Wafer sem image defect type detection method, device, medium and product
By extracting defect images from wafer SEM images and judging their similarity with adjacent normal images, graphic and non-graphic defects can be distinguished, solving the problem of the inability to distinguish defect types in existing technologies and achieving efficient and accurate defect detection.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- DONGFANG JINGYUAN ELECTRON LTD
- Filing Date
- 2026-02-26
- Publication Date
- 2026-06-02
AI Technical Summary
Existing wafer surface defect detection methods cannot distinguish between pattern defects caused by poor photolithography, etching processes, or equipment problems and non-pattern defects caused by foreign particles, resulting in high costs and low efficiency for manual inspection.
By extracting defect images from wafer SEM images and detecting their similarity to adjacent normal images, the defect type can be determined as graphic or non-graphic, reducing the number of defects that need to be manually identified.
It reduces labor costs, improves defect detection efficiency and accuracy, simplifies the sample library and model construction process, and enhances the stability and accuracy of detection.
Smart Images
Figure CN122134677A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of semiconductor technology, and in particular to a method, device, medium, and product for detecting defect types in wafer SEM images. Background Technology
[0002] In modern integrated circuit manufacturing, various wafer surface defects have become a key factor affecting the yield of integrated circuit manufacturing. Therefore, wafer defect detection is a crucial step in the integrated circuit manufacturing process. Currently, the method for detecting wafer surface defects typically involves using a scanning electron microscope (SEM) to acquire SEM images of the wafer surface. The presence of defects is determined by comparing the SEM images with the design layout or by comparing images of adjacent dies within the SEM image.
[0003] However, existing wafer surface defect detection methods can only determine whether defects exist on the wafer surface, but cannot identify whether the defects are pattern defects caused by poor photolithography or etching processes or equipment problems, or non-pattern defects caused by foreign particles. Therefore, all detected defects need to be manually inspected, which is costly and inefficient. Summary of the Invention
[0004] One object of the present invention is to provide a method, device, medium and product for detecting defect types in wafer SEM images that helps reduce the labor cost of wafer surface defect detection and improve detection efficiency.
[0005] Specifically, this invention provides a method for detecting defect types in wafer SEM images, comprising: Identify defects in the wafer SEM image; Extract the defect image corresponding to the defect from the wafer SEM image; The system detects whether the defect image meets preset requirements. If it does, the defect is determined to be a graphic defect; otherwise, the defect is determined to be a non-graphic defect.
[0006] Optionally, the step of detecting whether the defect image meets preset requirements includes: Extract normal images adjacent to the defective images from the wafer SEM images to obtain a control image; The similarity between the defective image and the comparison image determines whether the defective image meets the preset requirements.
[0007] Optionally, the step of determining whether the defective image meets the preset requirements based on the similarity between the defective image and the reference image includes: Extract the image feature values of each pixel in the defective image and the control image respectively; By counting the number of pixels for each of the image feature values, the distribution of image feature values between the defective image and the control image is obtained. The distribution of image feature values is used to characterize the number of pixels corresponding to different numerical ranges of the image feature values in the image. Whether the defective image meets the preset requirements is determined by comparing the image feature value distribution of the defective image and the control image.
[0008] Optionally, the step of determining whether the defective image meets the preset requirements by comparing the image feature value distributions of the defective image and the control image includes: The distribution graphs of the image feature values of the comparison image and the image feature values of the defective image are obtained in a set coordinate system, wherein the set coordinate system uses the image feature values and the number of pixels as the horizontal and vertical axes, respectively. Calculate the overlap area between the distribution patterns of the control image and the distribution patterns of the defect image; Determine whether the overlapping area is greater than a set area threshold. If yes, the defective image is determined to meet the preset requirements. If no, the defective image is determined not to meet the preset requirements.
[0009] Optionally, the image feature values include image grayscale values or image internal gradient values.
[0010] Optionally, the step of obtaining the comparison image includes: Determine the positional relationship of the defect relative to the corresponding reference graphic in the design layout; If the area of the defect located outside the reference pattern is larger than the area located inside the reference pattern, the comparison image is determined to be an image of a normal pattern adjacent to the defect image in the wafer SEM image; If the area of the defect located within the reference pattern is greater than the area located outside the reference pattern, the comparison image is determined to be the background region adjacent to the defect image in the wafer SEM image.
[0011] Optionally, the step of determining defects in the wafer SEM image includes: Align the wafer SEM image with the design layout; Defects in the wafer SEM image are determined based on the degree of difference between the wafer SEM image and the image of the same location on the design layout.
[0012] In another aspect of the invention, a computer device is also provided, including a memory, a processor, and a computer-executable program stored in the memory and running on the processor, wherein the processor executes the computer-executable program to implement a defect type detection method for wafer SEM images according to any of the preceding claims.
[0013] In another aspect of the invention, a computer-readable storage medium is also provided, on which a computer-executable program is stored, which, when executed by a processor, implements the defect type detection method for wafer SEM images according to any one of the preceding claims.
[0014] In another aspect of the invention, a computer program product is also provided, comprising a computer executable program that, when executed by a processor, implements the defect type detection method for wafer SEM images according to any of the preceding claims.
[0015] The defect type detection method for wafer SEM images of the present invention involves extracting the corresponding defect image from the wafer SEM image after identifying a defect. The method then checks whether the defect image meets preset requirements. If it does, the defect is determined to be a graphic defect; otherwise, it is determined to be a non-graphic defect. In other words, after confirming the existence of a defect, its type (graphic or non-graphic) can be determined by extracting the corresponding defect image separately. This eliminates the need for manual verification of non-graphic defects, reducing the number of defects requiring manual inspection, thus saving labor costs and improving defect detection and subsequent processing efficiency.
[0016] Furthermore, the defect type detection method for wafer SEM images of the present invention extracts a normal image adjacent to the defective image from the wafer SEM image to obtain a reference image. The similarity between the defective image and the reference image determines whether the defective image meets preset requirements. In other words, it can determine whether a defective image is a graphic defect or a non-graphic defect using only information from the wafer SEM image. Compared to judging solely based on the characteristics of the defective image itself, this eliminates the need for a cumbersome sample library and model construction process to prepare corresponding type standards for comparison in advance, reducing resource consumption. On the other hand, the judgment standards for the same type of defect may vary significantly in different scenarios, and using a fixed standard is prone to error. By using the similarity between the defective image and the reference image to determine the defect type, each defective image is compared with its adjacent normal image, achieving dynamic verification, which helps improve the accuracy and stability of type detection.
[0017] The above and other objects, advantages and features of the present invention will become more apparent to those skilled in the art from the following detailed description of specific embodiments of the invention in conjunction with the accompanying drawings. Attached Figure Description
[0018] The following sections will describe some specific embodiments of the invention in detail by way of example and not limitation, with reference to the accompanying drawings. The same reference numerals in the drawings denote the same or similar parts or portions. Those skilled in the art should understand that these drawings are not necessarily drawn to scale. In the drawings: Figure 1 This is a schematic flowchart of a defect type detection method for wafer SEM images according to an embodiment of the present invention; Figure 2 This is a schematic flowchart illustrating the method for detecting defects in a wafer SEM image according to an embodiment of the present invention. Figure 3 This is a schematic diagram comparing a wafer SEM image and a design layout according to an embodiment of the present invention; Figure 4 This is a schematic flowchart illustrating whether a defect image meets preset requirements in a method for detecting defect types in a wafer SEM image according to an embodiment of the present invention. Figure 5 A schematic flowchart illustrating the method for detecting defect types in wafer SEM images according to an embodiment of the present invention, which determines whether a defect image meets preset requirements based on the similarity between the defect image and a reference image; Figure 6 A schematic flowchart illustrating the method for detecting defect types in wafer SEM images according to an embodiment of the present invention, showing how a reference image is obtained. Figure 7 This is a schematic diagram of a wafer SEM image with defect removal according to an embodiment of the present invention; Figure 8 A schematic flowchart illustrating a method for detecting defect types in wafer SEM images according to an embodiment of the present invention, which determines whether a defect image meets preset requirements by comparing the image feature value distributions of a defect image and a control image; Figure 9 This is a schematic diagram of the distribution of image feature values of a defective image in a set coordinate system according to an embodiment of the present invention. Figure 10 This is a schematic diagram of the distribution of image feature values of a comparison image in a set coordinate system according to an embodiment of the present invention. Figure 11 This is a schematic diagram of a computer device according to an embodiment of the present invention; Figure 12This is a schematic diagram of a computer-readable storage medium according to an embodiment of the present invention; Figure 13 This is a schematic diagram of a computer program product according to an embodiment of the present invention. Detailed Implementation
[0019] Those skilled in the art should understand that the embodiments described below are merely a part of the embodiments of the present invention, and not all of the embodiments of the present invention. These partial embodiments are intended to explain the technical principles of the present invention and are not intended to limit the scope of protection of the present invention. Based on the embodiments provided by the present invention, all other embodiments obtained by those skilled in the art without creative effort should still fall within the scope of protection of the present invention.
[0020] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus or device (such as a computer-based system, a processor-included system or other system that can fetch and execute instructions from, an instruction execution system, apparatus or device).
[0021] The flowchart provided in this embodiment is not intended to indicate that the operations of the method will be performed in any particular order, or that all operations of the method are included in every case. Furthermore, the method may include additional operations. Within the scope of the technical concept provided by the method in this embodiment, additional variations can be made to the above method.
[0022] like Figure 1 As shown, in one embodiment, the defect type detection method for wafer SEM images generally includes: Step S101: Identify defects in the wafer SEM image.
[0023] Reference Figure 2 As shown, in one embodiment, this step includes: Step S201: Align the wafer SEM image with the design layout.
[0024] Specifically, the design layout can be understood as the ideal image that needs to be obtained on the wafer surface. The wafer SEM image is then aligned with the same location on the design layout.
[0025] Step S202: Determine the defects in the wafer SEM image based on the degree of difference between the wafer SEM image and the image of the same location on the design layout.
[0026] Specifically, by comparing the wafer SEM image with the image of the same location on the design layout one by one, the degree of image difference is determined, thereby obtaining the difference between different locations of the wafer SEM image and the ideal image, and thus identifying defects in the wafer SEM image. The method of comparing the degree of difference between the wafer SEM image and the image of the same location on the design layout is existing technology and will not be elaborated here.
[0027] It should be noted that the method for determining defects in this embodiment can also be called the D2DB (Die-to-Database Inspection) method. In other embodiments, other methods can also be used to determine defects in wafer SEM images, such as die-to-die or cell-to-cell inspection methods.
[0028] Step S102: Extract the defect image corresponding to the defect from the wafer SEM image.
[0029] Specifically, this involves extracting the defect image corresponding to the defect from the wafer SEM image based on the defect's location and size. (Refer to...) Figure 3 As shown, defect image 210 and defect image 410 are extracted from wafer SEM images.
[0030] Step S103: Detect whether the defect image meets the preset requirements. If yes, proceed to step S104; otherwise, proceed to step S105.
[0031] Step S104: Determine that the defect is a graphic defect.
[0032] Step S105: Determine that the defect is a non-graphical defect.
[0033] Specifically, this means determining whether a defect is a graphic defect caused by a process or equipment problem, or a non-graphic defect caused by foreign particles, by detecting whether the defect image meets preset requirements.
[0034] In this embodiment, after identifying defects in a wafer SEM image, a defect image corresponding to the defect is extracted from the wafer SEM image. The defect image is then checked to see if it meets preset requirements. If it does, the defect is determined to be a graphic defect; otherwise, it is determined to be a non-graphic defect. In other words, after confirming the existence of a defect, its type (graphic or non-graphic) can be determined by extracting the defect image separately. This eliminates the need for manual verification of non-graphic defects, reducing the number of defects requiring manual inspection, thus saving labor costs and improving defect detection and subsequent processing efficiency.
[0035] like Figure 4 As shown, in one embodiment, the step of detecting whether a defective image meets preset requirements includes: Step S401: Extract the normal image adjacent to the defect image from the wafer SEM image to obtain the comparison image.
[0036] Specifically, a wafer SEM image includes the pattern and the background outside the pattern. The normal image adjacent to the defect image includes the image of the normal pattern (the part of the pattern that is not identified as a defect) adjacent to the defect image and the image of the background area adjacent to the defect image.
[0037] Step S402: Determine whether the defect image meets the preset requirements based on the similarity between the defect image and the reference image.
[0038] Specifically, this means determining whether a defective image meets preset requirements by checking the similarity between the defective image and the control image.
[0039] By extracting adjacent normal images from wafer SEM images, a control image is obtained. The similarity between the defect image and the control image determines whether the defect image meets preset requirements. In other words, it is possible to determine whether a defect image is a graphic or non-graphic defect using only information from the wafer SEM image. Compared to judging solely based on the characteristics of the defect image itself, this eliminates the need for cumbersome sample library and model construction processes to prepare corresponding type standards for comparison in advance, reducing resource consumption. Furthermore, the judgment criteria for the same type of defect may vary significantly in different scenarios, and using a fixed standard is prone to errors. By determining the defect type using the similarity between the defect image and the control image, each defect image is compared with its adjacent normal image, achieving dynamic verification, which helps improve the accuracy and stability of type detection.
[0040] like Figure 5 As shown, in one embodiment, the step of determining whether a defective image meets a preset requirement based on the similarity between the defective image and a control image includes: Step S501: Extract the image feature values of each pixel in the defect image and the control image respectively.
[0041] Specifically, image feature values can be image grayscale values, image internal gradient values, and other feasible image feature values, such as texture feature values. Taking image grayscale values as an example, this means extracting the image feature value of each pixel in the defective image and extracting the image feature value of each pixel in the comparison image.
[0042] Step S502: By counting the number of pixels for each image feature value, the distribution of image feature values between the defective image and the control image is obtained. The distribution of image feature values is used to characterize the number of pixels corresponding to different numerical ranges of image feature values in the image.
[0043] A numerical range can be a single image feature value or a range of values. Pixel values can be absolute values (i.e., the actual number of pixels within the numerical range) or relative values (i.e., the percentage of the actual number of pixels within the numerical range relative to the total number of pixels in the region).
[0044] Step S503: Determine whether the defect image meets the preset requirements by comparing the image feature value distribution of the defect image and the control image.
[0045] Specifically, this method characterizes the similarity between a defective image and a control image by analyzing the distribution of their image feature values. This approach not only helps avoid extensive preliminary learning processes, simplifying the detection process and thus improving detection efficiency, but also ensures good accuracy.
[0046] like Figure 6 As shown, in one embodiment, the step of obtaining the comparison image includes: Step S601: Determine the positional relationship of the defect relative to the corresponding reference graphic in the design layout.
[0047] If the area of the defect located outside the reference pattern is larger than the area located inside the reference pattern, proceed to step S602 to determine the reference image as the image of the normal pattern adjacent to the defect image in the wafer SEM image.
[0048] If the area of the defect located within the reference pattern is larger than the area located outside the reference pattern, proceed to step S603 to determine the reference image as the background area adjacent to the defect image in the wafer SEM image.
[0049] For example, refer to Figure 3 and Figure 7 The diagram shows a alignment of a wafer SEM image and a design layout. The design layout contains two graphics, reference graphic 110 and reference graphic 120. The wafer SEM image contains two graphics, graphic 200 and graphic 300 (comprising two parts). Both graphic 200 and graphic 300 in the wafer SEM image have defects, corresponding to defect image 210 and defect image 410, respectively.
[0050] Pattern 200 corresponds to reference pattern 110. For the defect in pattern 200, the area outside reference pattern 110 is larger than the area inside reference pattern 110. Therefore, the reference image is determined to be image 220 of a normal pattern adjacent to the defect image 210 in the wafer SEM image.
[0051] Pattern 300 corresponds to reference pattern 120. For the defect in pattern 300, the area within reference pattern 120 is larger than the area outside reference pattern 120. Therefore, the reference image is determined to be image 420 of the background region adjacent to defect image 410 in the wafer SEM image.
[0052] It should be noted that when adopting Figure 2 Under the condition that the defect is determined in the embodiment shown, the positional relationship between the defect and the corresponding reference graphic in the design layout can be determined at the same time as the defect is determined, which helps to further improve the detection efficiency.
[0053] Furthermore, referring to Figure 8 As shown, the steps for determining whether a defective image meets preset requirements by comparing the image feature value distributions of the defective image and the control image include: Step S801: Obtain the distribution graphs of image feature values of the comparison image and the defective image in a set coordinate system, with the image feature values and the number of pixels as the horizontal and vertical axes, respectively.
[0054] Specifically, refer to Figure 9 and Figure 10 As shown, in this embodiment, the coordinate system is set with image feature values as the horizontal axis and the number of pixels (referred to as pixel count in the figure) as the vertical axis.
[0055] Specifically, as mentioned above, pixel values can be absolute values (i.e., the actual number of pixels within a numerical range). Absolute values can be used when the areas of the defective image and the control image are not significantly different. Alternatively, they can be relative values (i.e., the percentage of the actual number of pixels within a numerical range to the total number of pixels in the region). Relative values are used when the areas of the defective image and the control image differ significantly. The distribution graph represents the region between the curve and the image feature value coordinate axis within a defined coordinate system. Specifically, it represents the region enclosed by the curve, the perpendicular lines from the two endpoints of the curve to the image feature value coordinate axis, and the image feature value coordinate axis itself.
[0056] Reference Figure 9 As shown, the distribution graph 211 of the image feature values of defect image 210 and defect image 410 in the set coordinate system is illustrated.
[0057] Reference Figure 10 As shown, the distribution graph 221 of the image feature values of the comparison image 220 and the distribution graph 421 of the image feature values of the comparison image 420 in the set coordinate system are illustrated.
[0058] Step S802: Calculate the overlap area between the distribution patterns of the control image and the defect image.
[0059] Reference Figure 9 and Figure 10 As shown, defect image 210 corresponds to control image 220. Therefore, the overlapping area of distribution pattern 221 of control image 220 and distribution pattern 211 of defect image 210 in the set coordinate system is calculated.
[0060] Since the defect image 410 corresponds to the reference image 420, the overlapping area of the distribution pattern 421 of the reference image 420 and the distribution pattern 411 of the defect image 410 in the set coordinate system is calculated.
[0061] Step S803: Determine whether the overlapping area is greater than a set area threshold. If yes, proceed to step S804; otherwise, proceed to step S805. Specifically, the set area threshold can be a fixed threshold or the area of the distribution pattern of the defect image multiplied by a set ratio.
[0062] Step S804: Determine that the defect image meets the preset requirements.
[0063] Step S805: Determine that the defect image does not meet the preset requirements.
[0064] Specifically, for defects whose area outside the reference image is larger than that inside the reference image, if the defect image is a graphic defect, it should be very similar to the control image (the image of the adjacent normal graphic). However, if the defect image is a non-graphic defect, it should be significantly different from the control image.
[0065] For defects whose area within the reference image is larger than its area outside the reference image, they are either non-graphical defects caused by foreign particles or graphic defects that are broken within the image. If the defect image is a graphic defect, it should be very similar to the control image (the image of the adjacent background area). If the defect image is a non-graphical defect, it should be significantly different from the control image.
[0066] Therefore, if the overlapping area is greater than the set area threshold, it means that the image feature value distributions of the reference image and the defective image are very similar, which means that the reference image and the defective image are very similar. Thus, the defective image is determined to meet the preset requirements and is identified as a graphic defect.
[0067] If the overlapping area is less than the set area threshold, it means that the image feature value distributions of the reference image and the defect image are not similar, which means that the reference image and the defect image are not similar. Therefore, it is determined that the defect image does not meet the preset requirements and is identified as a non-graphical defect.
[0068] By acquiring the distribution patterns of image feature values of both the control image and the defective image within a set coordinate system, and calculating the overlap area between the two distribution patterns, the method determines whether the defective image meets preset requirements based on whether the overlap area exceeds a set area threshold. This similarity judgment method is more intuitive and convenient, helping to ensure the accuracy of the detection results.
[0069] It should be noted that in some other embodiments, the step of determining whether a defective image meets the preset requirements by comparing the distribution of image feature values of the defective image and the control image may also include detecting whether the difference between the statistical parameters (such as the mean, variance, or standard deviation) of the image feature values of the defective image and the control image is less than a preset threshold. If so, the defective image is determined to meet the preset requirements; otherwise, the defective image is determined not to meet the preset requirements.
[0070] Additionally, it should be noted that for defects whose area outside the reference image is larger than that inside the reference image, if the defect image is a graphic defect, it should be very similar to the image of the adjacent normal graphic. However, if the defect image is a non-graphic defect, it will differ significantly from both the image of the adjacent normal graphic and the image of the adjacent background region.
[0071] For defects whose area within the reference graphic is larger than its area outside the reference graphic, they are either non-graphical defects caused by foreign particles or graphic defects that are broken within the graphic. If the defect image is a graphic defect, it should be very similar to the image of the adjacent background area. However, if the defect image is a non-graphical defect, it will differ significantly from both the image of the adjacent normal graphic and the image of the adjacent background area.
[0072] Therefore, in some embodiments, different Figure 6 The scheme shown allows the comparison image to simultaneously include images of normal graphics adjacent to the defective image and images of the background region adjacent to the defective image. In this scheme, the judgment can also be made by the overlapping area of the distribution patterns of the image feature values of the comparison image and the defective image within a set coordinate system. However, the comparison image now contains two distribution patterns, and the overlapping area is the total overlapping area between the distribution patterns of the defective image and the two distribution patterns of the comparison image.
[0073] This is because when images are dissimilar, the distribution patterns of the defective image and the control image are generally far apart. The difference between the overlapping area of one distribution pattern and the overlapping area of two distribution patterns is not too large. Therefore, the overlapping area of the distribution patterns of the image feature values of the control image and the image feature values of the defective image in the set coordinate system can still be used for judgment.
[0074] In addition, when using the difference of statistical parameters for judgment, the difference of the statistical parameters (such as mean, variance or standard deviation) of the image feature values of the two regions in the defective image and the control image is greater than the preset threshold. If it is, the defective image is determined not to meet the preset requirements. If not, the defective image is determined to meet the preset requirements.
[0075] In one embodiment, a computer device and a computer-readable storage medium are also provided. Figure 11 This is a schematic diagram of a computer device 10 according to an embodiment of the present invention. Figure 12 This is a schematic diagram of a computer-readable storage medium 20 according to an embodiment of the present invention.
[0076] The computer device 10 may include a memory 11, a processor 12, and a computer-executable program 13 stored on the memory 11 and running on the processor 12. When the processor 12 executes the computer-executable program 13, it implements the defect type detection method for wafer SEM images of any of the above embodiments.
[0077] The computer-readable storage medium 20 stores a computer-executable program 13 thereon, which, when executed by a processor, implements the defect type detection method for wafer SEM images of any of the above embodiments.
[0078] This embodiment also provides a computer program product. Figure 13 This is a schematic diagram of a computer program product 30 according to an embodiment of the present invention. The computer program product 30 includes a computer executable program 13, which, when executed by a processor 12, implements the defect type detection method for wafer SEM images of any of the embodiments described above.
[0079] Specifically, the computer executable program 13 used to perform the operations of the present invention may be assembly instructions, instruction set architecture (ISA) instructions, computer instructions, computer-related instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages.
[0080] For the purposes of this embodiment, the computer-readable storage medium 20 can be any means capable of containing, storing, communicating, propagating, or transmitting a program for use by or in conjunction with an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, the computer-readable storage medium 20 can even be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.
[0081] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system.
[0082] Computer device 10 can be, for example, a server, desktop computer, laptop computer, tablet computer, or smartphone. In some examples, computer device 10 can be a cloud acquisition node. Computer device 10 can be described in the general context of computer system executable instructions (such as program modules) executed by a computer system. Typically, program modules can include routines, programs, object programs, components, logic, data structures, etc., that perform specific tasks or implement specific abstract data types. Computer device 10 can be implemented in a distributed cloud acquisition environment where tasks are performed by remote processing devices linked via a communication network. In a distributed cloud acquisition environment, program modules can reside on local or remote acquisition system storage media, including storage devices.
[0083] Computer device 10 may include a processor 12 adapted to execute stored instructions and a memory 11 that provides temporary storage space for the operation of said instructions during operation. Processor 12 may be a single-core processor, a multi-core processor, an acquisition cluster, or any other configuration. Memory 11 may include random access memory (RAM), read-only memory, flash memory, or any other suitable storage system.
[0084] The processor 12 can be connected via a system interconnect (e.g., PCI, PCI-Express, etc.) to an I / O interface (input / output interface) suitable for connecting the computer device 10 to one or more I / O devices (input / output devices). I / O devices may include, for example, a keyboard and indicating devices, where indicating devices may include a touchpad or touchscreen, etc. I / O devices may be built into the computer device 10 or may be external devices connected to the acquisition device.
[0085] The processor 12 may also be linked via a system interconnect to a display interface suitable for connecting the computer device 10 to a display device. The display device may include a display screen that is a built-in component of the computer device 10. The display device may also include a computer monitor, television, or projector, etc., externally connected to the computer device 10. Furthermore, a network interface controller (NIC) may be adapted to connect the computer device 10 to a network via a system interconnect. In some embodiments, the NIC may use any suitable interface or protocol (such as an Internet Minicomputer System Interface) to transmit data. The network may be a cellular network, a radio network, a wide area network (WAN), a local area network (LAN), or the Internet, etc. Remote devices may connect to the computer device via the network.
[0086] Therefore, those skilled in the art should recognize that although numerous exemplary embodiments of the present invention have been shown and described in detail herein, many other variations or modifications conforming to the principles of the present invention can be directly determined or derived from the disclosure of the present invention without departing from the spirit and scope of the invention. Thus, the scope of the present invention should be understood and construed as covering all such other variations or modifications.
Claims
1. A method for detecting defect types in wafer SEM images, comprising: Identify defects in the wafer SEM image; Extract the defect image corresponding to the defect from the wafer SEM image; The system detects whether the defect image meets preset requirements. If it does, the defect is determined to be a graphic defect; otherwise, the defect is determined to be a non-graphic defect.
2. The defect type detection method for wafer SEM images according to claim 1, wherein... The step of detecting whether the defective image meets the preset requirements includes: Extract normal images adjacent to the defective images from the wafer SEM images to obtain a control image; The similarity between the defective image and the comparison image determines whether the defective image meets the preset requirements.
3. The defect type detection method for wafer SEM images according to claim 2, wherein... The step of determining whether the defective image meets the preset requirements based on the similarity between the defective image and the reference image includes: Extract the image feature values of each pixel in the defective image and the control image respectively; By counting the number of pixels for each of the image feature values, the distribution of image feature values between the defective image and the control image is obtained. The distribution of image feature values is used to characterize the number of pixels corresponding to different numerical ranges of the image feature values in the image. Whether the defective image meets the preset requirements is determined by comparing the image feature value distribution of the defective image and the control image.
4. The defect type detection method for wafer SEM images according to claim 3, wherein... The step of determining whether the defective image meets the preset requirements by comparing the image feature value distribution of the defective image and the control image includes: The distribution graphs of the image feature values of the comparison image and the image feature values of the defective image are obtained in a set coordinate system, wherein the set coordinate system uses the image feature values and the number of pixels as the horizontal and vertical axes, respectively. Calculate the overlap area between the distribution patterns of the control image and the distribution patterns of the defect image; Determine whether the overlapping area is greater than a set area threshold. If yes, the defective image is determined to meet the preset requirements. If no, the defective image is determined not to meet the preset requirements.
5. The defect type detection method for wafer SEM images according to claim 3, wherein... The image feature values include image grayscale values or image internal gradient values.
6. The defect type detection method for wafer SEM images according to claim 2, wherein... The step of obtaining the comparison image includes: Determine the positional relationship of the defect relative to the corresponding reference graphic in the design layout; If the area of the defect located outside the reference pattern is larger than the area located inside the reference pattern, the comparison image is determined to be an image of a normal pattern adjacent to the defect image in the wafer SEM image; If the area of the defect located within the reference pattern is greater than the area located outside the reference pattern, the comparison image is determined to be the background region adjacent to the defect image in the wafer SEM image.
7. The defect type detection method for wafer SEM images according to claim 1, wherein... The step of determining defects in the wafer SEM image includes: Align the wafer SEM image with the design layout; Defects in the wafer SEM image are determined based on the degree of difference between the wafer SEM image and the image of the same location on the design layout.
8. A computer device comprising a memory, a processor, and a computer-executable program stored in the memory and running on the processor, wherein the processor, when executing the computer-executable program, implements the defect type detection method for wafer SEM images according to any one of claims 1 to 7.
9. A computer-readable storage medium having a computer-executable program stored thereon, wherein the computer-executable program, when executed by a processor, implements the defect type detection method for wafer SEM images according to any one of claims 1 to 7.
10. A computer program product comprising a computer executable program, wherein the computer executable program, when executed by a processor, implements the defect type detection method for wafer SEM images according to any one of claims 1 to 7.