Wafer defect detection method, device, equipment, medium and product

Abstract

The embodiments of the present disclosure disclose a wafer defect detection method, device, equipment, medium and product, comprising: acquiring a wafer image to be detected; the wafer image to be detected includes texture and defects; identifying the texture of the wafer image to be detected according to a preset texture recognition algorithm to obtain a texture recognition result; and determining a detection result of the defects according to the texture recognition result. The technical scheme eliminates the influence of the texture on the wafer defect recognition, improves the accuracy and recognition efficiency of the wafer defect detection on the wafer surface, and improves the yield of the product.

Description

TECHNICAL FIELD

[0001] Embodiments of the present disclosure relate to the technical field of intelligent detection, and in particular to a wafer defect detection method, device, equipment, medium and product. BACKGROUND

[0002] With the deepening and popularization of industrial automation and intelligentization, the automatic optical inspection (AOI) equipment has replaced the traditional manual inspection and has become a trend of technology development. At present, wafer defect detection is an important subfield. The wafer post-process detection problem is the most prominent. In the detection process, most wafers have been processed by multiple processes and are attached with multiple materials. Due to the variety of materials, the surface forms various textures due to the diversification of plating materials. The texture on a specific material coating has certain characteristics, but due to the different angles at which the actual measured wafers are placed and the different surface flatness, the distribution position of the texture on the wafer surface is irregular. Unlike the ordinary color difference, there is a certain gradual change in gray scale, and the texture color difference has certain randomness and anisotropy.

[0003] In the field of traditional semiconductor process equipment, the pattern wafer (wafer forming circuit pattern) is detected by the D2R (die-to-reference) detection method. This method takes a die or a shot as the smallest repeating unit, collects images of about ten smallest repeating units, trains a mean value map and a variance map, and provides a reference for subsequent detection.

[0004] The wafer with a reflective feature forms some randomly distributed textures, and this kind of color difference will cause the value of the variance map to be very large in the training process. Further, it will cause difficulties in wafer defect detection. After the variance fluctuation becomes large, the method considers that the case where the difference from the mean value map is large is also normal, which causes the detection sensitivity to decrease and the wafer key defect to be missed. After reducing the influence of the variance fluctuation, a large number of over-inspections are caused due to the texture color difference on the surface, so that the work of the machine user in inspection, evaluation and audit is difficult to perform. SUMMARY

[0005] Embodiments of the present disclosure provide a wafer defect detection method, device, equipment, medium and product, which eliminate the influence of texture on wafer defect recognition and improve the accuracy of wafer defect detection.

[0006] In a first aspect, a wafer defect detection method is provided, and the method comprises:

[0007] obtaining a wafer image to be detected; the wafer image to be detected includes texture and defects;

[0008] identify the texture of the wafer image to be detected according to a preset texture recognition algorithm, to obtain a texture recognition result;

[0009] determine the detection result of the defect according to the texture recognition result.

[0010] In a second aspect, a wafer defect detection device is provided, comprising:

[0011] an image acquisition module configured to acquire a wafer image to be detected, wherein the wafer image to be detected comprises texture and defects;

[0012] a texture recognition module configured to identify the texture of the wafer image to be detected according to a preset texture recognition algorithm, to obtain a texture recognition result;

[0013] a defect detection result determination module configured to determine the detection result of the defect according to the texture recognition result.

[0014] In a third aspect, an electronic device is provided, comprising:

[0015] at least one processor; and

[0016] a memory communicatively connected to the at least one processor; wherein

[0017] the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the wafer defect detection method according to the first aspect.

[0018] In a fourth aspect, a computer-readable storage medium is provided, which stores a computer program executable by a processor to implement the wafer defect detection method according to the first aspect.

[0019] In a fifth aspect, a computer program product is provided, which comprises a computer program executable by a processor to implement the wafer defect detection method according to the first aspect.

[0020] The embodiments of the present disclosure disclose a wafer defect detection method, device, equipment, medium and product, comprising: acquiring a wafer image to be detected; the wafer image to be detected comprises texture and defects; identifying the texture of the wafer image to be detected according to a preset texture recognition algorithm, to obtain a texture recognition result; determining the detection result of the defect according to the texture recognition result. The technical solution eliminates the influence of texture on wafer defect recognition by identifying the texture of the wafer image to be detected according to a preset texture recognition algorithm, and further determines the wafer defect detection result, improves the accuracy and recognition efficiency of wafer defect detection on the wafer surface, and improves the yield of products.

[0021] It should be understood that the content described in this part is not intended to identify key or important features of the embodiments of the present disclosure, nor to limit the scope of the embodiments of the present disclosure. Other features of the embodiments of the present disclosure will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings needed to be used in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present disclosure, and other drawings can be obtained by those skilled in the art without creative effort on the basis of these drawings.

[0023] Figure 1 is a flowchart of a wafer defect detection method provided by an embodiment of the present disclosure;

[0024] Figure 2 is a schematic diagram of a test image provided by an embodiment of the present disclosure;

[0025] Figure 3 is a schematic diagram of a mean value map provided by an embodiment of the present disclosure;

[0026] Figure 4 is a schematic diagram of a preset texture reference map provided by an embodiment of the present disclosure;

[0027] Figure 5 is a schematic diagram of a detection result meeting the over-check requirement provided by an embodiment of the present disclosure;

[0028] Figure 6 is a schematic diagram of a detection result provided by an embodiment of the present disclosure;

[0029] Figure 7 is a schematic diagram of an execution process of a wafer defect detection method provided by an embodiment of the present disclosure;

[0030] Figure 8 is a structural schematic diagram of a wafer defect detection device provided by an embodiment of the present disclosure;

[0031] Figure 9 is a structural schematic diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0032] In order for those skilled in the art to better understand the solutions of the embodiments of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, not all. Based on the embodiments in the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without creative labor should fall within the scope of protection of the present disclosure.

[0033] It should be noted that the terms "first", "second" and the like in the description and claims of the embodiments of the present disclosure and the above drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or a chronological sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present disclosure described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0034] Embodiment one

[0035] Figure 1 A flowchart of a wafer defect detection method provided by the first embodiment of the present disclosure, the present embodiment can be applicable to the case of detecting wafer defects, which can be executed by a wafer defect detection device. The wafer defect detection device can be realized in the form of hardware and / or software, and can be configured in an electronic device, such as a computer, a computer, a terminal, a server and other devices with data processing capability. As shown in the figure, the method comprises: Figure 1

[0036] S110, obtaining a wafer image to be detected; the wafer image to be detected includes texture and defects;

[0037] In the present embodiment, the wafer image to be detected can be a wafer image containing texture and defects. The texture of the wafer can refer to the microstructure of the wafer surface. The defects of the wafer can be defects or flaws that do not meet the requirements on the wafer surface or inside in the semiconductor manufacturing process. These defects can have a serious impact on the performance and reliability of the wafer, so it is very important to classify and analyze the wafer defects. Wafer defects can be classified into several categories, such as impurities, cracks, pollution, etc

[0038] S120, identifying the texture of the wafer image to be detected according to a preset texture identification algorithm to obtain a texture identification result;​

[0039] In this embodiment, the preset texture recognition algorithm can be an algorithm for identifying textures in advance. The texture of the wafer image to be detected can be identified according to the preset texture recognition algorithm, and a texture recognition result of the wafer image to be detected can be obtained.

[0040] In S130, a detection result of the defect is determined according to the texture recognition result.

[0041] Specifically, the wafer image to be detected includes textures and defects. After obtaining the texture recognition result, a defect detection result of the wafer image to be detected can be determined. For example, after identifying the texture recognition result of the wafer image to be detected, the un-identified result is the detection result of the defect.

[0042] The embodiment of the present disclosure discloses a wafer defect detection method, including: obtaining a wafer image to be detected; the wafer image to be detected includes textures and defects; identifying the textures of the wafer image to be detected according to a preset texture recognition algorithm, and obtaining a texture recognition result; and determining a detection result of the defect according to the texture recognition result, eliminating the influence of textures on wafer defect recognition, improving the accuracy and recognition efficiency of wafer defect detection on the wafer surface, and improving the yield of products.

[0043] As an optional implementation of the embodiment, the wafer defect detection method for texture detection provided by the embodiment further includes determining target combination parameters of the preset texture recognition algorithm.

[0044] The process of determining the target combination parameters of the preset texture recognition algorithm includes:

[0045] 1) obtaining a test image by using an image acquisition device, and obtaining a mean value image of the test image;

[0046] In this embodiment, the test image can also be obtained by using an image acquisition device. The image acquisition device can be a camera, and the test image can be an image captured by the camera. After obtaining the test image, the test image can be processed by using a mean value image acquisition algorithm to obtain a mean value image of the test image. For example, the mean value image can be generated by calculating the average value of each pixel position in all images. The mean value image can also be obtained by using an image processing library that provides a direct calculation function of the mean value image.

[0047] Figure 2 A schematic diagram of a test image provided by the embodiment is shown in FIG. 1. Figure 2 As shown in FIG. 1, the test image can be obtained by using a camera. Figure 3 A schematic diagram of a mean value image provided by the embodiment is shown in FIG. 2. Figure 3As shown, the mean map can be obtained by processing the test image using a mean map acquisition algorithm. The mean map does not contain texture.

[0048] 2) Determine a preset texture reference image using the test image and the mean image;

[0049] Specifically, the difference between the mean image and the test image can be used as a preset texture reference image.

[0050] Figure 4 This is a schematic diagram of a preset texture reference image provided in this embodiment, such as... Figure 4 As shown, the red highlight indicates the location of the texture. Figure 5 This embodiment provides a schematic diagram of a detection result that meets the requirements for passing the inspection. Figure 5 It can be seen that, Figure 4 In addition to the texture location, other texture locations were also detected.

[0051] 3) Determine the combined parameters based on the image processing parameters and the convolution kernel processing parameters; the combined parameters include an image size, an image grayscale, and a convolution kernel size;

[0052] In this embodiment, the image processing parameters can be parameters of the image processing process. For example, image processing parameters can include image size and image grayscale, and can include at least one set of image size and image grayscale. Image size can be the width and height of the image, and is typically measured in pixels, such as 800×600. Image grayscale can be the number of different gray levels in the image, which determines the amount of detail in the image's brightness and darkness; image grayscale typically includes 256 (0-255) gray levels. The convolution kernel processing parameters can include at least one convolution kernel size, which can be the size of the convolution kernel. It should be noted that the convolution kernel processing parameters can also include information such as the type of convolution kernel.

[0053] As described above, the combined parameters can be determined based on the image processing parameters and the convolution kernel processing parameters. The combined parameters include an image size, an image grayscale, and a convolution kernel size.

[0054] 4) For each of the combined parameters, the test image is detected based on the convolution kernel of the machine learning algorithm and the combined parameters to obtain the detection result;

[0055] It is known that, after determining multiple combination parameters, for any set of combination parameters, the size of the convolution kernel in the machine learning algorithm can be set to the size of the convolution kernel in the combination parameters, and the image size and image grayscale can be set to the image size and image grayscale in the combination parameters. This allows detection to be performed on the test image to obtain the detection result. The test image can be an image acquired by an image acquisition device, such as a camera; the detection result can be the detected texture, which can be highlighted.

[0056] For example, in the combined parameters, the image size is 100×50, the image grayscale is 5, and the convolution kernel size is one-dimensional convolution kernel × one-dimensional convolution kernel. The image can be processed based on the image size and the image grayscale, and the convolution kernel of the machine learning algorithm can be set to one-dimensional convolution kernel × one-dimensional convolution kernel.

[0057] 5) The detection results are evaluated using an evaluation system and a preset texture reference image to obtain an evaluation result;

[0058] In this embodiment, the evaluation system can be a standard for evaluating the detection results. For example, the evaluation system can be a scoring rule. The preset texture reference map can be a reference map with known texture positions. For example, the preset texture reference map can be a mean map, which can be generated by calculating the average value of each pixel position in all images; or the mean map can be obtained by an image processing library that provides the function of directly calculating the mean map.

[0059] Specifically, by using a preset texture reference image and a test image, the first texture position of the test image can be obtained; the second texture position, detected by combining parameters and a machine learning algorithm, can be obtained from the detection results. After obtaining the first and second texture positions, the overlap between the first and second texture positions can be calculated, and the overlap can be scored using an evaluation system to obtain an evaluation result. For example, if the overlap between the first and second texture positions is 75%, the evaluation result is 75 points.

[0060] 6) Determine target combination parameters based on the evaluation results corresponding to each of the aforementioned combination parameters and the actual detection requirements, and use the target combination parameters to construct a preset texture recognition algorithm.

[0061] Specifically, by combining various parameters and machine learning algorithms, corresponding evaluation results can be obtained. Based on these evaluation results and actual detection requirements, target combination parameters can be determined. Among these, textures exist in different categories, and the evaluation results can determine the sensitivity of each combination parameter to different types of textures.

[0062] As described above, the actual detection requirements may include pre-set over-detection and under-detection requirements for preset texture categories; wherein, the over-detection requirement is the requirement that the evaluation result is higher than a preset threshold; the under-detection requirement is the requirement that the evaluation result is lower than a preset threshold, the preset threshold being a pre-set threshold, for example, the preset threshold can be 100, if the evaluation result is 125, it is an over-detection requirement, which can be understood as the current combination of parameters being able to detect textures other than those in the preset texture reference image; if the evaluation result is 80, it is an under-detection requirement, which can be understood as the current combination of parameters being able to detect some textures in the preset texture reference image.

[0063] For example, the target combination function can be determined according to the detection requirements of different types of textures, or it can be determined according to the actual detection requirements of the preset texture category, such as the need to construct the preset texture category.

[0064] As an optional implementation of this embodiment, the image processing parameters include at least one set of image sizes and image grayscale; the convolution kernel processing parameters include at least one convolution kernel size;

[0065] The determination of combined parameters based on image processing parameters and convolution kernel processing parameters includes:

[0066] Each image processing parameter and each convolution kernel processing parameter are matched separately to obtain multiple combined parameters.

[0067] Specifically, each image processing parameter and each convolution kernel processing parameter can be matched separately to obtain multiple combined parameters. These combined parameters can include a set of image dimensions, an image grayscale level, and the size of the convolution kernel.

[0068] As an optional implementation of this embodiment, the detection of the test image by the convolution kernel based on the machine learning algorithm and the combined parameters to obtain the detection result includes:

[0069] 1) Perform image preprocessing on the test image using the image size and image grayscale to determine the preprocessed test image;

[0070] Specifically, after determining the image size and grayscale, image preprocessing can be performed on the test image based on these parameters to determine the final test image. It should be noted that the image preprocessing process can also include: splitting the color channels and performing data augmentation on both the mean image and the image to be detected from multiple dimensions, enabling the effective information of the image to be identified from multiple dimensions.

[0071] 2) Determine the convolution kernel of the machine learning algorithm based on the convolution kernel size;

[0072] Specifically, after the convolution kernel size is determined, the size of the convolution kernel for the machine learning algorithm can be set based on the convolution kernel size. It should be noted that the type of convolution kernel can also be selected. The types of convolution kernels can include gradient kernels, point kernels, line kernels, ripple kernels, and oscillating kernels, etc.

[0073] 3) The preprocessed test image is detected by a convolutional kernel based on a machine learning algorithm to obtain the detection result, which includes the location of the texture.

[0074] As we know, once the convolution kernel is determined, the test image can be tested based on the convolution kernel of the machine learning algorithm. This yields the detection results, which include texture locations, which can be displayed through highlighting. Figure 6 This is a schematic diagram of a detection result provided in this embodiment, such as... Figure 6 As shown, the red highlighted areas represent the detected texture locations.

[0075] As an optional implementation of this embodiment, the step of evaluating the detection results using an evaluation system and a preset texture reference image to obtain an evaluation result includes:

[0076] 1) Determine the degree of overlap between the texture positions of the preset texture reference image and the detection result; the texture in the preset texture reference image is a preset texture category;

[0077] Specifically, after obtaining the detection result, the degree of overlap between the texture position of the preset texture reference image and the texture position of the detection result can be determined. For example, the degree of overlap is 75%. The texture in the preset texture reference image is a preset texture category, which can be a pre-determined texture category.

[0078] 2) Based on the evaluation system, the degree of overlap of the texture positions is quantified to obtain the evaluation result.

[0079] In this embodiment, the evaluation system can be a scoring rule, which can be determined based on historical detection results. Based on the evaluation system, the degree of texture position overlap can be quantified, thereby obtaining an evaluation result. For example, if the texture position overlap is 75%, after quantification by the evaluation system, the evaluation result is a score, such as 75.

[0080] As an optional implementation of this embodiment, determining the target combination parameters based on the evaluation results corresponding to each combination parameter and the actual detection requirements includes:

[0081] 1) Determine initial combination parameters based on the actual detection requirements; the initial combination parameters include at least one set of combination parameters; the actual detection requirements include pre-set over-detection requirements and under-detection requirements for the preset texture category; the over-detection requirement is the requirement that the evaluation result is higher than a preset threshold; the under-detection requirement is the requirement that the evaluation result is lower than a preset threshold;

[0082] Specifically, actual detection requirements can be based on the actual operating environment. For example, in time detection, texture over-detection is required, and the actual detection requirement can be an over-detection requirement. After obtaining the evaluation results, the sensitivity of each combined parameter to the preset texture category can be obtained. The higher the evaluation result of the combined parameter, the higher the sensitivity of the combined parameter to the preset texture category; the lower the evaluation result of the combined parameter, the lower the sensitivity of the combined parameter to the preset texture category. The initial combined parameters can be determined through the actual detection requirements; the initial combined parameters include at least one set of combined parameters.

[0083] For example, if the actual testing requirement is an over-test requirement, the requirement in the evaluation result that is higher than the preset threshold can be used as the initial combination parameter.

[0084] 2) Select the combination parameters that meet the preset conditions from the initial combination parameters as the target combination parameters.

[0085] Specifically, after obtaining the initial combination parameters, the initial combination parameters can be filtered based on preset conditions, and the combination parameters that meet the preset conditions can be used as target combination parameters. The preset conditions can be pre-set filtering conditions. For example, the preset conditions can be to select the combination parameter with the largest corresponding evaluation result among the initial combination parameters as the target combination parameter.

[0086] Figure 7 This is a schematic diagram illustrating the execution process of a wafer defect detection method provided in this embodiment, as shown below. Figure 7As shown, a camera captures an image (test image), and a mean image is obtained from the test image. By comparing the two images (test image and mean image), the location and region of the texture (preset texture reference image) are obtained. Then, texture learning is performed on the test image. Image preprocessing (image scaling and grayscale compression) and convolution kernel preprocessing (selection of convolution kernel size and combination of one-dimensional convolution kernels) are performed based on combined parameters. Finally, texture learning is performed on the image to be detected (test image) based on machine learning algorithms to obtain the detection result. Based on the preset texture reference image, a scoring mechanism (evaluation system) is set according to the texture location and recorded. At the same time, different image sizes, image grayscales, and convolution kernel sizes can be matched separately. The image to be detected (test image) is repeatedly tested according to each matched combination parameter to obtain multiple scoring results. The parameters with the highest scores (combined parameters) are selected as the learning results (and the detection results are evaluated to obtain evaluation results, and the target combination parameters are determined based on the evaluation results and actual detection requirements). After the target combination parameters are determined, this recognition step is added to the subsequent text detection process. This increases the detection capability and improves the over-detection rate, especially when D2R detection is already sensitive. By recognizing the texture in the image being detected and then filtering the detection results based on location information, a single detection is completed. This allows the D2R detection method to maintain high sensitivity and a good over-detection rate even in high color difference (texture) scenes.

[0087] Example 2

[0088] Figure 8 This is a schematic diagram of the structure of a wafer defect detection device provided in Embodiment 2 of this disclosure; as shown Figure 8 As shown, the device includes: an image acquisition module 210, a texture recognition module 220, and a defect detection result determination module 230.

[0089] The image acquisition module 210 is used to acquire an image of the wafer to be inspected; the image of the wafer to be inspected includes texture and defects.

[0090] The texture recognition module 220 is used to recognize the texture of the wafer image to be detected according to a preset texture recognition algorithm, and obtain the texture recognition result;

[0091] The defect detection result determination module 230 is used to determine the detection result of the defect based on the texture recognition result.

[0092] Embodiment 2 of this disclosure provides a wafer defect detection device that eliminates the influence of texture on wafer defect identification and improves the accuracy of wafer defect detection.

[0093] Furthermore, the device further includes: a target combination parameter determination module; the target combination parameter determination module includes:

[0094] An image acquisition unit is used to acquire test images using an image acquisition device and to acquire the mean image of the test images;

[0095] A preset texture reference map determination unit is used to determine a preset texture reference map using the test image and the mean map;

[0096] A combined parameter determination unit is used to determine combined parameters based on image processing parameters and convolution kernel processing parameters; the combined parameters include an image size, an image grayscale, and a convolution kernel size.

[0097] The detection unit is used to detect the test image based on the convolution kernel of the machine learning algorithm and the combined parameters for each of the combined parameters, and obtain the detection result;

[0098] The evaluation result determination unit is used to evaluate the detection results using an evaluation system and a preset texture reference image to obtain an evaluation result;

[0099] The construction unit is used to determine the target combination parameters based on the evaluation results and actual detection requirements corresponding to each of the combined parameters, so as to construct a preset texture recognition algorithm using the target combination parameters.

[0100] Furthermore, the image processing parameters include at least one set of image dimensions and image grayscale; the convolution kernel processing parameters include at least one convolution kernel size;

[0101] The combined parameter determination unit is also used for

[0102] Each image processing parameter and each convolution kernel processing parameter are matched separately to obtain multiple combined parameters.

[0103] Furthermore, the detection unit is also used for

[0104] The test image is preprocessed using the image size and image grayscale to determine the preprocessed test image;

[0105] The convolution kernel of the machine learning algorithm is determined based on the convolution kernel size;

[0106] A convolutional kernel based on a machine learning algorithm is used to detect the preprocessed test image to obtain detection results, which include the location of textures.

[0107] Furthermore, the evaluation result determination unit is also used for:

[0108] Determine the degree of overlap between the texture positions of the preset texture reference image and the detection result; the texture in the preset texture reference image is a preset texture category;

[0109] Based on the evaluation system, the degree of overlap of the texture positions is quantified to obtain the evaluation result.

[0110] Furthermore, building blocks are also used for:

[0111] The initial combination parameters are determined based on the actual detection requirements; the initial combination parameters include at least one set of combination parameters; the actual detection requirements include pre-set over-detection requirements and under-detection requirements for the preset texture category; the over-detection requirement is the requirement that the evaluation result is higher than a preset threshold; the under-detection requirement is the requirement that the evaluation result is lower than a preset threshold;

[0112] The combination parameters that satisfy the preset conditions from the initial combination parameters are taken as the target combination parameters.

[0113] The wafer defect detection device provided in this disclosure can execute the wafer defect detection method provided in any embodiment of this disclosure, and has the corresponding functional modules and beneficial effects for executing the method.

[0114] Example 3

[0115] Figure 9 A schematic diagram of the structure of an electronic device 10 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the embodiments of the present disclosure described and / or claimed herein.

[0116] like Figure 9 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0117] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0118] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microprocessor, etc. Processor 11 performs the various methods and processes described above, such as wafer defect detection methods.

[0119] In some embodiments, the wafer defect detection method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or mounted on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the wafer defect detection method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the wafer defect detection method by any other suitable means (e.g., by means of firmware).

[0120] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0121] Computer programs for implementing the methods of embodiments of this disclosure may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0122] In the context of embodiments of this disclosure, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0123] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0124] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0125] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0126] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the embodiments of this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of the embodiments of this disclosure can be achieved, and this document does not impose any limitations.

[0127] The specific embodiments described above do not constitute a limitation on the scope of protection of the embodiments disclosed herein. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the embodiments disclosed herein should be included within the scope of protection of the embodiments disclosed herein.

[0128] This disclosure also provides a computer program product, including a computer program and / or instructions, which, when executed by a processor, implements the wafer defect detection method provided in any embodiment of this application.

[0129] In implementing a computer program product, computer program code for performing the operations of the embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0130] Note that the above are merely preferred embodiments and the technical principles applied in this disclosure. Those skilled in the art will understand that this disclosure is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the protection scope of this disclosure. Therefore, although the embodiments of this disclosure have been described in detail above, this disclosure is not limited to the above embodiments. More other equivalent embodiments may be included without departing from the concept of this disclosure, and the scope of this disclosure is determined by the scope of the appended claims.

Claims

1. A method for detecting wafer defects, characterized in that, The method includes: Acquire an image of the wafer to be inspected; the image of the wafer to be inspected includes texture and defects; The texture of the wafer image to be detected is identified according to a preset texture recognition algorithm to obtain the texture recognition result; The detection result of the defect is determined based on the texture recognition result; The method further includes determining the target combination parameters of the preset texture recognition algorithm; The process of determining the target combination parameters of the preset texture recognition algorithm includes: The test image is acquired using an image acquisition device, and the mean image of the test image is obtained. A preset texture reference image is determined using the test image and the mean image; The combined parameters are determined based on image processing parameters and convolution kernel processing parameters; the combined parameters include an image size, an image grayscale, and a convolution kernel size. For each of the combined parameters, the test image is detected based on the convolution kernel of the machine learning algorithm and the combined parameters to obtain the detection result; The detection results are evaluated using an evaluation system and a preset texture reference image to obtain an evaluation result; Target combination parameters are determined based on the evaluation results corresponding to each of the combined parameters and the actual detection requirements, so as to construct a preset texture recognition algorithm using the target combination parameters.

2. The method according to claim 1, characterized in that, The image processing parameters include at least one set of image dimensions and image grayscale; the convolution kernel processing parameters include at least one convolution kernel size; The determination of combined parameters based on image processing parameters and convolution kernel processing parameters includes: Each image processing parameter and each convolution kernel processing parameter are matched separately to obtain multiple combined parameters.

3. The method according to claim 1, characterized in that, The convolutional kernel based on the machine learning algorithm and the combined parameters are used to detect the test image to obtain detection results, including: The test image is preprocessed using the image size and image grayscale to determine the preprocessed test image; The convolution kernel of the machine learning algorithm is determined based on the convolution kernel size; A convolutional kernel based on a machine learning algorithm is used to detect the preprocessed test image to obtain detection results, which include the location of textures.

4. The method according to claim 1, characterized in that, The evaluation result is obtained by evaluating the detection result using an evaluation system and a preset texture reference image, including: Determine the degree of overlap between the texture positions of the preset texture reference image and the detection result; the texture in the preset texture reference image is a preset texture category; Based on the evaluation system, the degree of overlap of the texture positions is quantified to obtain the evaluation result.

5. The method according to claim 4, characterized in that, The determination of the target combination parameters based on the evaluation results corresponding to each combination parameter and the actual detection requirements includes: The initial combination parameters are determined based on the actual detection requirements; the initial combination parameters include at least one set of combination parameters; the actual detection requirements include pre-set over-detection requirements and under-detection requirements for the preset texture category; the over-detection requirement is the requirement that the evaluation result is higher than a preset threshold; the under-detection requirement is the requirement that the evaluation result is lower than a preset threshold; The combination parameters that satisfy the preset conditions in the initial combination parameters are taken as the target combination parameters.

6. A wafer defect detection device, characterized in that, include: The image acquisition module is used to acquire images of the wafer to be inspected. The image of the wafer to be inspected includes texture and defects; The texture recognition module is used to recognize the texture of the wafer image to be detected according to a preset texture recognition algorithm, and obtain the texture recognition result; A defect detection result determination module is used to determine the detection result of the defect based on the texture recognition result; The device further includes: a target combination parameter determination module; the target combination parameter determination module includes: An image acquisition unit is used to acquire test images using an image acquisition device and to acquire the mean image of the test images; A preset texture reference map determination unit is used to determine a preset texture reference map using the test image and the mean map; A combined parameter determination unit is used to determine combined parameters based on image processing parameters and convolution kernel processing parameters; the combined parameters include an image size, an image grayscale, and a convolution kernel size. The detection unit is used to detect the test image based on the convolution kernel of the machine learning algorithm and the combined parameters for each of the combined parameters, and obtain the detection result; The evaluation result determination unit is used to evaluate the detection results using an evaluation system and a preset texture reference image to obtain an evaluation result; The construction unit is used to determine the target combination parameters based on the evaluation results and actual detection requirements corresponding to each of the combined parameters, so as to construct a preset texture recognition algorithm using the target combination parameters.

7. An electronic device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the wafer defect detection method as described in any one of claims 1-5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the wafer defect detection method as described in any one of claims 1-5.

9. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the wafer defect detection method as described in any one of claims 1-5.

Images