Defect detection methods and devices for wafers

Abstract

This invention provides a method and apparatus for wafer defect detection, belonging to the field of semiconductor manufacturing technology. The wafer defect detection method includes: extracting defect boundaries from the wafer to be inspected to determine the morphological information of the defects, including the number of breakpoints, defect distribution density, length, and angle of each defect segment; generating a defect density matrix including multiple feature vectors based on the morphological information, the defect density matrix being used to characterize the density variation and breakpoint distribution of defects; evaluating the defect density matrix using a defect assessment model to obtain a risk score for the wafer; comparing the risk score of the wafer with a preset risk threshold, and identifying wafers with risk scores greater than the risk threshold as wafers with discontinuous special pattern defects. This invention can identify wafers with discontinuous special pattern defects.

Description

Technical Field

[0001] This invention relates to the field of semiconductor manufacturing technology, and in particular to a method and apparatus for detecting defects in wafers. Background Technology

[0002] With the continuous advancement of semiconductor wafer manufacturing technology, higher precision requirements are being placed on the detection and interception of defect morphology on wafer surfaces. During the semiconductor wafer manufacturing process, complex morphological defects may appear on the wafer surface, such as discontinuous special patterns. These defects often exhibit a discontinuous distribution pattern, and due to the discontinuity, complex length, and density characteristics of such special patterns, it is difficult to quantify and define them and train models in actual production.

[0003] Currently, widely used ADC (Automatic Defect Classification) models mainly rely on existing defect morphologies in sample libraries for training and judgment. However, because non-continuous special pattern defect samples are difficult to collect in actual production, and their distribution is complex and difficult to quantify, traditional ADC models cannot accurately identify and effectively intercept such defects, resulting in missed detections or false positives. Summary of the Invention

[0004] This invention provides a method and apparatus for detecting defects in wafers, capable of identifying wafers with discontinuous special pattern defects.

[0005] To achieve the above objectives, the technical solution adopted in the embodiments of the present invention is as follows:

[0006] A method for detecting defects in a wafer, comprising:

[0007] Defect boundaries are extracted from the wafer to be judged to determine the morphological information of the defects of the wafer. The morphological information includes the number of breakpoints, defect distribution density, length and angle of each defect segment.

[0008] A defect density matrix comprising multiple feature vectors is generated based on the morphological information. The defect density matrix is ​​used to characterize the density variation and breakpoint distribution of defects.

[0009] The defect density matrix is ​​evaluated using a defect assessment model to obtain the risk score of the wafer;

[0010] The risk score of the wafer is compared with a preset risk threshold, and wafers with a risk score greater than the risk threshold are identified as wafers with discontinuous special pattern defects.

[0011] In some embodiments, before extracting the defect boundaries of the wafer to be determined, the method further includes:

[0012] The wafers to be tested are screened using an ADC model, and abnormal wafers are intercepted to obtain the wafers to be judged.

[0013] In some embodiments, the number of defect assessment models is multiple, and the step of using the defect assessment models to evaluate the defect density matrix and obtain the risk score of the wafer includes:

[0014] The defect density matrix is ​​input into multiple defect determination models to obtain multiple detection results.

[0015] The multiple detection results are fused according to the weights of the multiple defect determination models to obtain the risk score of the wafer.

[0016] In some embodiments, the method further includes:

[0017] The manufacturing process parameters of the wafer to be determined are matched with the manufacturing process parameters of the first wafer in the historical database;

[0018] The risk threshold is adjusted based on the matching degree between the manufacturing process parameters of the first wafer and the manufacturing process parameters of the second wafer, wherein the first wafer is determined to have a non-continuous special pattern defect.

[0019] In some embodiments, adjusting the risk threshold based on the matching degree between the manufacturing process parameters of the first wafer and the manufacturing process parameters of the second wafer includes:

[0020] If the matching degree between the manufacturing process parameters of the wafer and the manufacturing process parameters of the first wafer is greater than or equal to a preset first matching degree, the risk threshold is reduced.

[0021] If the matching degree between the manufacturing process parameters of the wafer and the manufacturing process parameters of the first wafer is less than or equal to a preset second matching degree, the risk threshold is increased.

[0022] Wherein, the second matching degree is less than the first matching degree.

[0023] In some embodiments, after determining that wafers with a risk score greater than the risk threshold are wafers with discontinuous special pattern defects, the method further includes:

[0024] The wafers identified as having discontinuous special pattern defects are used as sample data for the defect determination model. The sample data is then amplified to generate discontinuous special pattern defect samples with different distribution characteristics, and the sample database of the defect determination model is updated.

[0025] The defect determination model is trained using the updated sample database.

[0026] In some embodiments, the step of extracting defect boundaries from the wafer to be judged and determining the morphological information of the defects in the wafer includes:

[0027] Multiple filters are used to extract the boundaries of discontinuous special pattern defects on the wafer. The curve shape of the defects is obtained by least squares method and feature point fitting algorithm. The length start point and end point of the defects, as well as the contour and breakpoint distribution of discontinuous special pattern defects are determined.

[0028] This invention also provides a wafer defect detection device, comprising:

[0029] The extraction module is used to extract the defect boundaries of the wafer to be judged and determine the morphological information of the defects of the wafer. The morphological information includes the number of breakpoints, defect distribution density, length and angle of each defect segment.

[0030] The generation module is used to generate a defect density matrix including multiple feature vectors based on the morphology information. The defect density matrix is ​​used to characterize the density variation and breakpoint distribution of defects.

[0031] The evaluation module is used to evaluate the defect density matrix using a defect determination model to obtain the risk score of the wafer;

[0032] The determination module is used to compare the risk score of the wafer with a preset risk threshold, and determine the wafer with a risk score greater than the risk threshold as a wafer with a discontinuous special pattern defect.

[0033] This invention also provides an electronic device, including: a processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the steps of the wafer defect detection method as described above.

[0034] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the wafer defect detection method described above.

[0035] This invention also provides a computer program product, including computer instructions, which, when executed by a processor, implement the steps of the wafer defect detection method as described above.

[0036] The beneficial effects of this invention are:

[0037] In this embodiment, by introducing a defect density matrix and various feature vectors, a more detailed description of the morphology of special defects can be achieved. This enables the defect determination model to accurately distinguish between discontinuous special pattern defect types with different densities and lengths, solving the problem of missed and over-judgment by the ADC model caused by insufficient samples of discontinuous special pattern defects in the existing wafer surface defect interception process. This ensures effective identification and interception of discontinuous special pattern defects on the wafer surface, while improving the accuracy and efficiency of detection. Attached Figure Description

[0038] Figure 1 A schematic flowchart illustrating the wafer defect detection method according to an embodiment of the present invention;

[0039] Figure 2 This is a schematic diagram illustrating the generation of discontinuous special defect samples with different distribution characteristics according to an embodiment of the present invention;

[0040] Figure 3 A schematic diagram showing the structure of a wafer defect detection device according to an embodiment of the present invention;

[0041] Figure 4 This is a schematic diagram showing the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention are within the scope of protection of the present invention.

[0043] When faced with the defect of discontinuous special patterns, traditional ADC models suffer from the following technical bottlenecks:

[0044] 1. Insufficient sample size: Non-continuous special pattern samples are difficult to collect in actual production. The small sample size leads to insufficient generalization ability of model training and poor judgment effect.

[0045] 2. Difficulty in quantifying morphology: Due to the discontinuous distribution of special pattern defects, traditional morphology parameters (such as length and width) cannot effectively describe their overall characteristics, which makes the model prone to making incorrect judgments when faced with changes in the morphology of the defect distribution.

[0046] 3. Lack of dynamic interception strategy: Existing ADC models usually use fixed thresholds for defect judgment, which cannot adapt to the changes in defect morphology caused by process fluctuations during wafer production, resulting in the inability to effectively intercept potentially risky wafers in actual production.

[0047] 4. Errors caused by human intervention: Traditional ADC models rely on human intervention for sample selection and feature labeling during sample labeling and model training, which can easily introduce human error and reduce the overall accuracy and stability of the judgment.

[0048] Therefore, there is an urgent need for a wafer defect detection solution to improve the sample diversity and model judgment capability of discontinuous special pattern defects, and to achieve accurate quantification and real-time interception of complex defect morphology on the wafer surface, thereby ensuring the overall quality stability of wafer shipments.

[0049] This invention provides a method and apparatus for detecting defects in wafers, capable of identifying wafers with discontinuous special pattern defects.

[0050] This invention provides a method for detecting defects in a wafer, such as... Figure 1 As shown, it includes:

[0051] Step 101: Extract the defect boundaries of the wafer to be judged and determine the morphological information of the defects of the wafer. The morphological information includes the number of breakpoints, defect distribution density, length and angle of each defect segment.

[0052] In this embodiment, multiple filters are used to extract the boundary of discontinuous special pattern defects on the wafer, and the curve shape of the defects is obtained by the least squares method and feature point fitting algorithm. The length start point and end point of the defects, as well as the outline and breakpoint distribution of discontinuous special pattern defects, are determined.

[0053] In this embodiment, image segmentation, edge detection, and feature quantization methods can be used to process the defect region of suspected discontinuous special pattern and extract the microscopic morphological features of the discontinuous special defect.

[0054] In some embodiments, before extracting the defect boundaries of the wafer to be determined, the method further includes:

[0055] The wafers to be tested are screened using an ADC model, and abnormal wafers are intercepted to obtain the wafers to be judged.

[0056] This embodiment can use a traditional ADC model to perform preliminary screening of all wafers, quickly detect wafer surface defects with high significance, and directly intercept wafers with obvious abnormalities, reducing the pressure of subsequent sample labeling and model training.

[0057] Step 102: Generate a defect density matrix including multiple feature vectors based on the morphology information. The defect density matrix is ​​used to characterize the density variation and breakpoint distribution of defects.

[0058] In this embodiment, a multi-dimensional defect density matrix can be established based on morphological information to quantitatively describe the morphology of special defects. The defect density matrix can describe the sparsity or density of discontinuous special defects. It includes multiple feature vectors and can quantitatively describe the morphological features of discontinuous special pattern defects. By performing sparsity analysis on the pixel distribution of each segment, the defect density matrix can describe the density variation and breakpoint distribution of discontinuous special pattern defects. The defect density matrix can reflect the overall morphology of discontinuous special defects and perform segment-by-segment evaluation, providing a basis for subsequent dynamic interception strategies.

[0059] Step 103: Evaluate the defect density matrix using a defect assessment model to obtain the risk score of the wafer;

[0060] In some embodiments, to improve detection accuracy, the number of defect determination models can be multiple. The step of using the defect determination models to evaluate the defect density matrix and obtain the risk score of the wafer includes:

[0061] The defect density matrix is ​​input into multiple defect determination models to obtain multiple detection results.

[0062] The multiple detection results are fused according to the weights of the multiple defect determination models to obtain the risk score of the wafer.

[0063] Multiple defect detection models can be trained independently. The weights of each model can be set according to the actual situation. The detection results from multiple defect detection models are then weighted and fused to obtain the final risk score. These multiple defect detection models can be random forest models, deep learning neural network-based models, etc.

[0064] Step 104: Compare the risk score of the wafer with a preset risk threshold, and determine the wafer with a risk score greater than the risk threshold as a wafer with a discontinuous special pattern defect.

[0065] Specifically, wafers with a risk score greater than the risk threshold are identified as wafers with discontinuous special pattern defects; wafers with a risk score not greater than the risk threshold are identified as wafers without discontinuous special pattern defects.

[0066] In this embodiment, after a wafer is identified as having a discontinuous special pattern defect, an alarm can be automatically triggered and the wafer can be intercepted to prevent unqualified products from flowing into subsequent process stages.

[0067] In this embodiment, by introducing a defect density matrix and various feature vectors, a more detailed description of the morphology of special defects can be achieved. This enables the defect determination model to accurately distinguish between discontinuous special pattern defect types with different densities and lengths, solving the problem of missed and over-judgment by the ADC model caused by insufficient samples of discontinuous special pattern defects in the existing wafer surface defect interception process. This ensures effective identification and interception of discontinuous special pattern defects on the wafer surface, while improving the accuracy and efficiency of detection.

[0068] In actual production, the surface morphology of wafers is affected by process fluctuations, and a fixed risk threshold cannot effectively adapt to morphology changes. In this embodiment, the risk threshold can be dynamically updated to ensure a high interception accuracy under different process conditions. The method also includes:

[0069] The manufacturing process parameters of the wafer to be determined are matched with the manufacturing process parameters of the first wafer in the historical database;

[0070] The risk threshold is adjusted based on the matching degree between the manufacturing process parameters of the first wafer and the manufacturing process parameters of the second wafer, wherein the first wafer is determined to have a non-continuous special pattern defect.

[0071] Specifically, adjusting the risk threshold based on the matching degree between the manufacturing process parameters of the first wafer and the manufacturing process parameters of the second wafer includes:

[0072] If the matching degree between the manufacturing process parameters of the wafer and the manufacturing process parameters of the first wafer is greater than or equal to a preset first matching degree, the risk threshold is reduced.

[0073] If the matching degree between the manufacturing process parameters of the wafer and the manufacturing process parameters of the first wafer is less than or equal to a preset second matching degree, the risk threshold is increased.

[0074] Wherein, the second matching degree is less than the first matching degree.

[0075] To address the issue of a small sample size for discontinuous special pattern defects, in this embodiment, after determining wafers with risk scores greater than the risk threshold as wafers exhibiting discontinuous special pattern defects, the method further includes:

[0076] The wafers identified as having discontinuous special pattern defects are used as sample data for the defect determination model. The sample data is then amplified to generate discontinuous special pattern defect samples with different distribution characteristics, and the sample database of the defect determination model is updated.

[0077] The defect determination model is trained using the updated sample database.

[0078] This embodiment performs data augmentation on a limited set of discontinuous special pattern defect data. It employs image data augmentation methods such as geometric transformations (e.g., rotation, scaling, translation) and random cropping to generate various discontinuous special pattern defect samples with different distribution characteristics, such as... Figure 2 As shown, it can change the appearance characteristics of defects, improve the diversity of the sample set, and make up for the problem of insufficient sample collection in the actual production process.

[0079] This embodiment can also dynamically update the sample library and model parameters based on newly collected defect samples during production. By introducing an automated data cleaning and preprocessing module, new samples are automatically classified and labeled, and merged with the original sample set. The ADC model is periodically updated and retrained online. This significantly improves the system's real-time performance and adaptability, effectively reduces errors caused by manual intervention, further improves the accuracy of identifying discontinuous special pattern defects, and enhances the ability to intercept discontinuous special pattern defects.

[0080] This embodiment reduces the risk of over- and under-judgment in traditional models when the sample size is insufficient, and improves the overall interception capability of risky wafers. It utilizes image data augmentation methods to generate diverse defect samples, and simultaneously introduces multiple machine learning algorithms to train the augmented samples. This allows for model optimization and updates in a shorter time, significantly shortening the model training cycle and effectively avoiding the inconsistencies caused by traditional manual annotation and model training, thus improving the stability of defect detection results. Furthermore, because this embodiment employs automated data processing and algorithm fusion techniques in sample augmentation, model training, and judgment processes, it significantly reduces reliance on manual annotation and sample selection, avoiding the introduction of human error and improving the consistency and stability of model judgments.

[0081] This invention also provides a wafer defect detection device, such as... Figure 3 As shown, it includes:

[0082] Extraction module 21 is used to extract the defect boundaries of the wafer to be judged and determine the morphological information of the defects of the wafer. The morphological information includes the number of breakpoints, defect distribution density, length and angle of each defect segment.

[0083] The generation module 22 is used to generate a defect density matrix including multiple feature vectors based on the morphology information. The defect density matrix is ​​used to characterize the density variation and breakpoint distribution of defects.

[0084] Evaluation module 23 is used to evaluate the defect density matrix using a defect determination model to obtain the risk score of the wafer;

[0085] The determination module 24 is used to compare the risk score of the wafer with a preset risk threshold, and determine the wafer with a risk score greater than the risk threshold as a wafer with a discontinuous special pattern defect.

[0086] In this embodiment, by introducing a defect density matrix and various feature vectors, a more detailed description of the morphology of special defects can be achieved. This enables the defect determination model to accurately distinguish between discontinuous special pattern defect types with different densities and lengths, solving the problem of missed and over-judgment by the ADC model caused by insufficient samples of discontinuous special pattern defects in the existing wafer surface defect interception process. This ensures effective identification and interception of discontinuous special pattern defects on the wafer surface, while improving the accuracy and efficiency of detection.

[0087] In some embodiments, the apparatus further includes:

[0088] The screening module is used to screen the wafers to be detected using the ADC model, and to obtain the wafers to be judged after intercepting abnormal wafers.

[0089] In some embodiments, there are multiple defect determination models. The evaluation module 23 is specifically used to input the defect density matrix into multiple defect determination models respectively to obtain multiple detection results; and to fuse the multiple detection results according to the weights of the multiple defect determination models to obtain the risk score of the wafer.

[0090] In some embodiments, the apparatus further includes:

[0091] An adjustment module is used to match the manufacturing process parameters of the wafer to be judged with the manufacturing process parameters of a first wafer in the historical database; and to adjust the risk threshold according to the matching degree between the manufacturing process parameters of the wafer and the manufacturing process parameters of the first wafer, wherein the first wafer is the wafer judged to have a discontinuous special pattern defect.

[0092] In some embodiments, the adjustment module is specifically used to reduce the risk threshold if the matching degree between the wafer's manufacturing process parameters and the first wafer's manufacturing process parameters is greater than or equal to a preset first matching degree; and to increase the risk threshold if the matching degree between the wafer's manufacturing process parameters and the first wafer's manufacturing process parameters is less than or equal to a preset second matching degree.

[0093] Wherein, the second matching degree is less than the first matching degree.

[0094] In some embodiments, the apparatus further includes:

[0095] The update module is used to take the wafers determined to have discontinuous special pattern defects as sample data of the defect determination model, perform data amplification processing on the sample data to generate discontinuous special pattern defect samples with different distribution characteristics, update the sample database of the defect determination model, and train the defect determination model using the updated sample database.

[0096] In some embodiments, the extraction module 21 is specifically used to extract the boundary of discontinuous special pattern defects on the wafer using multiple filters, and to obtain the curve shape of the defect by least squares method and feature point fitting algorithm, and to determine the length start point and end point of the defect, as well as the outline and breakpoint distribution of the discontinuous special pattern defect.

[0097] Please refer to Figure 4The present invention also provides an electronic device 30, including a processor 31, a memory 32, and a computer program stored in the memory 32 and executable on the processor 31. When the computer program is executed by the processor 31, it implements the various processes of the above-described wafer defect detection method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0098] This invention also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the various processes of the above-described wafer defect detection method embodiments and achieves the same technical effects. To avoid repetition, it will not be described again here. The computer-readable storage medium includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media 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 memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage, or any other non-transmission medium that can be used to store information accessible to the terminal device under test. As defined in this article, computer-readable storage media do not include transient media, such as modulated data signals and carrier waves.

[0099] This application also provides a computer program product, including computer instructions, which, when executed by a processor, implement the above-described... Figure 1 The various processes of the method embodiments shown can achieve the same technical effect, and will not be described again here to avoid repetition.

[0100] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0101] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0102] In the various method embodiments of this disclosure, the sequence numbers of each step are not intended to limit the order of the steps. For those skilled in the art, any changes in the order of the steps are within the scope of protection of this disclosure without any creative effort.

[0103] It should be noted that the various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, since the embodiments are basically similar to the product embodiments, the descriptions are relatively simple, and the relevant parts can be referred to the descriptions of the product embodiments.

[0104] The above description represents the preferred embodiments of this disclosure. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles described herein, and these improvements and modifications should also be considered within the scope of protection of this disclosure.

Claims

1. A method for detecting defects in a wafer, characterized in that, include: Defect boundaries are extracted from the wafer to be judged to determine the morphological information of the defects of the wafer. The morphological information includes the number of breakpoints, defect distribution density, length and angle of each defect segment. A defect density matrix comprising multiple feature vectors is generated based on the morphological information. The defect density matrix is ​​used to characterize the density variation and breakpoint distribution of defects. The defect density matrix is ​​evaluated using a defect assessment model to obtain the risk score of the wafer; The risk score of the wafer is compared with a preset risk threshold, and the wafer with a risk score greater than the risk threshold is determined to be a wafer with a discontinuous special pattern defect; The number of defect assessment models is multiple, and the evaluation of the defect density matrix using the defect assessment models to obtain the risk score of the wafer includes: The defect density matrix is ​​input into multiple defect determination models to obtain multiple detection results. The multiple detection results are fused according to the weights of the multiple defect determination models to obtain the risk score of the wafer; The method further includes: The manufacturing process parameters of the wafer to be determined are matched with the manufacturing process parameters of the first wafer in the historical database; The risk threshold is adjusted based on the matching degree between the manufacturing process parameters of the first wafer and the manufacturing process parameters of the second wafer, wherein the first wafer is determined to have a non-continuous special pattern defect.

2. The wafer defect detection method according to claim 1, characterized in that, Before extracting the defect boundaries of the wafer to be judged, the method further includes: The wafers to be tested are screened using an ADC model, and abnormal wafers are intercepted to obtain the wafers to be judged.

3. The wafer defect detection method according to claim 1, characterized in that, The step of adjusting the risk threshold based on the matching degree between the manufacturing process parameters of the first wafer and the manufacturing process parameters of the second wafer includes: If the matching degree between the manufacturing process parameters of the wafer and the manufacturing process parameters of the first wafer is greater than or equal to a preset first matching degree, the risk threshold is reduced. If the matching degree between the manufacturing process parameters of the wafer and the manufacturing process parameters of the first wafer is less than or equal to a preset second matching degree, the risk threshold is increased. Wherein, the second matching degree is less than the first matching degree.

4. The wafer defect detection method according to claim 1, characterized in that, After determining that wafers with a risk score greater than the risk threshold are wafers with discontinuous special pattern defects, the method further includes: The wafers identified as having discontinuous special pattern defects are used as sample data for the defect determination model. The sample data is then amplified to generate discontinuous special pattern defect samples with different distribution characteristics, and the sample database of the defect determination model is updated. The defect determination model is trained using the updated sample database.

5. The wafer defect detection method according to claim 1, characterized in that, The step of extracting defect boundaries from the wafer to be evaluated and determining the morphological information of the defects in the wafer includes: Multiple filters are used to extract the boundaries of discontinuous special pattern defects on the wafer. The curve shape of the defects is obtained by least squares method and feature point fitting algorithm, and the length start point, end point, contour and breakpoint distribution of the discontinuous special pattern defects are determined.

6. A wafer defect detection device, characterized in that, include: The extraction module is used to extract the defect boundaries of the wafer to be judged and determine the morphological information of the defects of the wafer. The morphological information includes the number of breakpoints, defect distribution density, length and angle of each defect segment. The generation module is used to generate a defect density matrix including multiple feature vectors based on the morphology information. The defect density matrix is ​​used to characterize the density variation and breakpoint distribution of defects. The evaluation module is used to evaluate the defect density matrix using a defect determination model to obtain the risk score of the wafer; The determination module is used to compare the risk score of the wafer with a preset risk threshold, and determine the wafer with a risk score greater than the risk threshold as a wafer with a discontinuous special pattern defect; An adjustment module is used to match the manufacturing process parameters of the wafer to be judged with the manufacturing process parameters of a first wafer in a historical database; and to adjust the risk threshold according to the matching degree between the manufacturing process parameters of the wafer and the manufacturing process parameters of the first wafer, wherein the first wafer is the wafer judged to have a discontinuous special pattern defect. The number of defect determination models is multiple. The evaluation module is specifically used to input the defect density matrix into multiple defect determination models respectively to obtain multiple detection results; and to fuse the multiple detection results according to the weights of the multiple defect determination models to obtain the risk score of the wafer.

7. An electronic device, characterized in that, include: A processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the steps of the wafer defect detection method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the wafer defect detection method as described in any one of claims 1 to 5.

9. A computer program product, characterized in that, It includes computer instructions that, when executed by a processor, implement the steps of the wafer defect detection method as described in any one of claims 1 to 5.

Images