Grain template automatic extraction method and system based on two-step contour segmentation

Abstract

The invention provides a grain template automatic extraction method and system based on two-step contour segmentation, and the method comprises the steps: estimating the periodicity of a grain, and determining the initial width and height of the grain according to the periodicity in the horizontal and vertical directions; enhancing edge information in the wafer image; extracting straight lines in four directions in the image, wherein the four directions comprise a vertical direction, a horizontal direction, a 45-degree direction and a 135-degree direction; setting an initial contour of the crystal grain; performing image segmentation by using the active contour model, and iteratively updating the fine contour of the crystal grain; repeating two-step contour segmentation on the plurality of areas on the wafer image to obtain segmentation results of the plurality of wafer images; and smoothing a plurality of obtained results, and outputting the results as a crystal grain template. According to the secondary contour extraction method, the defects that the accuracy of the crystal grain contour based on linear extraction is low and an initial value needs to be manually set for an active contour model are overcome. According to the invention, the high-precision template of the crystal grain can be automatically extracted, so that real-time wafer defect detection is facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of wafer defect detection, and particularly to an automatic extraction method and system for grain templates based on two-step contour segmentation. Background Art

[0002] A high-precision grain template is the basis for real-time wafer defect detection. Through the template matching algorithm, it can be detected whether the grains on the wafer image meet the expectations, and any place that does not match the grain template may indicate the existence of defects. The generation of the grain template is an important link in formulating the wafer recipe.

[0003] Existing wafer recipe formulation methods all require the equipment to stop normal production work, and then manual configuration by operators. This not only reduces work efficiency, but also is prone to errors when operators lack experience, resulting in more serious consequences.

[0004] Existing technologies have designed shape-based template matching methods to segment repetitive structures on images. The disadvantage is that it is easy to lack grain segmentation bands, which may lead to incomplete wafer images. There are also technologies that only extract periodic and intersection points, and the accuracy of the grain contour is poor, which is not suitable for wafer defect detection based on templates.

[0005] Currently, there is no method for automatically generating high-precision grain templates on wafer images in the semiconductor equipment industry. Summary of the Invention

[0006] The object of the present invention is achieved through the following technical solutions.

[0007] Therefore, in view of the above problems, the present invention designs an automatic extraction method for grain templates. First, straight lines on the image are extracted, and the connected straight line regions that meet the requirements are used as the initial contour. Then, the contour is iteratively optimized, and finally a high-precision grain template is extracted with an error less than 1 pixel.

[0008] Specifically, according to the first aspect of the present invention, an automatic extraction method for grain templates based on two-step contour segmentation is provided, including: S1. Estimate the periodicity of the grains, and determine the initial width and height of the grains according to the periodicity in the horizontal and vertical directions; S2. Enhance the edge information in the wafer image; S3. Extract straight lines in four directions of the wafer image, and the four directions include vertical, horizontal, 45 degrees, and 135 degrees; S4. Set the initial contour of the grains; S5. Use the active contour model for image segmentation and iteratively update the fine contour of the grains; S6. Repeat steps S4 - S5 for multiple regions on the wafer image to obtain segmentation results of multiple wafer images; S7. Smooth the multiple results obtained in step S6, and output the result as the die template.

[0009] According to the second aspect of the present invention, there is also provided an automatic die template extraction system based on two - step contour segmentation, including: An estimation period module, configured to estimate the periodicity of the die, and determine the initial width and height of the die according to the periodicity in the horizontal and vertical directions; An edge enhancement module, configured to enhance the edge information in the wafer image; A straight line extraction module, configured to extract straight lines in four directions in the wafer image, and the four directions include vertical, horizontal, 45 - degree, and 135 - degree directions; An initial contour module, configured to set the initial contour of the die; A fine contour module, configured to perform image segmentation using an active contour model and iteratively update the fine contour of the die; A repeated extraction module, configured to repeatedly extract contours in multiple regions on the wafer image to obtain segmentation results of multiple wafer images; A smoothing output module, configured to smooth the multiple results obtained and output the result as the die template.

[0010] The advantages of the present invention are as follows: The present invention can automatically extract a high - precision template of the die for real - time wafer defect detection. The present invention designs a method of secondary contour extraction. After obtaining the periodicity of the die pattern, the initial contour of the die is obtained by using a method based on straight line extraction, and then the fine contour of the die is extracted by using an image segmentation algorithm based on an active contour model. This solves both the problem of low accuracy of the die contour based on straight line extraction and the drawback that the active contour model requires manual setting of the initial value. Description of the Drawings

[0011] By reading the following detailed description of the preferred embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered as a limitation of the present invention. Moreover, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings: Figure 1 Shows a flowchart of an automatic die template extraction method based on two - step contour segmentation according to an embodiment of the present invention.

[0012] Figure 2 Shows a flowchart of selecting an initial contour according to an embodiment of the present invention.

[0013] Figure 3Shows the composition diagram of the automatic extraction system of the grain template based on two-step contour segmentation according to an embodiment of the present invention.

[0014] Figure 4 Shows a schematic structural diagram of an electronic device provided by an embodiment of the present invention.

[0015] Figure 5 Shows a schematic diagram of a storage medium provided by an embodiment of the present invention. Detailed implementation manners

[0016] Hereinafter, the exemplary embodiments of the present disclosure will be described in more detail with reference to the drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be completely conveyed to those skilled in the art.

[0017] Due to the very high precision requirements of the wafer template (the error is less than 1 pixel) and the existence of a large number of periodic patterns, the existing automatic image segmentation methods are difficult to meet the requirements. Semi-automatic segmentation algorithms such as the active contour model method need to manually set appropriate initial values to obtain appropriate results. However, due to the need for downtime operations in industrial production, the practicality is not good.

[0018] The high-precision automatic extraction method of the grain contour of the present invention first estimates the periodicity, then extracts the straight lines, selects the effective straight lines to generate the initial contour, uses the active contour algorithm to iteratively update the contour to obtain the fine contour, and finally performs median filtering to obtain the template. Specifically, as Figure 1 shown, the present invention includes the following steps: S1. Estimate the periodicity of the grains, and determine the initial width and height of the grains according to the periodicity in the horizontal and vertical directions; S2. Enhance the edge information in the image; S3. Extract the straight lines (vertical, horizontal, 45-degree, and 135-degree directions) in the image; S4. Set the initial contour of the grains; S5. Use the active contour model for image segmentation to iteratively update the fine contour of the grains; S6. Repeat steps 4-6 in multiple regions on the image to obtain the segmentation results of multiple grains (not exceeding 15); S7. Smooth the multiple results obtained in step 6, and output the result as the grain template. Embodiment

[0019] Specifically, for example, an embodiment of the present invention is as follows: Let the original wafer image be , with width and height being m and n respectively, diagonal length being l, and horizontal and vertical coordinates being x and y respectively. When estimating the periodicity of grains, first perform a discrete Fourier transform on the image using Equation 1, where u and v are the coordinates on the frequency spectrum diagram.

[0020] (1)

[0021] Then calculate the periodic energy distributions in the horizontal and vertical directions using Equations 2 and 3 and .

[0022] (2)

[0023] (3)

[0024] Finally, take the coordinates at the maximum of the periodic energy in the horizontal and vertical directions as the initial width and height of the grains and .

[0025] When enhancing the edge information of the image, first use two operators and to convolve the image , obtaining the convolved images and , and then use Equation 4 to obtain the image with enhanced edge information .

[0026] (4)

[0027] When extracting straight lines from the image, first create an accumulator matrix with a width of 2* and a height of 180, and the values of all points in the matrix are 0. Then traverse each point in the image . When the value exceeds the minimum of the length threshold (an empirical parameter, set to 128 in the present invention), according to its coordinates (x, y), for each angle , calculate the distance according to Equation 5 .

[0028] (5)

[0029] Increment the value at , in the accumulator matrix by 1. After the image traversal is completed, traverse the points in the accumulator matrix that exceed the threshold as candidate straight lines, and then calculate the two endpoints q1, q2 of the candidate straight line according to Equation 6, where the coordinates of q1 are ( , ), and the coordinates of q2 are ( , ). is half of the width of the accumulator matrix.

[0030] (6)

[0031] After obtaining the endpoints of the straight line, calculate the angle and length d of the straight line according to Formulas 7 and 8.

[0032] (7)

[0033] (8)

[0034] Take the straight lines whose angles and lengths are both within the threshold range (when or ), or , or , or ) as valid straight lines.

[0035] The method for setting the initial contour of the grain is as Figure 2 shown. If the number of valid straight lines is greater than or equal to 4 (the minimum number of grain boundaries), starting from any endpoint, traverse the straight lines based on the depth-first principle. During the traversal, if the distance between the endpoints of two straight lines is less than the threshold (an empirical parameter, set to 5 pixels in this experiment), they are considered the same endpoint. If the traversed path can return to the starting endpoint, the closed loop formed by this path is taken as the closed polygon.

[0036] If the number of valid straight lines is greater than or equal to 4, but they cannot be combined into a closed polygon, count the closed polygons that can be formed by the endpoints of all valid straight lines, and then take the closed polygon with the largest area as the initial contour of the grain. The calculation method of the closed polygon is as follows: First, sort the coordinates of the endpoints in ascending order in the horizontal direction (x coordinate) (if x is the same, sort in ascending order of y). Then traverse the sorted points from left to right, retaining the points that meet the non-right-turn condition to form the lower polygon. Then traverse the sorted points from right to left, retaining the points that meet the non-right-turn condition to form the upper polygon. Finally, merge the lower polygon and the upper polygon and remove the duplicate vertices to form the complete closed polygon.

[0037] If the number of valid straight lines is less than 4, first perform binary filtering on the image (an empirical parameter, the threshold is 125 in the present invention). The pixels greater than or equal to the threshold are edge pixels, and the pixels less than the threshold are non-edge pixels, obtaining the binary edge map . Then set a sliding window (in the present invention, the width and height of the sliding window are 1 / 20 of the initial width and height of the grain, and the step size of the sliding window is half of the width and height), and count the edge pixel density of each sliding window. Merge the windows with edge pixel density greater than or equal to a threshold (an empirical parameter, which is 0.3 in the present invention). Then use The circular kernel of is alternately used to perform morphological dilation and erosion operations for multiple times (an empirical parameter, which is 3 times in the present invention) to fill holes and remove burrs. Then use the skeleton of the edge pixels as effective straight lines, and calculate the closed polygon by the method in the previous step. Take the closed polygon with the largest area as the initial contour of the grain.

[0038] When calculating the closed polygon, if multiple closed polygons are obtained, take the closed polygon with the largest area as the initial contour of the grain.

[0039] When optimizing the contour of the grain using the active contour algorithm, first set the weight parameters α, β of the contour C and , and then for the i-th point on the contour C, update the position of the contour point using Equation 9-17. Where and respectively represent the internal energy and external energy of the contour. These two symbols are used to express intermediate variables, that is, the partial derivatives of the energy in the horizontal and vertical directions, indicating the change of the contour in the horizontal and vertical directions. If the difference between two updates is less than the threshold (such as 1 pixel), stop the update.

[0040] (9)

[0041] (10)

[0042] (11)

[0043] (12)

[0044] (13)

[0045] (14)

[0046] (15)

[0047] (16)

[0048] (17)

[0049] For the I detected grain contours, let the width and height of the i-th contour be respectively And , first, use Equation 18 and Equation 19 to calculate the widths and the average values of the heights .

[0050] (18)

[0051] (19)

[0052] And use them as the width and height of the grain contour template. Then, align the upper left corners of these grain contours, and perform median filtering on each point at the same position. Use the result after median filtering as the value of the current position of the grain contour template.

[0053] As Figure 3 shown, an automatic grain template extraction system based on two-step contour segmentation includes: An estimation period module 301, configured to estimate the periodicity of grains, and determine the initial width and height of grains according to the periodicity in the horizontal and vertical directions; An edge enhancement module 302, configured to enhance the edge information in the wafer image; A straight line extraction module 303, configured to extract straight lines in four directions in the wafer image, and the four directions include vertical, horizontal, 45 degrees, and 135 degrees; An initial contour module 304, configured to set the initial contour of grains; A fine contour module 305, configured to perform image segmentation using an active contour model and iteratively update the fine contour of grains; A repeated extraction module 306, configured to repeatedly extract contours in multiple regions on the wafer image to obtain segmentation results of multiple wafer images; A smoothing output module 307, configured to smooth the obtained multiple results and output the results as a grain template.

[0054] The automatic grain template extraction system based on two-step contour segmentation provided by the above embodiments of the present invention and the automatic grain template extraction method based on two-step contour segmentation provided by the embodiments of the present invention are based on the same inventive concept and have the same beneficial effects as the methods adopted, run, or implemented by the application programs stored therein.

[0055] The embodiments of the present invention also provide an electronic device corresponding to the automatic grain template extraction method based on two-step contour segmentation provided by the foregoing embodiments to execute the automatic grain template extraction method based on two-step contour segmentation. The embodiments of the present invention are not limited.

[0056] Please refer to Figure 4 , which shows a schematic diagram of an electronic device provided by some embodiments of the present invention. AsFigure 4 As shown, the electronic device 20 includes: a processor 200, a memory 201, a bus 202, and a communication interface 203. The processor 200, the communication interface 203, and the memory 201 are connected through the bus 202. A computer program that can run on the processor 200 is stored in the memory 201. When the processor 200 runs the computer program, it executes the automatic extraction method of the crystal grain template based on two-step contour segmentation provided in any of the foregoing embodiments of the present invention.

[0057] Among them, the memory 201 may include a high-speed random access memory (RAM: Random Access Memory), and may also include a non-volatile memory, such as at least one disk memory. Through at least one communication interface 203 (which can be wired or wireless), a communication connection is realized between this system network element and at least one other network element, and the Internet, wide area network, local area network, metropolitan area network, etc. can be used.

[0058] The bus 202 can be an ISA bus, a PCI bus, an EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. Among them, the memory 201 is used to store a program. After receiving an execution instruction, the processor 200 executes the program. The automatic extraction method of the crystal grain template based on two-step contour segmentation disclosed in any of the foregoing embodiments of the present invention can be applied to the processor 200 or implemented by the processor 200.

[0059] The processor 200 may be an integrated circuit chip with the ability to process signals. In the implementation process, each step of the above method can be completed by the integrated logic circuit of the hardware in the processor 200 or the instructions in the form of software. The above-mentioned processor 200 may be a general-purpose processor, including a central processing unit (CPU for short), a network processor (NP for short), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. It can implement or execute the various methods, steps and logic block diagrams disclosed in the embodiments of the present invention. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present invention can be directly embodied as being executed and completed by the hardware decoding processor, or executed and completed by the combination of the hardware and software modules in the decoding processor. The software module may be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. This storage medium is located in the memory 201, and the processor 200 reads the information in the memory 201 and combines its hardware to complete the steps of the above method.

[0060] The electronic device provided by the embodiments of the present invention and the method for automatically extracting a grain template based on two-step contour segmentation provided by the embodiments of the present invention are based on the same inventive concept and have the same beneficial effects as the method adopted, run or implemented by it.

[0061] The embodiments of the present invention also provide a computer-readable storage medium corresponding to the method for automatically extracting a grain template based on two-step contour segmentation provided by the foregoing embodiments. Please refer to Figure 5 , which shows that the computer-readable storage medium is an optical disc 30, on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it will execute the method for automatically extracting a grain template based on two-step contour segmentation provided by any of the foregoing embodiments.

[0062] It should be noted that examples of the computer-readable storage medium may also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical and magnetic storage media, which will not be elaborated here one by one.

[0063] The computer-readable storage medium provided by the above embodiments of the present invention and the method for automatically extracting a grain template based on two-step contour segmentation provided by the embodiments of the present invention are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application programs stored therein.

[0064] It should be noted that: The algorithms and displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings provided herein. The structure required to construct such systems will be apparent from the above description. In addition, the present invention is not directed to any particular programming language. It should be understood that the content of the present invention described herein can be implemented using various programming languages, and the description of the specific language above is to disclose the best mode of the present invention.

[0065] In the specification provided herein, a large number of specific details are set forth. However, it can be understood that the embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0066] Similarly, it should be understood that, in order to streamline the present invention and assist in understanding one or more of the various inventive aspects, in the above description of the exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, the disclosed methods should not be construed as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as reflected in the following claims, the inventive aspects lie in less than all the features of the single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of the present invention.

[0067] Those skilled in the art can understand that the modules in the devices in the embodiments can be adaptively changed and disposed in one or more devices different from the embodiments. The modules or units or components in the embodiments can be combined into one module or unit or component, and in addition, they can be divided into multiple sub-modules or sub-units or sub-components. Except that at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all the features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all the processes or units of any method or device so disclosed. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) can be replaced by an alternative feature that provides the same, equivalent, or similar purpose.

[0068] In addition, those skilled in the art will appreciate that although some of the embodiments described herein include certain features included in other embodiments rather than other features, combinations of features of different embodiments are meant to be within the scope of the present invention and form different embodiments. For example, in the following claims, any one of the claimed embodiments can be used in any combination.

[0069] Each component embodiment of the present invention can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art should understand that a microprocessor or a digital signal processor (DSP) can be used in practice to implement some or all of the functions of some or all of the components in the virtual machine creation system according to the embodiments of the present invention. The present invention can also be implemented as a device or system program (e.g., a computer program and a computer program product) for performing part or all of the methods described herein. Such a program implementing the present invention can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, or provided on a carrier signal, or in any other form.

[0070] It should be noted that the above embodiments illustrate rather than limit the present invention, and those skilled in the art can design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in the claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention can be implemented by means of hardware including several different elements and by means of a suitably programmed computer. In the unit claims listing several systems, several of these systems can be embodied by the same item of hardware. The use of the words first, second, and third, etc. does not denote any order. These words can be interpreted as names.

[0071] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. An automatic extraction method for grain templates based on two-step contour segmentation, characterized in that, Including: S1. Estimate the periodicity of the grains, and determine the initial width and height of the grains according to the periodicity in the horizontal and vertical directions; S2. Enhance the edge information in the wafer image; S3. Extract the straight lines in four directions in the wafer image, and the four directions include the vertical, horizontal, 45-degree and 135-degree directions; S4. Set the initial contour of the grains; S5. Use the active contour model for image segmentation, and iteratively update the fine contour of the grains; S6. Repeat steps S4 - S5 in multiple regions on the wafer image to obtain the segmentation results of multiple wafer images; S7. Smooth the multiple results obtained in step S6, and output the result as the grain template.

2. The automatic grain template extraction method based on two-step contour segmentation according to claim 1, characterized in that the estimation of the periodicity of the grains, and the determination of the initial width and height of the grains according to the periodicity in the horizontal and vertical directions, includes: When estimating the periodicity of the grains, first perform a discrete Fourier transform on the wafer image, then calculate the periodic energy distribution in the horizontal and vertical directions, and finally use the coordinates of the points with the maximum periodic energy in the horizontal and vertical directions as the initial width and height of the grains.

3. The automatic grain template extraction method based on two-step contour segmentation according to claim 1 or 2, characterized in that the enhancement of the edge information in the wafer image, includes: First, convolve the wafer image with two operators respectively to obtain the convolved image, and then calculate the image with enhanced edge information according to the convolved image.

4. The automatic grain template extraction method based on two-step contour segmentation according to claim 3, characterized in that the extraction of the straight lines in four directions in the wafer image, includes: First, create an accumulator matrix, and then traverse each point in the wafer image with enhanced edge information. When the value exceeds the minimum length threshold, calculate the distance for each angle according to the coordinates of the point; After the image traversal is completed, traverse the points in the accumulator matrix that exceed the minimum length threshold as candidate straight lines, and then calculate the two endpoints of the candidate straight lines; After obtaining the endpoints of the candidate straight lines, calculate the angle and length of the candidate straight line; The candidate straight lines with both the angle and length within the threshold range are used as valid straight lines.

5. The automatic grain template extraction method based on two-step contour segmentation according to claim 4, characterized in that the setting of the initial contour of the grains, includes: If the number of valid straight lines is greater than or equal to 4, start from any endpoint and traverse the straight lines based on the depth-first principle; during the traversal, if the distance between the endpoints of two straight lines is less than the threshold, they are considered to be the same endpoint; if the traversed path can return to the starting endpoint, the closed loop formed by this path is used as a closed polygon; If the number of valid straight lines is greater than or equal to 4, but a closed polygon cannot be formed, count the closed polygons that can be formed by the endpoints of all valid straight lines, and then use the closed polygon with the largest area as the initial contour of the grains. If the number of effective straight lines is less than 4, first perform binary filtering on the image after edge information enhancement. Pixels greater than or equal to the threshold are edge pixels, and pixels less than the threshold are non-edge pixels to obtain a binary edge map. Then set a sliding window and count the edge pixel density of each sliding window. Merge the windows with edge pixel density greater than or equal to the threshold. Then alternately perform morphological dilation and erosion operations using a circular kernel, repeating multiple times to fill holes and remove burrs. Then use the skeleton of the edge pixels as effective straight lines, calculate the closed polygon, and use the closed polygon with the largest area as the initial contour of the grain.

6. The automatic extraction method of a grain template based on two-step contour segmentation according to claim 5, characterized in that The use of the active contour model for image segmentation and iteratively updating the fine contour of the grain includes: First, set the weight parameter of the contour. Then, for each point on the contour, update the position of the contour point. If the difference between two updates is less than the threshold, stop the update.

7. The automatic extraction method of a grain template based on two-step contour segmentation according to claim 6, characterized in that The smoothing of the multiple results obtained in step S6 and outputting the results as the grain template includes: For the multiple detected grain contours, first calculate the width mean and height mean of the multiple grain contours; And use the width mean and height mean as the width and height of the grain contour template. Then align the upper left corners of the multiple grain contours, and then perform median filtering on each point at the same position. Use the result after median filtering as the value of the current position of the grain contour template.

8. An automatic extraction system for grain templates based on two-step contour segmentation, characterized in that, Including: An estimation period module for estimating the periodicity of the grain and determining the initial width and height of the grain according to the periodicity in the horizontal and vertical directions; An edge enhancement module for enhancing the edge information in the wafer image; A straight line extraction module for extracting straight lines in four directions in the wafer image, and the four directions include vertical, horizontal, 45 degrees, and 135 degrees; An initial contour module for setting the initial contour of the grain; A fine contour module for using the active contour model for image segmentation and iteratively updating the fine contour of the grain; A repeated extraction module for repeatedly extracting contours in multiple regions on the wafer image to obtain segmentation results of multiple wafer images; A smoothing output module for smoothing the multiple obtained results and outputting the results as the grain template.

9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor runs the computer program to implement the method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the method according to any one of claims 1-7.

Images