Wafer defect detection method and system based on machine vision

Abstract

The invention discloses a wafer defect detection method and system based on machine vision. The method comprises the following steps: collecting a plurality of segmentation image blocks for multiple times of detection of a to-be-detected wafer based on machine vision equipment; determining a real-time defect distribution coefficient corresponding to each segmented image block based on the number of times that each segmented image block is marked as a defective image block in multiple times of detection; and determining an abnormal image block according to the real-time defect distribution coefficient corresponding to each segmented image block and the historical defect distribution coefficient corresponding to each segmented image block. The problem that the wafer manufacturing efficiency is affected due to the fact that abnormal changes of the defect positions cannot be found in time due to the fact that the abnormality of the defect distribution positions cannot be analyzed in the prior art can be solved, the abnormality of the defect distribution positions in the wafer manufacturing process can be accurately analyzed and found in time, the production process is optimized, and the production efficiency is improved. And the production efficiency of the wafer is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of machine vision, and in particular to a wafer defect detection method and system based on machine vision. Background Art

[0002] With the advancement of technology, the demand for chips continues to increase. In the process of chip manufacturing, wafers are the main material. Machine vision is increasingly used in intelligent manufacturing, and its development trends include higher resolution, faster processing speed, stronger intelligence level and better integration capabilities. Especially in the field of intelligent manufacturing, machine vision plays an irreplaceable role.

[0003] During the wafer production process, due to the limitations of technology or the production equipment itself, there are certain inherent defects in the manufacturing of wafers. The defects are caused by the limitations of technology or the production equipment itself. Therefore, the distribution positions of the defective parts on the manufactured wafers are generally not much different. When machine vision detects wafer images, it is easy to obtain defective blocks, but it is unable to analyze the abnormalities in the defect distribution positions, resulting in the inability to timely detect abnormal changes in the defect positions, which in turn affects the production efficiency of the wafers. Summary of the invention

[0004] The present invention provides a wafer defect detection method and system based on machine vision, which are used to solve the problem in the prior art that the abnormality of the defect distribution position cannot be analyzed, resulting in the inability to timely detect abnormal changes in the defect position, affecting the wafer manufacturing efficiency, thereby improving the wafer production efficiency.

[0005] In a first aspect, the present invention provides a wafer defect detection method based on machine vision, comprising:

[0006] Collect multiple segmented blocks for multiple inspections of the wafer to be inspected based on machine vision equipment;

[0007] Determine a real-time defect distribution coefficient corresponding to each segmented image block based on the number of times each segmented image block is marked as a defective image block in the multiple inspections;

[0008] The abnormal block is determined according to the real-time defect distribution coefficient corresponding to each segmented block and the historical defect distribution coefficient corresponding to each segmented block.

[0009] According to the wafer defect detection method based on machine vision provided by the present invention, the method collects multiple segmented blocks for multiple inspections of the wafer to be inspected based on the machine vision device, including:

[0010] Collecting multiple inspection images of the wafer to be inspected based on the machine vision device;

[0011] Each of the detection images is segmented to obtain a plurality of segmented image blocks.

[0012] According to the wafer defect detection method based on machine vision provided by the present invention, the real-time defect distribution coefficient corresponding to each segmented image block is determined based on the number of times each segmented image block is marked as a defective image block in the multiple detections, including:

[0013] Determine a real-time defect coefficient corresponding to each segmented image block based on the number of times each segmented image block is marked as a defective image block in the multiple detections;

[0014] A real-time defect distribution coefficient corresponding to each of the segmented blocks is determined based on the real-time defect coefficient.

[0015] According to the wafer defect detection method based on machine vision provided by the present invention, the real-time defect distribution coefficient corresponding to each segmented image block is determined based on the real-time defect coefficient, including:

[0016] The real-time defect distribution coefficient corresponding to each segmented image block is determined according to the following formula:

[0017]

[0018] Wherein, Bi is the real-time defect coefficient, FB is the real-time defect distribution coefficient, a is the number of segmented blocks, Bp is the mean value of the real-time defect coefficient, and a≥i≥1.

[0019] According to the wafer defect detection method based on machine vision provided by the present invention, before determining the abnormal block according to the real-time defect distribution coefficient corresponding to each segmented block and the historical defect distribution coefficient corresponding to each segmented block, the method further includes:

[0020] Determine a historical defect coefficient corresponding to each of the segmented blocks based on a historical inspection image of the wafer to be inspected;

[0021] The historical defect distribution coefficient is determined using the historical defect coefficient according to the following formula:

[0022]

[0023] Wherein, Wi is the historical defect coefficient, FA is the historical defect distribution coefficient, a is the number of segmented blocks, Wp is the mean value of the historical defect coefficient, and a≥i≥1.

[0024] According to the wafer defect detection method based on machine vision provided by the present invention, the abnormal block is determined according to the real-time defect distribution coefficient corresponding to each segmented block and the historical defect distribution coefficient corresponding to each segmented block, including:

[0025] Determine a difference between a real-time defect distribution coefficient corresponding to each segmented image block and a historical defect distribution coefficient corresponding to each segmented image block;

[0026] The segmented image block whose difference value is greater than a preset abnormal value is determined as the abnormal image block.

[0027] According to the wafer defect detection method based on machine vision provided by the present invention, the wafer defect detection method based on machine vision further includes:

[0028] Collecting working parameters of the abnormal block;

[0029] Based on the historical operating parameters of the abnormal block, the operating parameters of the abnormal block are adjusted.

[0030] In a second aspect, the present invention further provides a wafer defect detection system based on machine vision, comprising:

[0031] An acquisition module, used for acquiring multiple segmented blocks for multiple inspections of the wafer to be inspected based on a machine vision device;

[0032] A determination module, configured to determine a real-time defect distribution coefficient corresponding to each segmented image block based on the number of times each segmented image block is marked as a defective image block in the multiple detections;

[0033] The defect detection module is used to determine the abnormal block according to the real-time defect distribution coefficient corresponding to each segmented block and the historical defect distribution coefficient corresponding to each segmented block.

[0034] In a third aspect, the present invention further provides an electronic device, comprising: a memory for storing a computer software program; and a processor for reading and executing the computer software program, thereby implementing the wafer defect detection method based on machine vision as described in the first aspect above.

[0035] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium, wherein the storage medium stores a computer software program, and when the computer software program is executed by a processor, the machine vision-based wafer defect detection method as described in the first aspect above is implemented.

[0036] In a fifth aspect, the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the wafer defect detection method based on machine vision as described in the first aspect above.

[0037] The machine vision-based wafer defect detection method provided by the present invention first collects multiple segmented blocks for multiple inspections of the wafer to be inspected based on the machine vision equipment, then determines the real-time defect distribution coefficient corresponding to each segmented block based on the number of times each segmented block is marked as a defective block in the multiple inspections, and finally determines the abnormal block according to the real-time defect distribution coefficient corresponding to each segmented block and the historical defect distribution coefficient corresponding to each segmented block. Therefore, it can solve the problem in the prior art that it is impossible to analyze the abnormality of the defect distribution position, resulting in the inability to timely discover the abnormal change of the defect position, thereby affecting the wafer manufacturing efficiency. It can realize accurate analysis and timely discovery of the abnormality of the defect distribution position in the wafer manufacturing process, thereby optimizing the production process, and improving the wafer production efficiency while ensuring the wafer quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a flow chart of a wafer defect detection method based on machine vision provided by the present invention;

[0039] Figure 2 It is a structural schematic diagram of a wafer defect detection system based on machine vision provided by the present invention;

[0040] Figure 3 A schematic diagram of an electronic device according to an embodiment of the present invention;

[0041] Figure 4 A schematic diagram of an embodiment of a computer-readable storage medium provided for an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0043] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0044] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in the present invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present invention.

[0045] Optional, see Figure 1 , Figure 1 : is a flow chart of the wafer defect detection method based on machine vision provided by the present invention. The execution subject of the wafer defect detection method based on machine vision provided by the present invention can be a defect detection system. Therefore, the wafer defect detection method based on machine vision includes the following steps:

[0046] Step 10: Collect multiple segmented image blocks of multiple inspections of the wafer to be inspected based on a machine vision device.

[0047] Optionally, firstly, a plurality of inspection images of the wafer to be inspected are collected by a machine vision device, and then each inspection image is segmented to obtain a plurality of segmented blocks. During the collection, the interval duration of the inspection can also be set, but this is not limited in the embodiment of the present invention.

[0048] When collecting the inspection image, the wafer to be inspected on the workbench can be photographed through machine vision equipment such as cameras, line scanners, CCDs, infrared cameras, 3D cameras, etc., and then the inspection blocks of the wafer to be inspected are obtained for collection to ensure the quality of the collected blocks is reliable. Of course, it can also be collected through other forms, and the embodiments of the present invention are not limited to this.

[0049] After the detection image is collected, it provides a basis for subsequent block processing and analysis. The collected blocks are divided into multiple segmented blocks according to the predetermined segmentation standard, and each segmented block is compared with its preset standard image to obtain the similarity between each segmented block and its preset standard image. The similarity is used to determine whether each segmented block is qualified. If qualified, no processing is performed. If unqualified, it is marked as a defective block. Specifically in the segmentation of fast images, you can choose threshold-based segmentation, edge-based segmentation, clustering-based segmentation, etc. Edge detection is one of the commonly used feature extraction methods in visual recognition. It can effectively detect the contours of objects in the block. Common edge detection algorithms include Sobel operator, Canny operator, etc., and other rules can also be used for block segmentation.

[0050] Step 20: determining a real-time defect distribution coefficient corresponding to each segmented image block based on the number of times each segmented image block is marked as a defective image block in multiple inspections.

[0051] Optionally, based on the number of times each segmented block is marked as a defective block in multiple inspections, a real-time defect coefficient corresponding to each segmented block is determined, and then a real-time defect distribution coefficient corresponding to each segmented block is determined according to the real-time defect coefficient. Specifically, after each predetermined number of inspections b, the inspection results corresponding to each segmented block are obtained, and the number of times each segmented block is judged as a defective block in the predetermined number of inspections b is marked as Ea, and the ratio between the number of times each segmented block is judged as a defective block in the predetermined number of inspections b and the predetermined number of inspections b is marked as the real-time defect coefficient corresponding to each segmented block, that is, the real-time defect coefficient Bi corresponding to each segmented block is calculated by the formula Ea / b=Bi, b is a preset value, which can be selected according to actual needs, b≥1, and then the real-time defect distribution coefficient of each segmented block is obtained by discrete calculation, specifically, the real-time defect distribution coefficient corresponding to each segmented block is determined according to the following formula:

[0052]

[0053] Among them, Bi is the real-time defect coefficient, FB is the real-time defect distribution coefficient, a is the number of segmented blocks, Bp is the mean of the real-time defect coefficient, and the real-time defect distribution coefficient reflects the real-time distribution status of defects, a≥i≥1.

[0054] Step 30 , determining an abnormal block according to the real-time defect distribution coefficient corresponding to each segmented block and the historical defect distribution coefficient corresponding to each segmented block.

[0055] Optionally, firstly, the difference between the real-time defect distribution coefficient corresponding to each segmented block and the historical defect distribution coefficient corresponding to each segmented block is determined, and then the segmented block whose difference is greater than a preset abnormal value is determined as an abnormal block.

[0056] In this step, firstly, the historical defect coefficient corresponding to each segmented block is determined based on the historical inspection image of the wafer to be inspected. Specifically, when determining the historical defect distribution coefficient, firstly, the defect detection results corresponding to each segmented block in the historical inspection times c are obtained, that is, whether it is a defective block, where c is a preset number of times. Then, according to the number of times each segmented block is judged as a defective block in the preset inspection times c, analysis is performed to obtain the historical defect coefficient corresponding to each segmented block. Specifically, firstly, the number of times each segmented block is judged as a defective block in the preset inspection times c is marked as Qa, where a is the number of segmented blocks, and a≥1; then the ratio between the number of times each segmented block is judged as a defective block in the preset inspection times c and the preset inspection times c is marked as the historical defect coefficient corresponding to each segmented block, that is, the historical defect coefficient Wi corresponding to each segmented block is calculated by the formula Qa / c=Wi. This coefficient reflects the defect frequency of the block in the past inspection. The larger the historical defect coefficient, the higher the frequency of defects in the corresponding segmented block.

[0057] The historical defect distribution coefficient FA corresponding to the defect coefficient Wi of each segmented block is obtained through discrete calculation, and the defect distribution of the benchmark is reflected through the historical defect distribution coefficient;

[0058] The specific method of discrete calculation is:

[0059]

[0060] Among them, Wi is the historical defect coefficient, FA is the historical defect distribution coefficient, a is the number of segmented blocks, Wp is the mean of the historical defect coefficient, and a≥i≥1.

[0061] After determining the abnormal block, the working parameters of the abnormal block can be collected again, and then the working parameters of the abnormal block can be adjusted based on the historical working parameters of the abnormal block. Specifically, the absolute value of the difference between the historical defect distribution coefficient and the real-time defect distribution coefficient is analyzed, and then the distribution abnormality signal is generated based on the analysis result. Specifically, by |FA-FB|=BI, the absolute value BI of the difference between the historical defect distribution coefficient and the real-time defect distribution coefficient can be obtained. When BI is greater than Y1, a distribution abnormality signal is generated. The generation of a distribution abnormality signal indicates that the distribution position of the defect of the wafer to be detected in each segmented block is abnormal, that is, the defect distribution position on the wafer to be detected has changed abnormally. When BI is less than or equal to Y1, there is no abnormality in the task, and no processing is performed. Y1 is the preset first abnormal value, and the specific value is set by relevant personnel according to actual needs.

[0062] In another embodiment, after generating a distribution abnormality signal, the absolute value of the difference between the real-time defect coefficient and the historical defect coefficient of each segmented block is obtained. When the absolute value of the difference is large, the corresponding segmented block is marked as an abnormal block. Specifically, for each segmented block, the absolute value of the difference between the real-time defect coefficient Bi and the historical defect coefficient Wi is calculated by the formula |Wi-Bi|=Ra. When Ra is greater than the threshold value Y2, it is judged that the difference is not within the normal fluctuation range, and the segmented block is marked as an abnormal block. Otherwise, no processing is performed. The threshold value Y2 is a preset second abnormal value, and the specific value is set by relevant personnel according to actual needs.

[0063] For the blocks marked as abnormal, the operation data of the manufacturing equipment performing a predetermined number of inspections b times on each abnormal block is collected. The operation data may include at least one of the equipment temperature, pressure and speed. Of course, other data types may also be added according to the actual operation conditions. The following is an example in which the operation data includes the equipment temperature, pressure and speed.

[0064] The average value of the equipment temperature, pressure and speed when the manufacturing equipment performs a predetermined number of detections b times of manufacturing operations on each abnormal block is used as the abnormal equipment temperature, pressure and speed corresponding to each abnormal block, and at the same time, the standard data when the same manufacturing equipment performs e times of manufacturing operations on the corresponding blocks of each abnormal block under standard working conditions is obtained. The standard data includes equipment temperature, pressure and speed, and e is a preset value. The specific value is set by relevant personnel according to actual needs. Then, the average value of the equipment temperature, pressure and speed when the manufacturing equipment performs e times of manufacturing operations on the corresponding blocks of each abnormal block is used as the standard equipment temperature, pressure and speed of the corresponding blocks of each abnormal block, and then the difference between the standard equipment temperature, pressure and speed of the corresponding blocks of each abnormal block and the abnormal equipment temperature, pressure and speed corresponding to each abnormal block is calculated, as well as the difference of each parameter corresponding to the manufacturing equipment when performing the manufacturing operation on each abnormal block, and then according to the difference of each parameter corresponding to each abnormal block, the manufacturing equipment performs compensatory control adjustment on each parameter corresponding to the manufacturing operation on each abnormal block.

[0065] In one example, when the parameter difference is positive, the corresponding parameter is increased and adjusted according to the corresponding value of the parameter difference, and when the parameter difference is negative, the corresponding parameter is decreased and adjusted according to the corresponding value of the parameter difference; if the temperature parameter difference of an abnormal block is positive, the pressure parameter difference is negative, and the speed parameter is 0, then when the manufacturing equipment performs a manufacturing operation on the block, the temperature parameter of the manufacturing equipment is increased according to the temperature parameter difference value, the pressure parameter of the manufacturing equipment is decreased according to the pressure parameter difference, and the speed parameter is not processed, thereby achieving the purpose of corresponding compensatory control and adjustment of various parameters of the manufacturing equipment when the manufacturing equipment performs a manufacturing operation on the abnormal block.

[0066] The embodiment of the present invention first collects multiple segmented blocks of multiple inspections of the wafer to be inspected based on a machine vision device, then determines the real-time defect distribution coefficient corresponding to each segmented block based on the number of times each segmented block is marked as a defective block in multiple inspections, and finally determines the abnormal block according to the real-time defect distribution coefficient corresponding to each segmented block and the historical defect distribution coefficient corresponding to each segmented block. Therefore, it can solve the problem in the prior art that it is impossible to analyze the abnormality of the defect distribution position, resulting in the inability to timely discover the abnormal changes in the defect position, thereby affecting the wafer manufacturing efficiency. It can realize accurate analysis and timely discovery of the abnormality of the defect distribution position in the wafer manufacturing process, thereby optimizing the production process, and improving the wafer production efficiency while ensuring the wafer quality.

[0067] The following is a description of the wafer defect detection system based on machine vision provided by the present invention. The wafer defect detection system based on machine vision described below and the wafer defect detection method based on machine vision described above can refer to each other.

[0068] See also Figure 2 , Figure 2 A schematic diagram of the structure of a wafer defect detection system based on machine vision provided by an embodiment of the present invention, wherein the wafer defect detection system based on machine vision includes:

[0069] The acquisition module 210 is used to acquire multiple segmented blocks for multiple inspections of the wafer to be inspected based on a machine vision device;

[0070] A determination module 220, configured to determine a real-time defect distribution coefficient corresponding to each segmented image block based on the number of times each segmented image block is marked as a defective image block in multiple inspections;

[0071] The defect detection module 230 is used to determine the abnormal block according to the real-time defect distribution coefficient corresponding to each segmented block and the historical defect distribution coefficient corresponding to each segmented block.

[0072] The machine vision-based wafer defect detection system provided by the present invention first collects multiple segmented blocks for multiple inspections of the wafer to be inspected based on the machine vision equipment, then determines the real-time defect distribution coefficient corresponding to each segmented block based on the number of times each segmented block is marked as a defective block in the multiple inspections, and finally determines the abnormal block according to the real-time defect distribution coefficient corresponding to each segmented block and the historical defect distribution coefficient corresponding to each segmented block. Therefore, it can solve the problem in the prior art that it is impossible to analyze the abnormality of the defect distribution position, resulting in the inability to timely discover the abnormal change of the defect position, thereby affecting the wafer manufacturing efficiency. It can realize accurate analysis and timely discovery of the abnormality of the defect distribution position in the wafer manufacturing process, thereby optimizing the production process, and improving the wafer production efficiency while ensuring the wafer quality.

[0073] In one embodiment, the acquisition module 210 is specifically used for:

[0074] Collect multiple inspection images of the wafer to be inspected for multiple inspections;

[0075] Each detection image is segmented to obtain multiple segmentation blocks.

[0076] In one embodiment, the determination module 220 is specifically configured to:

[0077] Determine a real-time defect coefficient corresponding to each segmented image block based on the number of times each segmented image block is marked as a defective image block in multiple inspections;

[0078] The real-time defect distribution coefficient corresponding to each segmented image block is determined based on the real-time defect coefficient.

[0079] In one embodiment, the determination module 220 is specifically used to determine the real-time defect distribution coefficient corresponding to each segmented image block according to the following formula:

[0080]

[0081] Wherein, Bi is the real-time defect coefficient, FB is the real-time defect distribution coefficient, a is the number of segmented blocks, Bp is the mean of the real-time defect coefficient, and a≥i≥1.

[0082] In one embodiment, the defect detection module 230 is further configured to:

[0083] Determine a historical defect coefficient corresponding to each segmented image block based on a historical inspection image of the wafer to be inspected;

[0084] The historical defect distribution coefficient is determined using the historical defect coefficient according to the following formula:

[0085]

[0086] Among them, Wi is the historical defect coefficient, FA is the historical defect distribution coefficient, a is the number of segmented blocks, Wp is the mean of the historical defect coefficient, and a≥i≥1.

[0087] In one embodiment, the defect detection module 230 is specifically used for:

[0088] Determine the difference between the real-time defect distribution coefficient corresponding to each segmented image block and the historical defect distribution coefficient corresponding to each segmented image block;

[0089] The segmented blocks whose difference values ​​are greater than a preset abnormal value are determined as abnormal blocks.

[0090] In one embodiment, the defect detection module 230 is specifically used for:

[0091] Collect working parameters of abnormal blocks;

[0092] Based on the historical working parameters of the abnormal block, the working parameters of the abnormal block are adjusted.

[0093] See also Figure 3 , Figure 3 Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 3 As shown, an embodiment of the present invention provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor 320. When the processor 320 executes the computer program 311, the following steps are implemented:

[0094] Collect multiple segmented blocks for multiple inspections of the wafer to be inspected based on machine vision equipment;

[0095] Determine a real-time defect distribution coefficient corresponding to each segmented image block based on the number of times each segmented image block is marked as a defective image block in multiple inspections;

[0096] The abnormal blocks are determined according to the real-time defect distribution coefficient corresponding to each segmented block and the historical defect distribution coefficient corresponding to each segmented block.

[0097] See also Figure 4 , Figure 4 Schematic diagram of an embodiment of a computer-readable storage medium provided in an embodiment of the present invention. Figure 4 As shown, this embodiment provides a computer-readable storage medium 400, on which a computer program 311 is stored. When the computer program 311 is executed by a processor, the following steps are implemented:

[0098] Collect multiple segmented blocks for multiple inspections of the wafer to be inspected based on machine vision equipment;

[0099] Determine a real-time defect distribution coefficient corresponding to each segmented image block based on the number of times each segmented image block is marked as a defective image block in multiple inspections;

[0100] The abnormal blocks are determined according to the real-time defect distribution coefficient corresponding to each segmented block and the historical defect distribution coefficient corresponding to each segmented block.

[0101] On the other hand, the present invention further provides a computer program product, which includes a computer program. The computer program can be stored in a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the wafer defect detection method based on machine vision provided by the above methods, which includes:

[0102] Collect multiple segmented blocks for multiple inspections of the wafer to be inspected based on machine vision equipment;

[0103] Determine a real-time defect distribution coefficient corresponding to each segmented image block based on the number of times each segmented image block is marked as a defective image block in multiple inspections;

[0104] The abnormal blocks are determined according to the real-time defect distribution coefficient corresponding to each segmented block and the historical defect distribution coefficient corresponding to each segmented block.

[0105] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0106] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0107] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0108] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0109] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0110] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0111] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A wafer defect detection method based on machine vision, characterized in that: include: Collect multiple segmented blocks for multiple inspections of the wafer to be inspected based on machine vision equipment; Determine a real-time defect distribution coefficient corresponding to each segmented image block based on the number of times each segmented image block is marked as a defective image block in the multiple inspections; The abnormal block is determined according to the real-time defect distribution coefficient corresponding to each segmented block and the historical defect distribution coefficient corresponding to each segmented block.

2. The wafer defect detection method based on machine vision according to claim 1, characterized in that: The method of collecting multiple segmented image blocks for multiple inspections of the wafer to be inspected based on a machine vision device includes: Collecting multiple inspection images of the wafer to be inspected based on the machine vision device; Each of the detection images is segmented to obtain a plurality of segmented image blocks.

3. The wafer defect detection method based on machine vision according to claim 2, characterized in that: The determining, based on the number of times each segmented image block is marked as a defective image block in the multiple detections, a real-time defect distribution coefficient corresponding to each segmented image block includes: Determine a real-time defect coefficient corresponding to each segmented image block based on the number of times each segmented image block is marked as a defective image block in the multiple detections; A real-time defect distribution coefficient corresponding to each of the segmented blocks is determined based on the real-time defect coefficient.

4. The wafer defect detection method based on machine vision according to claim 3, characterized in that: The determining the real-time defect distribution coefficient corresponding to each segmented block based on the real-time defect coefficient includes: The real-time defect distribution coefficient corresponding to each segmented image block is determined according to the following formula: Wherein, Bi is the real-time defect coefficient, FB is the real-time defect distribution coefficient, a is the number of segmented blocks, Bp is the mean value of the real-time defect coefficient, and a≥i≥1.

5. The wafer defect detection method based on machine vision according to claim 4, characterized in that: Before determining the abnormal block according to the real-time defect distribution coefficient corresponding to each segmented block and the historical defect distribution coefficient corresponding to each segmented block, the method further includes: Determine a historical defect coefficient corresponding to each of the segmented blocks based on a historical inspection image of the wafer to be inspected; The historical defect distribution coefficient is determined using the historical defect coefficient according to the following formula: Wherein, Wi is the historical defect coefficient, FA is the historical defect distribution coefficient, a is the number of segmented blocks, Wp is the mean value of the historical defect coefficient, and a≥i≥1.

6. The method for wafer defect detection based on machine vision according to claim 5, characterized in that: The determining of the abnormal block according to the real-time defect distribution coefficient corresponding to each segmented block and the historical defect distribution coefficient corresponding to each segmented block comprises: Determine a difference between a real-time defect distribution coefficient corresponding to each segmented image block and a historical defect distribution coefficient corresponding to each segmented image block; The segmented image block whose difference value is greater than a preset abnormal value is determined as the abnormal image block.

7. The wafer defect detection method based on machine vision according to claim 1, characterized in that: Also includes: Collecting working parameters of the abnormal block; Based on the historical operating parameters of the abnormal block, the operating parameters of the abnormal block are adjusted.

8. A wafer defect detection system based on machine vision, characterized in that: include: An acquisition module, used for acquiring multiple segmented blocks for multiple inspections of the wafer to be inspected based on a machine vision device; A determination module, configured to determine a real-time defect distribution coefficient corresponding to each segmented image block based on the number of times each segmented image block is marked as a defective image block in the multiple detections; The defect detection module is used to determine the abnormal block according to the real-time defect distribution coefficient corresponding to each segmented block and the historical defect distribution coefficient corresponding to each segmented block.

9. An electronic device, characterized in that: include: Memory for storing computer software programs; A processor is used to read and execute the computer software program, thereby implementing the wafer defect detection method based on machine vision as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, characterized in that: The storage medium stores a computer software program, and when the computer software program is executed by the processor, the machine vision-based wafer defect detection method as described in any one of claims 1 to 7 is implemented.

Images