can be counted, i.e., the number of pixels with the same second contrast value. In some embodiments, the second contrast information may further include the correspondence between the second contrast values ​​and their corresponding number of pixels.

[0039] In S140, a matching index of the first measurement signal boundary and the second measurement signal boundary is obtained based on the first contrast information and the second contrast information. The matching index is used to indicate the degree of matching between the measurement feature size boundary and the first measurement signal boundary.

[0040] In this embodiment of the disclosure, by calculating the correlation between the first contrast information and the second contrast information, since the second contrast information comes from the second measurement signal boundary of the standard image, and the second measurement signal boundary coincides with the standard feature size boundary in the standard image, the correlation between the first contrast information and the second contrast information can be used to obtain a matching index between the first measurement signal boundary fitted in the measurement image and the second measurement signal boundary fitted in the standard image. This matching index can indicate the degree of matching between the actual measurement feature size boundary in the measurement image and the fitted first measurement signal boundary.

[0041] In an exemplary embodiment, the method provided in this disclosure further includes: obtaining the measurement feature size of the semiconductor pattern in the semiconductor structure based on the first measurement signal boundary.

[0042] In this embodiment of the disclosure, after obtaining the first measurement signal boundary, the measurement feature dimension of the semiconductor pattern in the measured semiconductor structure can be obtained based on the width between the left and right sides of the first measurement signal boundary. For example, if the semiconductor pattern is a groove, the measurement CD of the groove is obtained. Based on the matching index obtained above, the accuracy of the measurement CD can be obtained.

[0043] The measurement verification method provided in this disclosure, on the one hand, obtains a matching index between the first measurement signal boundary and the second measurement signal boundary by using the first contrast information of the first measurement signal boundary of the measurement image and the second contrast information of the second measurement signal boundary of the standard image. This matching index can then be used to determine the degree of matching between the first measurement signal boundary detected from the measurement image and the actual measurement feature size boundary, thereby detecting the accuracy of the measurement CD measured based on the first measurement signal boundary. On the other hand, when detecting the accuracy of the measurement CD, only the first contrast information of the first measurement signal boundary and the second contrast information of the second measurement signal boundary need to be calculated, resulting in a small computational load. Furthermore, the calculation of the matching index is independent of the size of the CD to be detected.

[0044] like Figure 2 As shown, the black portion in the standard image (golden image) 200 represents the groove 250, meaning it is assumed here that the semiconductor pattern to be detected is a groove. The white dots represent the boundary of the groove 250, i.e., the standard feature size boundary 220. Here, it is assumed that the standard feature size boundary 220 extends in the second direction (Y-axis) and has one on each side. When measuring the feature size of the groove 250, the measurement range is first determined. Here, it is assumed that a measurement frame 210 is used to determine the measurement range, thereby reducing the measurement range, reducing the computational load, and improving the detection accuracy. By calculating the contrast value of each pixel within the measurement frame 210, a pattern can be formed as shown below. Figure 2The contrast signal 230 shown extends along the first direction (X-axis) and is wavy due to the different contrast values ​​of pixels in the same row. The white pixels at the standard feature size boundary 220 in the standard image 200 have higher pixel values, and therefore larger contrast values. As can be seen in the figure, the contrast signal 230 exhibits two peaks at the standard feature size boundaries 220 on both the left and right sides. These two peaks identify two pixels in the row that approximately fall on the standard feature size boundaries 220 on both sides. Similarly, the contrast signal for each row and the two pixels in each row can be obtained. Then, by connecting pixels with approximately or identical X-axis values ​​along the Y-axis direction, the second measurement signal boundary 240 can be obtained. Figure 2 (As shown by the two dashed lines in the middle). By calculating the second measurement signal boundary 240 on the left and right sides, the standard feature size 270 can be obtained. Figure 2 The gray area 260 represents other structures within the semiconductor structure. From... Figure 2 As can be seen, the boundary 240 of the second measurement signal measured in the standard image 200 almost coincides with the actual standard feature size boundary 220. Therefore, the standard feature size 270 measured through the standard image 200 is also consistent with or approximately consistent with the actual CD of the groove 250. That is, the accurate boundary of the second measurement signal can be captured through the standard image, i.e., the dotted line is exactly aligned with the white position of the standard image. The white position is the boundary of the CD to be measured. The white is the effect captured by the CD-SEM machine, which is the actual edge of the CD, such as the edge of the groove.

[0045] like Figure 3 As shown, the black portion in the measurement image (raw image, hereinafter also referred to as the original image) 300 represents the groove. The white dots represent the boundary of the groove, i.e., the measurement feature size boundary 320. The standard feature size boundary 320 verifies the extension in the second direction (Y-axis), with one boundary on each of the left and right sides. When measuring the feature size of the groove, the measurement range is first determined. Here, it is assumed that a measurement frame 310 is used to determine the measurement range, thereby narrowing the measurement range, reducing the computational load, and improving the detection accuracy. By calculating the contrast value of each pixel within the measurement frame 310, a shape can be formed as shown below. Figure 3The contrast signal 330 shown extends along the first direction (X-axis) and is wavy due to the different contrast values ​​of pixels in the same row. The white pixels at the measurement feature size boundary 320 in the measurement image 300 have higher pixel values, and therefore larger contrast values. As can be seen in the figure, the contrast signal 330 exhibits two peaks at the measurement feature size boundaries 320 on both the left and right sides. These two peaks can be used to identify two pixels in that row. Due to noise in image processing, these two pixels do not fall on the measurement feature size boundaries 320 on either side. Similarly, the contrast signal for each row and the two pixels in each row can be obtained. Then, by connecting pixels with approximately or identical X-axis values ​​along the Y-axis direction, the first measurement signal boundary 340 can be obtained. Figure 3 (As shown by the two dashed lines in the middle). By calculating the first measurement signal boundary 340 on the left and right sides, the measurement feature size 370 can be obtained. Figure 3 The gray area 360 represents other structures within the semiconductor structure. From... Figure 3 It can be seen that the first measurement signal boundary 340 measured in measurement image 300 does not coincide with the actual measurement feature size boundary 320, and there is a certain deviation. That is, the phenomenon of the first measurement signal boundary 340 being captured incorrectly occurs, and the dotted line exceeds the white position. Therefore, the measurement feature size 370 measured by measurement image 300 also deviates from the actual CD of the groove, causing the CD measurement to be incorrect. Here, the measurement feature size 370 is larger than the actual feature size.

[0046] according to Figure 3 It is known that due to equipment problems, inconsistencies between the actual scanned measurement pattern and the golden image set in the recipe, the measurement CD boundary in the measurement image does not match the first measurement signal boundary, ultimately resulting in an incorrect CD value. In related technologies, there is no index to characterize the degree of matching between the measurement CD boundary and the first measurement signal boundary in the measurement image; it can only be judged manually. The method provided in this disclosure achieves the purpose of detecting the quality of CD measurements by a CD-SEM machine by obtaining a matching index between the first and second measurement signal boundaries.

[0047] In an exemplary embodiment, obtaining a measurement image of a semiconductor structure includes: setting a measurement mode of a feature-size scanning electron microscope; if the measurement image including the boundary of the measured feature size is not obtained using the feature-size scanning electron microscope under the measurement mode, then changing the measurement mode of the feature-size scanning electron microscope; and obtaining the measurement image of the boundary of the measured feature size using the changed measurement mode.

[0048] like Figure 4As shown, the method provided in this disclosure embodiment may include the following steps.

[0049] In S410, the wafer enters the CD-SEM machine for measurement. Here, it is assumed that the wafer is the semiconductor structure to be inspected, and the measurement machine is illustrated using a CD-SEM machine.

[0050] In S420, it is determined whether a measurement image has been acquired; if no measurement image has been acquired, S430 is executed; if a measurement image has been acquired, S440 is executed.

[0051] Determine whether the CD-SEM machine can scan the wafer to generate a measurement image that includes the boundary of the measurement feature size. In other words, the CD-SEM machine can obtain the required measurement image and data for calculating the matching index.

[0052] In S430, measurement processing (MET (Metrology) handle) is performed. Then, the process jumps back to S410 above for further processing.

[0053] In this embodiment of the disclosure, MET refers to the result measurement after the semiconductor process is completed, and the data is usually generated by a measurement instrument. If the required measurement image cannot be obtained, the measurement method is changed, such as changing any one or more of the measurement site, measurement direction, or value change, until the required measurement image can be obtained for the next step of correlation calculation.

[0054] In S440, CD (i.e., measured feature dimension) and GOF (i.e., matching index) are calculated, and it is determined whether GOF conforms to the specification. If GOF does not conform to the specification, then S430 is executed; if GOF conforms to the specification, then S450 is executed.

[0055] In the S450, CD and GOF are recorded into the system.

[0056] In this embodiment, the GOF (God of Optimal Value) is used to determine the accuracy of CD-SEM measurements. If the GOF value is close to 0, it indicates that the CD measurement is inaccurate; if the GOF value is close to 1, it indicates that the CD measurement is accurate. That is, the GOF value represents the correlation or correlation degree between the measured image and the standard image. If the GOF value is incorrect (too low), the measurement method is changed again, for example, the measurement range defined by the measurement frame, until a greater correlation is obtained. The measured CD value at this point is then uploaded to the system.

[0057] In this embodiment, if the GOF value is less than the set spec, the GOF value is determined to be too low; if the GOF value is greater than or equal to the set spec, a high correlation is determined to be obtained. The spec can be set by collecting historical data of semiconductor patterns, and different specs can be set for different semiconductor patterns. This disclosure does not limit this.

[0058] In S460, the normal process is executed.

[0059] In this embodiment of the disclosure, by setting spec to monitor and manage GOF values, the frequency of manual inspection of measurement fitting problems can be reduced, saving labor costs and improving verification accuracy and efficiency.

[0060] In an exemplary embodiment, processing the measurement image to obtain a first measurement signal boundary in the measurement image includes: obtaining contrast information in the measurement image; and determining the first measurement signal boundary in the measurement image based on the contrast information in the measurement image.

[0061] In an exemplary embodiment, obtaining contrast information in the measurement image includes: accumulating the pixel values ​​of each pixel in the same column of the measurement image along a second direction to obtain the contrast value of each pixel, wherein the second direction is the extension direction of the boundary of the measurement feature size; drawing a contrast signal for each row based on the contrast values ​​of each pixel in the same row along a first direction, wherein the first direction is perpendicular to the second direction, and the contrast information includes the contrast signal.

[0062] The step of determining the first measurement signal boundary in the measurement image based on the contrast information in the measurement image includes: obtaining the maximum contrast value in the contrast signal of each row; determining the contrast threshold of the contrast signal of the corresponding row based on the maximum contrast value; determining the target pixel in the contrast signal of the corresponding row based on the contrast threshold; and connecting the target pixel along the second direction to form the first measurement signal boundary.

[0063] Contrast refers to the difference in brightness between different areas of an image. In this embodiment of the disclosure, the contrast information in the measurement image is obtained by calculating the superposition of the pixel values ​​of each pixel in the measurement image in the second direction (Y-axis). That is, T(x) = t(x,1) + t(x,2) + t(x,3) + ... + t(x,n), where T(x) represents the contrast value of the pixel in the column X = x on the X-axis of the measurement image, t(x,1) represents the pixel value of pixel (x,1), t(x,2) represents the pixel value of pixel (x,2), t(x,3) represents the pixel value of pixel (x,3), and t(x,n) represents the pixel value of pixel (x,n). x and n are both positive integers greater than or equal to 1, and the maximum value of x and n depends on the number of pixels in the measurement range. The contrast value of pixel (x,1) is equal to the pixel value of pixel (x,1), the contrast value of pixel (x,2) is equal to the sum of the pixel values ​​of pixel (x,1) and pixel (x,2), and the contrast value of pixel (x,3) is equal to the sum of the pixel values ​​of pixel (x,1) and pixel (x,2) and pixel (x,3). In an exemplary embodiment, the contrast values ​​of each pixel can also be normalized so that the contrast value range is between [-1,1]. After calculating the contrast value of each pixel, the contrast values ​​in the same row can be plotted as a wavy line according to their contrast value magnitude, thereby forming multiple contrast signals extending along the first direction (X-axis). For each contrast signal, the maximum contrast value can be found, and a predetermined percentage of the maximum contrast value, such as 80% (this is only for illustration and can be set according to actual needs, such as 70%, 90%, 95%, etc.), is determined as the contrast threshold. Then select the pixel points in each row whose contrast value is greater than the contrast threshold as the target pixel points. Connect the target pixel points in the same column or approximately the same column along the Y-axis to form the first measurement signal boundary on the left and right sides, which can be used for subsequent calculation of measurement CD.

[0064] In an exemplary embodiment, obtaining first contrast information of the boundary of the first measurement signal includes: extracting first contrast values ​​of each pixel on the boundary of the first measurement signal; obtaining the number of pixels on the boundary of the first measurement signal under the same first contrast value; establishing a first coordinate based on the first contrast value; establishing a second coordinate based on the number of pixels; wherein the first contrast information includes the correspondence between the number of pixels and the first contrast value in the first coordinate and the second coordinate.

[0065] like Figure 5 As shown, the above S440 may further include the following steps.

[0066] In S441a, the CD-SEM machine calculates the contrast of the original image.

[0067] The CD-SEM machine processes the measurement image and calculates the contrast value of each pixel. The magnitude of the contrast value reflects the brightness and darkness in the measurement image; the smaller the contrast value of black areas, the larger the contrast value of white areas.

[0068] In S442a, the first measurement signal boundary is generated based on the contrast.

[0069] Since the first measurement signal boundary has a large pixel value, or a large contrast value, the CD-SEM machine draws two first measurement signal boundaries on the measurement image based on the contrast value of each pixel in the measurement image.

[0070] In S443a, the first contrast value of the boundary position of the first measurement signal is extracted.

[0071] That is, extract the contrast value of each pixel at the boundary position of the first measurement signal, which is called the first contrast value, and remove the contrast values ​​of pixels in other areas of the measurement image.

[0072] In S444a, the first contrast value at the boundary position of the first measurement signal is converted into a digital signal.

[0073] For example, the boundary of the first measurement signal can be uniformly sampled to obtain the digitized first contrast value and its corresponding pixel.

[0074] In S445a, the correlation coefficient, or GOF, is calculated.

[0075] Based on the digitized first contrast value and its corresponding pixel, the correspondence between the first contrast value and its corresponding pixel count can be obtained, which can be used as the first contrast information.

[0076] In this embodiment, a similar process is performed on the standard image: the CD-SEM machine processes the standard image and calculates the contrast value of each pixel. The CD-SEM machine draws two second measurement signal boundaries on the standard image based on the contrast values ​​of each pixel. The contrast values ​​of each pixel at the second measurement signal boundary positions are extracted and referred to as the second contrast values. The contrast values ​​of pixels in other areas of the standard image are then discarded. Based on the digitized second contrast values ​​and their corresponding pixels, the correspondence between the second contrast values ​​and their corresponding pixel counts can be obtained as second contrast information. Then, the correlation between the first and second contrast information is calculated as the GOF value. The GOF value in this embodiment is a relative GOF value, not an absolute GOF value.

[0077] For example, such as Figure 6 First, obtain a measurement image of 600. Then, as... Figure 7 As shown, the measurement frame 610 is determined, and then processed to obtain the first measurement signal boundary 630. The CD-SEM instrument calculates the first contrast value at the position of the first measurement signal boundary 630, as shown below. Figure 8 The relationship between the number of pixels and the first contrast value shown is expressed as the number of pixels - contrast value 1. Figure 9 (represented by d1 in Chinese). For example... Figure 9 As shown, the correspondence between the number of pixels and the second contrast value (represented as the number of pixels - contrast value 2, denoted by d2) is also plotted in the same coordinate system for calculating the GOF value.

[0078] Figure 8 and Figure 9 In the diagram, the horizontal axis (the first coordinate) represents the contrast value (which includes both the first and second contrast values), and the vertical axis represents the number of pixels.

[0079] In an exemplary embodiment, processing the measurement image to obtain a first measurement signal boundary in the measurement image includes: generating a contrast image including an initial measurement signal boundary based on the measurement image; generating a gradient image of the measurement image, the gradient image containing gradient values ​​of each pixel; generating a mapping image based on the contrast image and the gradient image, the mapping image containing an initial measurement signal boundary with a gradient value not equal to 0, and using the initial measurement signal boundary with a gradient value not equal to 0 as the first measurement signal boundary.

[0080] In this embodiment of the disclosure, the image gradient refers to the directional change of image intensity or color. The image can be viewed as a two-dimensional discrete function, where each pixel represents a value. The image gradient is simply the derivative of this two-dimensional discrete function.

[0081] G(m, n)=dx(m, n) i + dy(m, n) j (1)

[0082] dx(m, n)=I(m+1, n)- I(m, n) (2)

[0083] dy(m, n)=I(m, n+1)- I(m, n) (3)

[0084] Where G(m, n) represents the gradient value of pixel (m, n) in the measured image; m and n are both positive integers greater than or equal to 1, where m represents the x-coordinate of the pixel and n represents the y-coordinate; dx(m, n) represents the gradient value of pixel (m, n) along the x-axis; dy(m, n) represents the gradient value of pixel (m, n) along the y-axis; and I(m, n) is the pixel value at position (m, n) in the image. i and j represent the first direction indicated by the x-axis and the second direction indicated by the y-axis, respectively, meaning the gradient value G(m, n) is a vector. The gradient is the derivative of the contrast.

[0085] In an exemplary embodiment, the first contrast information includes the correspondence between the first contrast value of each pixel on the boundary of the first measurement signal and the number of pixels thereon; the second contrast information includes the correspondence between the second contrast value of each pixel on the boundary of the second measurement signal and the number of pixels thereon. The matching index includes a first boundary matching index and a second boundary matching index.

[0086] The method of obtaining a matching index for the boundary of the first measurement signal and the boundary of the second measurement signal based on the first contrast information and the second contrast information includes: dividing the first contrast information into first boundary measurement contrast information and second boundary measurement contrast information based on whether the first contrast value is 0; dividing the second contrast information into first boundary standard contrast information and second boundary standard contrast information based on whether the second contrast value is 0; obtaining the first boundary matching index based on the first boundary measurement contrast information and the first boundary standard contrast information; and obtaining the second boundary matching index based on the second boundary measurement contrast information and the second boundary standard contrast information.

[0087] Figure 10 It shows Figure 4 A flowchart of S440 in another exemplary embodiment.

[0088] like Figure 10 As shown, the above S440 may further include the following steps.

[0089] In S441b, the CD-SEM machine calculates the contrast of the original image.

[0090] For example, such as Figure 11 As shown, the CD-SEM machine processes the measurement image 600 and calculates the contrast value of each pixel. The CD-SEM machine first determines the measurement frame 610 in the measurement image 600, and then draws two initial measurement signal boundaries 630 on the measurement image based on the contrast value of each pixel in the measurement frame 610. At this time, the measurement image can be named the contrast image.

[0091] In S442b, image gradients are calculated.

[0092] For example, such as Figure 11 As shown, the CD-SEM machine calculates the gradient values ​​of each pixel in the measurement image 600 to form a gradient image 700.

[0093] In S443b, the initial measurement signal boundary is mapped onto the gradient image to generate a mapped image.

[0094] like Figure 11 As shown, the contrast image and gradient image 700 are combined to generate a mapped image 800. In this embodiment, since the boundary of the measurement feature size is the intersection of white, black, and gray, its pixel value changes, or its contrast value changes. Therefore, the gradient value at the boundary of the measurement feature size is not 0, while the gradient value of the black area (groove) and gray area, where the pixel value or contrast value does not change, is 0. Therefore, a feature size boundary can also be detected by whether the gradient value is 0. If the feature size boundary found by the gradient value coincides with the initial measurement signal boundary found by the contrast value, it indicates that the initial measurement signal boundary is accurately located; if the feature size boundary found by the gradient value does not coincide with the initial measurement signal boundary found by the contrast value, it indicates that the initial measurement signal boundary is incorrectly located. When the location is incorrect, the measurement method can be changed again until the feature size boundary found by the gradient value coincides with the initial measurement signal boundary found by the contrast value. That is, this embodiment can use the gradient image to verify the accuracy of the initial measurement signal boundary location, thereby improving the accuracy of the first measurement signal boundary.

[0095] Figure 11 In the gradient image 700, it is only necessary to calculate the gradient value in the second direction corresponding to the Y-axis, because it is assumed here that the measurement CD extends in the second direction corresponding to the Y-axis.

[0096] In other embodiments, only the initial measurement signal boundaries with non-zero gradient values ​​can be used as the first measurement signal boundaries, while the initial measurement signal boundaries with zero gradient values ​​can be discarded.

[0097] In S444b, the mapped image is converted into a digital signal.

[0098] For example, the first contrast value on the boundary of the first measurement signal can be sampled uniformly.

[0099] In S445b, the correlation coefficient, or GOF, is calculated.

[0100] In this embodiment of the disclosure, similar processing is performed on the standard image: the CD-SEM machine processes the standard image and calculates the contrast value of each pixel. Based on the contrast values ​​of each pixel in the standard image, the CD-SEM machine draws two initial second measurement signal boundaries on the standard image. The CD-SEM machine calculates the gradient value of each pixel on the standard image. Only the initial second measurement signal boundaries with non-zero gradient values ​​are used as the second measurement signal boundaries, and the initial second measurement signal boundaries with zero gradient values ​​are discarded. The second contrast values ​​on the second measurement signal boundaries are sampled uniformly.

[0101] Obtain the first contrast value on the boundary of the first measurement signal, and remove the contrast values ​​of pixels in other areas of the measurement image; obtain the second contrast value on the boundary of the second measurement signal, and remove the contrast values ​​of pixels in other areas of the standard image to calculate the GOF value.

[0102] For example, the GOF value can be calculated using the following formula in the embodiments of this disclosure:

[0103]

[0104] In the above formula, z k This represents the number of pixels corresponding to the k-th first contrast value on the boundary of the first measurement signal, where k is a positive integer greater than or equal to 1 and less than or equal to N, N is a positive integer greater than or equal to 1, and N is the number of first contrast values ​​on the boundary of the first measurement signal. Here, it is assumed that the number of second contrast values ​​on the boundary of the second measurement signal is also N. h represents the average number of pixels corresponding to the N first contrast values; k This represents the number of pixels corresponding to the k-th second contrast value on the boundary of the second measurement signal; This represents the average number of pixels corresponding to the N second contrast values.

[0105] Because gradients have directionality, the GOF value of a boundary can be calculated separately.

[0106] For example, such as Figure 12 The figure shows a histogram between the first contrast value and its number of pixels. Figure 13 This is a histogram showing the relationship between the second contrast value and its pixel count. Combined with... Figure 12 and Figure 13 You can get Figure 14 , Figure 14 In the diagram, d1 represents the correspondence between the first contrast value and the number of pixels (pixels - contrast value 1), and d2 represents the correspondence between the second contrast value and the number of pixels (pixels - contrast value 2).

[0107] from Figure 15(b) It can be seen that the right boundary (first boundary) of a certain line in the mapped image (i.e., the boundary of the first measurement signal) has a contrast value of less than 0, which is a dark area; the left boundary (second boundary) has a contrast value of greater than 0, which is a bright area. That is, the left and right boundaries of a certain line can be distinguished based on the contrast value of 0 in the gradient image. Therefore, the GOF values ​​(0≤GOF value≤1) of the left and right boundaries can be calculated separately with a contrast value of 0 as the boundary. These are called the first boundary matching index and the second boundary matching index, respectively. Figure 15 (a) is the left boundary pixel-contrast value, i.e. the first boundary measurement contrast information. Figure 15 (c) represents the right boundary pixel-contrast value, i.e., the second boundary measurement contrast information. Similarly, the first boundary standard contrast information and the second boundary standard contrast information of the standard image can be obtained.

[0108] like Figure 16 As shown, the left side is measurement image 600, and the right side is a magnified schematic diagram of a portion 680 of the measurement image 600.

[0109] Based on the same inventive concept, this disclosure also provides a measurement and verification device, as described in the following embodiments. Since the principle by which this device solves the problem is similar to that of the method embodiments described above, the implementation of this device embodiment can refer to the implementation of the method embodiments described above, and repeated details will not be repeated.

[0110] Figure 17 A schematic diagram of a measurement and verification device according to an embodiment of this disclosure is shown. Figure 17 As shown, the measurement and verification device 1700 may include an acquisition unit 1710 and a processing unit 1720.

[0111] The acquisition unit 1710 is used to acquire a measurement image and a standard image of a semiconductor structure, wherein the measurement image includes measurement feature size boundaries. The processing unit 1720 is used to process the measurement image and the standard image respectively to obtain a first measurement signal boundary in the measurement image and a second measurement signal boundary in the standard image. The acquisition unit 1710 is further used to acquire first contrast information of the first measurement signal boundary and second contrast information of the second measurement signal boundary. The acquisition unit 1710 is further used to acquire a matching index between the first measurement signal boundary and the second measurement signal boundary based on the first contrast information and the second contrast information, wherein the matching index indicates the degree of matching between the measurement feature size boundary and the first measurement signal boundary.

[0112] In an exemplary embodiment, the obtaining unit 1710 is further configured to: set a measurement mode of the feature size scanning electron microscope; if the measurement image including the boundary of the measurement feature size is not obtained using the feature size scanning electron microscope under the measurement mode, then change the measurement mode of the feature size scanning electron microscope; and obtain the measurement image of the boundary of the measurement feature size using the changed measurement mode.

[0113] In an exemplary embodiment, the obtaining unit 1710 is further configured to: obtain contrast information in the measurement image; and determine the boundary of the first measurement signal in the measurement image based on the contrast information in the measurement image.

[0114] In an exemplary embodiment, the obtaining unit 1710 is further configured to: accumulate the pixel values ​​of each pixel in the same column of the measurement image along a second direction to obtain the contrast value of each pixel, wherein the second direction is the extension direction of the boundary of the measurement feature size; and draw the contrast signal of each row based on the contrast value of each pixel in the same row along a first direction, wherein the first direction is perpendicular to the second direction, and the contrast information includes the contrast signal.

[0115] The obtaining unit 1710 is further configured to: obtain the maximum contrast value in the contrast signal of each row; determine the contrast threshold of the contrast signal of the corresponding row according to the maximum contrast value; determine the target pixel in the contrast signal of the corresponding row according to the contrast threshold; and connect the target pixel along the second direction to form the first measurement signal boundary.

[0116] In an exemplary embodiment, the obtaining unit 1710 is further configured to: extract a first contrast value for each pixel on the boundary of the first measurement signal; obtain the number of pixels on the boundary of the first measurement signal under the same first contrast value; establish a first coordinate based on the first contrast value; establish a second coordinate based on the number of pixels; wherein the first contrast information is included in the first coordinate and the second coordinate, and the correspondence between the number of pixels and the first contrast value.

[0117] In an exemplary embodiment, the processing unit 1720 is further configured to: generate a contrast image including an initial measurement signal boundary based on the measurement image; generate a gradient image of the measurement image, the gradient image including gradient values ​​of each pixel; generate a mapping image based on the contrast image and the gradient image, the mapping image including an initial measurement signal boundary with a gradient value not equal to 0, and use the initial measurement signal boundary with a gradient value not equal to 0 as the first measurement signal boundary.

[0118] In an exemplary embodiment, the first contrast information includes the correspondence between the first contrast value of each pixel on the boundary of the first measurement signal and the number of pixels thereon; the second contrast information includes the correspondence between the second contrast value of each pixel on the boundary of the second measurement signal and the number of pixels thereon; the matching index includes a first boundary matching index and a second boundary matching index.

[0119] The obtaining unit 1710 is further configured to: divide the first contrast information into first boundary measurement contrast information and second boundary measurement contrast information according to whether the first contrast value is 0; divide the second contrast information into first boundary standard contrast information and second boundary standard contrast information according to whether the second contrast value is 0; obtain the first boundary matching index according to the first boundary measurement contrast information and the first boundary standard contrast information; and obtain the second boundary matching index according to the second boundary measurement contrast information and the second boundary standard contrast information.

[0120] In an exemplary embodiment, the obtaining unit 1710 is further configured to: obtain the measurement feature size of the semiconductor pattern in the semiconductor structure based on the first measurement signal boundary.

[0121] See Figure 18 , Figure 18 This is a schematic diagram of the structure of a computer device provided in an embodiment of this disclosure. Figure 18 As shown, the computer device in this embodiment may include one or more processors 1801, a memory 1802, and an input / output interface 1803. The processor 1801, memory 1802, and input / output interface 1803 are connected via a bus 1804. The memory 1802 stores a computer program, which includes program instructions. The input / output interface 1803 receives and outputs data, such as for data interaction between the host machine and the computer device, or for data interaction between various virtual machines within the host machine. The processor 1801 executes the program instructions stored in the memory 1802.

[0122] The processor 1801 can perform the following operations: obtaining a measurement image and a standard image of a semiconductor structure, wherein the measurement image includes measurement feature size boundaries; processing the measurement image and the standard image respectively to obtain a first measurement signal boundary in the measurement image and a second measurement signal boundary in the standard image; obtaining first contrast information of the first measurement signal boundary and second contrast information of the second measurement signal boundary; and obtaining a matching index of the first measurement signal boundary and the second measurement signal boundary based on the first contrast information and the second contrast information, wherein the matching index is used to indicate the degree of matching between the measurement feature size boundary and the first measurement signal boundary.

[0123] The memory 1802 may include read-only memory and random access memory, and provides instructions and data to the processor 1801 and input / output interface 1803. A portion of the memory 1802 may also include non-volatile random access memory. In specific implementations, the computer device can execute the implementation methods provided by the steps in any of the above method embodiments through its built-in functional modules. For details, please refer to the implementation methods provided by the steps in the figures shown in the above method embodiments, which will not be repeated here.

[0124] This disclosure provides a computer device including a processor, an input / output interface, and a memory. The processor retrieves a computer program from the memory and executes the steps of the method shown in any of the above embodiments.

[0125] This disclosure also provides a computer-readable storage medium storing a computer program that implements the methods described above. The computer program is adapted to be loaded by a processor and execute the measurement and verification methods provided in any of the above embodiments.

[0126] This disclosure also provides a computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in various alternative embodiments described above.

Claims

1. A metrology verification method, characterized by, include: Obtaining a measurement image and a standard image of a semiconductor structure, wherein the measurement image includes measurement feature size boundaries, and the standard image refers to an image of the designed semiconductor structure that is desired to be achieved during the manufacture of the semiconductor structure, the standard image having standard feature size boundaries, the process of obtaining the measurement image of the semiconductor structure specifically includes: setting the measurement mode of the feature size scanning electron microscope; if the measurement image including the measurement feature size boundaries is not obtained using the feature size scanning electron microscope under the measurement mode, then changing the measurement mode of the feature size scanning electron microscope; and obtaining the measurement image of the measurement feature size boundaries using the changed measurement mode. The measurement image and the standard image are processed separately to obtain a first measurement signal boundary in the measurement image and a second measurement signal boundary in the standard image. The first measurement signal boundary refers to the boundary of the feature size to be detected in the measurement image through image processing, and the second measurement signal boundary refers to the boundary of the feature size to be detected in the standard image through image processing. The processing of the measurement image to obtain the first measurement signal boundary includes: generating a contrast image including the initial measurement signal boundary based on the measurement image; generating a gradient image of the measurement image, the gradient image containing the gradient value of each pixel; generating a mapping image based on the contrast image and the gradient image, the mapping image containing the initial measurement signal boundary with a non-zero gradient value, and using the initial measurement signal boundary with a non-zero gradient value as the first measurement signal boundary. Obtain first contrast information of the boundary of the first measurement signal and second contrast information of the boundary of the second measurement signal. The first contrast information includes the correspondence between the first contrast value of each pixel on the first measurement signal boundary and the number of pixels thereon. The second contrast information includes the correspondence between the second contrast value of each pixel on the second measurement signal boundary and the number of pixels thereon. Based on the first contrast information and the second contrast information, a matching index is obtained between the boundary of the first measurement signal and the boundary of the second measurement signal. The matching index includes a first boundary matching index and a second boundary matching index, which are used to indicate the degree of matching between the boundary of the measurement feature size and the boundary of the first measurement signal, specifically including: Based on whether the first contrast value is 0, the first contrast information is divided into first boundary measurement contrast information and second boundary measurement contrast information; based on whether the second contrast value is 0, the second contrast information is divided into first boundary standard contrast information and second boundary standard contrast information; based on the first boundary measurement contrast information and the first boundary standard contrast information, the first boundary matching index is obtained; based on the second boundary measurement contrast information and the second boundary standard contrast information, the second boundary matching index is obtained.

2. The method of claim 1, wherein, Processing the measurement image to obtain the first measurement signal boundary in the measurement image includes: Obtain the contrast information in the measured image; The boundary of the first measurement signal in the measurement image is determined based on the contrast information in the measurement image.

3. The method as described in claim 2, characterized in that, Obtaining the contrast information in the measured image includes: The pixel values ​​of each pixel in the same column of the measured image are accumulated along the second direction to obtain the contrast value of each pixel. The second direction is the extension direction of the boundary of the measured feature size. Based on the contrast values ​​of each pixel in the same row along the first direction, a contrast signal for each row is drawn, wherein the first direction is perpendicular to the second direction, and the contrast information includes the contrast signal. Determining the boundary of the first measurement signal in the measurement image based on the contrast information in the measurement image includes: Obtain the maximum contrast value in the contrast signal of each row; The contrast threshold of the contrast signal for the corresponding row is determined based on the maximum contrast value. The target pixel in the corresponding row is determined in the contrast signal of the corresponding row based on the contrast threshold. The target pixel points are connected along the second direction to form the first measurement signal boundary.

4. The method as described in claim 1, characterized in that, Obtaining the first contrast information of the boundary of the first measurement signal includes: Extract the first contrast value of each pixel on the boundary of the first measurement signal; The number of pixels at the same first contrast value on the boundary of the first measurement signal is obtained; A first coordinate is established based on the first contrast value, and a second coordinate is established based on the number of pixels. The first contrast information is included in the first coordinate and the second coordinate, and the correspondence between the number of pixels and the first contrast value is also included.

5. The method as described in claim 1, characterized in that, Also includes: The measurement feature dimensions of the semiconductor pattern in the semiconductor structure are obtained based on the boundary of the first measurement signal.

6. A computer device, characterized in that, include: One or more processors; A memory configured to store one or more programs, which, when executed by one or more processors, cause the computer device to perform the method as described in any one of claims 1 to 5.

7. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is run on a computer, it causes the computer to perform the method as described in any one of claims 1 to 5.

Images