Sampling inspection methods, apparatus and applications for wafers

Abstract

This invention discloses a sampling inspection method, apparatus, and application for wafers. The sampling inspection method includes: acquiring historical defect data of the wafer, the historical defect data including defect coordinates and defect level factors corresponding to the defect coordinates; acquiring a dynamic defect weight for each coordinate on the current wafer based on the defect coordinates and the corresponding defect level factors; and dividing each coordinate on the wafer to be inspected into at least two inspection regions for sampling inspection based on the dynamic defect weights, wherein the sampling rates of the at least two inspection regions are different. This invention divides the inspection regions based on the correlation between historical wafer defect coordinates and the current wafer in terms of distance and time, and configures a corresponding sampling rate for each inspection region to balance sampling efficiency and sampling quality, avoiding over-sampling or under-sampling of some areas on the wafer.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of wafer detection, and particularly relates to a sampling detection method, device and application for a wafer. BACKGROUND

[0002] Wafer detection is a key link in semiconductor manufacturing, but full detection is time-consuming and expensive, and a sampling method is usually used to select part of the Die for detection to balance efficiency and quality, but the existing sampling detection method has problems such as poor representativeness of sampling results, excessive sampling or insufficient sampling in some areas, unstable representativeness of sampling, and chaotic sampling path.

[0003] Therefore, in view of the above technical problems, it is necessary to provide a sampling detection method, device and application for a wafer. SUMMARY

[0004] The present application aims to provide a sampling detection method, device and application for a wafer, which can construct a defect dynamic weight based on the correlation degree of historical wafer defect coordinates and the current wafer, divide detection areas and configure corresponding sampling rates for each detection area, so as to balance the sampling efficiency and sampling quality and avoid excessive sampling or insufficient sampling in some areas on the wafer.

[0005] In order to achieve the above-mentioned purpose, the technical scheme provided by an embodiment of the present application is as follows:

[0006] A sampling detection method for a wafer, comprising:

[0007] Obtaining historical defect data of a wafer, the historical defect data comprising defect coordinates and defect level factors corresponding to the defect coordinates;

[0008] Based on the defect coordinates and the defect level factors corresponding to the defect coordinates, obtaining a defect dynamic weight of each coordinate on a current wafer;

[0009] Based on the defect dynamic weight, dividing each coordinate on a wafer to be detected into at least two detection areas for sampling detection, wherein the sampling rates of the at least two detection areas are different.

[0010] In one or more embodiments of the present application, based on the defect coordinates and the defect level factors corresponding to the defect coordinates, obtaining a defect dynamic weight of each coordinate on a current wafer, comprises:

[0011] Based on the positional relationship between the current coordinate and each of the defect coordinates, obtaining a defect probability of the current coordinate;

[0012] Based on the defect level factors corresponding to the defect coordinates, obtaining a defect influence factor corresponding to the current coordinate;

[0013] Based on the defect probability of the current coordinate and the defect influence factor corresponding to the current coordinate, a defect dynamic weight of each coordinate on the current wafer is obtained.

[0014] In one or more embodiments of the present application, based on the positional relationship between the current coordinate and each of the defect coordinates, the defect probability of the current coordinate is obtained, including:

[0015] Based on the first distance between the defect coordinate and the current coordinate, a defect influence weight between the defect coordinate and the current coordinate is determined.

[0016] The defect influence weights between all defect coordinates and the current coordinate are normalized to determine the defect probability of the current coordinate.

[0017] In one or more embodiments of the present application, the defect influence factor corresponding to the current coordinate is obtained based on the defect level factor corresponding to the defect coordinate, including:

[0018] Based on the wafer piece difference corresponding to the defect coordinate and the current coordinate, a first adjustment factor of each wafer historical piece is determined, and the first adjustment factor is inversely related to the value of the wafer piece difference.

[0019] Based on the defect level corresponding to each defect coordinate and the first adjustment factor of the corresponding wafer piece, the defect influence factor corresponding to the current coordinate is determined.

[0020] In one or more embodiments of the present application, based on the defect probability of the current coordinate and the defect influence factor corresponding to the current coordinate, a defect dynamic weight of each current coordinate is obtained, including:

[0021] The defect probability of the current coordinate and the defect influence factor corresponding to the current coordinate are respectively weighted and linearly combined to obtain the defect dynamic weight of each current coordinate, wherein the coefficient corresponding to the defect probability of the current coordinate is greater than or equal to the coefficient corresponding to the defect influence factor of the current coordinate; and / or, the sum of the coefficient corresponding to the defect probability of the current coordinate and the coefficient corresponding to the defect influence factor of the current coordinate is 1.

[0022] In one or more embodiments of the present application, the sampling detection method further includes: based on the defect influence factor corresponding to the current coordinate, the current coordinate is divided into a corresponding detection area for sampling detection.

[0023] In one or more embodiments of the present application, the sampling detection includes first sampling detection, and the first sampling detection includes:

[0024] All coordinates in the current detection area are taken as sampling points; and / or,

[0025] All coordinates in the current detection area and neighboring coordinates of all coordinates are taken as sampling points.

[0026] In one or more embodiments of the present application, the sampling detection includes second sampling detection, and the second sampling detection includes:

[0027] Based on the total number of coordinates in the current detection area and the average value of the defect dynamic weight of all coordinates in the current detection area, the number of sampling points corresponding to the current detection area is calculated.

[0028] All coordinates in the current detection area are sorted in descending order of their defect dynamic weight, and the first sampling point number of coordinates are extracted as sampling points of the current detection area.

[0029] In one or more embodiments of the present application, the sampling detection includes third sampling detection, and the third sampling detection includes:

[0030] Based on the total number of coordinates in the current detection area, the number of sampling points corresponding to the current detection area is calculated.

[0031] Randomly extracting sampling point number of coordinates in the current detection area for detection; and / or,

[0032] Based on the number of sampling points corresponding to the current detection area, the current detection area is divided into a plurality of grids, the number of secondary sampling points is calculated based on the number of grids and the number of sampling points, and the secondary sampling point number of coordinates in each grid is extracted for detection.

[0033] In one or more embodiments of the present application, the defect dynamic weight includes the defect probability of the current coordinate, and the sampling detection method further includes:

[0034] The ratio between the total number of defect coordinates detected in the historical wafer and the total number of sampling points of the historical wafer is taken as a normalization factor;

[0035] Based on the normalization factor and the number of wafers before updating, the defect probability of the current coordinate is updated to obtain the defect probability of the current coordinate after updating.

[0036] In one or more embodiments of the present application, the defect dynamic weight includes the defect probability of the current coordinate, and based on the sampling detection result, the defect dynamic weight is dynamically updated, including:

[0037] ;

[0038] The defect probability of the current coordinate after updating the defect dynamic weight;

[0039] The defect probability of the current coordinate before updating the defect dynamic weight​ The probability of defects;

[0040] This represents the total number of defect coordinates detected in the L wafers preceding the current wafer in this batch.

[0041] This represents the total number of sampling points for the L wafers preceding the current wafer in this batch.

[0042] In one or more embodiments of the present invention, the sampling detection method further includes:

[0043] When a wafer defect is missed within a batch, the dependency of the dynamic defect weight at each current coordinate on all historical defect data is increased; and / or,

[0044] When the defect density growth rate per unit area of ​​wafers within a batch exceeds a first set threshold, the influence of recent historical defect data in the dynamic defect weight of each current coordinate is strengthened.

[0045] In one or more embodiments of the present invention, the sampling inspection method further includes: when the dynamic weight of defects at the same coordinate on multiple consecutive wafers within the same batch increases, the coordinate on the current wafer is directly divided into the inspection area with the highest sampling rate; and / or,

[0046] When the growth rate of the number of coordinates in the detection area with the highest sampling rate on two adjacent wafers exceeds the second set threshold, the process is stopped for inspection.

[0047] A specific embodiment of the present invention provides a sampling and inspection device for wafers, the device comprising:

[0048] The data acquisition module is used to acquire historical defect data of the wafer, the historical defect data including defect coordinates and defect level factors corresponding to the defect coordinates;

[0049] The data processing module is used to obtain the dynamic weight of the defect at each coordinate on the current wafer based on the defect coordinate and the defect level factor corresponding to the defect coordinate.

[0050] The sampling inspection module is used to divide each coordinate on the wafer to be inspected into at least two inspection regions for sampling inspection based on the defect dynamic weight, wherein the sampling rates of the at least two inspection regions are different.

[0051] A specific embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the aforementioned sampling and inspection method for wafers.

[0052] The embodiment of the present application provides a computer readable medium, wherein computer execution instructions are carried in the computer readable medium, and the computer execution instructions are used for realizing the sampling detection method for a wafer when executed by a processor.

[0053] Compared with the prior art, the sampling detection method, device and application for a wafer can obtain the defect dynamic weight of each coordinate on a current wafer based on defect coordinates and defect level factors corresponding to the defect coordinates, so that serious defects are preferentially detected, the representativeness of a sampling result is improved, and yield risk is effectively reduced.

[0054] By dividing each coordinate on a wafer to be detected into different detection regions and setting corresponding sampling rates, the representativeness of sampling detection points is improved, and over-sampling or insufficient sampling in some regions is avoided. BRIEF DESCRIPTION OF DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort on the basis of these drawings.

[0056] Figure 1 The flow chart of the sampling detection method for a wafer in an embodiment of the present application;

[0057] Figure 2 The schematic diagram of wafer coordinate arrangement in an embodiment of the present application;

[0058] Figure 3 The structural schematic diagram of the sampling detection device for a wafer in an embodiment of the present application;

[0059] Figure 4 The structural schematic diagram of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0060] In order to make the person skilled in the art better understand the technical solutions in the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described in the present disclosure with reference to the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only some of the embodiments of the present disclosure, not all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those skilled in the art without creative effort should belong to the protection scope of the present disclosure.

[0061] As described in the background, wafer detection is a key link in semiconductor manufacturing, but full detection is time-consuming and expensive, and usually a sampling method is used to select part of the Die for detection to balance efficiency and quality, but the existing sampling detection method has problems such as poor representativeness of sampling results, over-sampling or insufficient sampling in some areas, unstable representativeness of sampling, and chaotic sampling path. The common wafer sampling detection in the existing design includes simple rule sampling, grid sampling and random sampling, all of which have various problems.

[0062] Simple rule sampling (for example, radial / annular sampling): points are taken at equal intervals along the radius direction of the wafer or concentric circles. The problem of simple rule sampling is that it cannot adapt to the non-uniformly distributed defects that may exist (such as edge intensive effect, local hot spots caused by specific processes), and when the defect distribution is non-uniform, the representativeness of the sampling result is poor.

[0063] Grid sampling: a uniform grid is overlaid on the wafer surface, and sampling detection is performed at the grid points or grid centers. The problem of grid sampling is that: as radial sampling, grid sampling may be insufficient in defect aggregation areas, and over-sampling in defect-free areas, so the sampling efficiency is low. In addition, grid sampling ignores the circular boundary of the wafer, and the boundary grid points may be invalid or have poor representativeness.

[0064] Random sampling: randomly select detection points in the effective area. The problem of random sampling is that: although theoretically unbiased, it may lead to uneven distribution of sampling points (such as accidental aggregation or sparseness) in actual operation, and the representativeness of the sampling result is unstable. In addition, the path of random sampling may be chaotic, resulting in a long moving path of the sampling detection equipment and frequent changes in the motion direction, which greatly reduces the detection efficiency.

[0065] Based on the in-depth analysis of the above technical problems, the present application provides a sampling detection method for a wafer, which combines a long-term historical defect rule to construct a defect dynamic weight, and the sampling detection method specifically comprises:

[0066] Obtaining historical defect data of the wafer, the historical defect data comprising defect coordinates and defect level factors corresponding to the defect coordinates;

[0067] Based on the defect coordinates and the defect level factors corresponding to the defect coordinates, obtaining a defect dynamic weight of each coordinate on the current wafer;

[0068] Based on the defect dynamic weight, dividing each coordinate on the wafer to be detected into at least two detection areas for sampling detection, wherein the sampling rates of the at least two detection areas are different.

[0069] It can be understood that the existing sampling detection method does not consider the defect priority, and different defect types or defect degrees have great difference in yield influence, for example, the influence degree of fatal short circuit and slight scratch on Die is different. In addition, random sampling or simple regular sampling ignores the historical defect law of Die, which leads to the possibility of missing detection in high-risk areas and over-detection in low-risk areas. The present application considers the influence of defect severity on wafer detection, dynamically adjusts the defect dynamic weight according to the defect severity, and ensures that serious defects can be detected first.

[0070] Further, in an embodiment, the defect dynamic weight of each coordinate on the current wafer is obtained based on the defect coordinate and the defect level factor corresponding to the defect coordinate, comprising:

[0071] obtaining the defect probability of the current coordinate based on the positional relationship between the current coordinate and each defect coordinate;

[0072] obtaining the defect impact factor corresponding to the current coordinate based on the defect level factor corresponding to the defect coordinate;

[0073] obtaining the defect dynamic weight of each coordinate on the current wafer based on the defect probability of the current coordinate and the defect impact factor corresponding to the current coordinate.

[0074] Further, in an embodiment, the defect probability of the current coordinate is obtained based on the positional relationship between the current coordinate and each defect coordinate, comprising:

[0075] determining the defect impact weight between the defect coordinate and the current coordinate based on the first distance between the defect coordinate and the current coordinate;

[0076] normalizing the defect impact weight between all defect coordinates and the current coordinate to determine the defect probability of the current coordinate.

[0077] Further, in an embodiment, the defect impact factor corresponding to the current coordinate is obtained based on the defect level factor corresponding to the defect coordinate, comprising:

[0078] determining the first adjustment factor of each wafer historical wafer based on the wafer difference of the wafer corresponding to the defect coordinate and the current coordinate, the first adjustment factor being inversely related to the value of the wafer difference;

[0079] determining the defect impact factor corresponding to the current coordinate based on the defect level corresponding to each defect coordinate and the first adjustment factor of the corresponding wafer.

[0080] The defect probability at the current coordinate represents the degree of correlation between the historical wafer defect coordinates and the current wafer in terms of distance, while the defect influence factor corresponding to the current coordinate represents the degree of correlation between the historical wafer defect coordinates and the current wafer in terms of time. This invention constructs a dynamic defect weight by comprehensively considering the defect probability at the current coordinate and the defect influence factor corresponding to the current coordinate, and divides the detection area based on the dynamic defect weight, and configures a corresponding sampling rate for each detection area to balance sampling efficiency and sampling quality, and avoid oversampling or undersampling in some areas of the wafer.

[0081] Furthermore, in one embodiment, the sampling inspection method also includes: dynamically updating the dynamic weight of defects based on the sampling inspection results.

[0082] Understandably, the sampling strategies in existing technologies are static and fixed, and have no learning ability within a batch. This invention optimizes the subsequent sampling strategy in real time based on the sampling and testing results of the first few wafers in the same batch, further reducing yield risk.

[0083] The present invention will be further described below with reference to specific embodiments.

[0084] Example 1:

[0085] like Figure 1 As shown, this embodiment provides a sampling inspection method for wafers, including:

[0086] S101, Obtain historical defect data of the wafer. The historical defect data includes defect coordinates and the defect level factor corresponding to the defect coordinates.

[0087] S102, based on the defect coordinates and the defect level factors corresponding to the defect coordinates, obtain the dynamic weight of the defect at each coordinate on the current wafer;

[0088] S103, based on the dynamic weight of defects, divide each coordinate on the wafer to be inspected into at least two inspection regions for sampling inspection, wherein the sampling rates of at least two inspection regions are different.

[0089] For example, the historical defect data in step S101 also includes the defect type corresponding to the defect coordinates. And the wafer number where the defect coordinates are located.

[0090] like Figure 2 As shown, a wafer includes several periodically arranged dies. Sampling inspection refers to selecting a number of dies from all the dies on a wafer for inspection. In this embodiment, coordinates are used to facilitate the differentiation of dies. Indicates that it is located at The position of the die. It should be noted that the relative position relationship of the same coordinates on each wafer in the same batch is unchanged, for example, for the kth wafer and the k+1th wafer in the same batch, because the wafer size is the same and the die size on the wafer is also the same, the coordinates The relative position on the wafer is also the same.

[0091] For example, according to the severity of the defect, the defect level corresponding to the defect coordinates is divided into three levels, and the defect level factor corresponding to each level is set to 1.0, 0.7 and 0.2 in turn, that is, the more serious the defect, the higher the corresponding defect level factor coefficient. In other alternative embodiments, the defect level is divided into five levels, and the defect level factor corresponding to each level is set to 1.1, 1.0, 0.7, 0.5 and 0.2 in turn. Those skilled in the art can set the defect level and the defect level factor according to the wafer state and experience, which is not limited in the present application.

[0092] Further, step S102 specifically includes:

[0093] S1021, obtaining a defect probability of the current coordinates based on the positional relationship between the current coordinates and each defect coordinates;

[0094] S1022, obtaining a defect influence factor corresponding to the current coordinates based on the defect level factor corresponding to the defect coordinates;

[0095] S1023, obtaining a defect dynamic weight of each coordinates on the current wafer based on the defect probability of the current coordinates and the defect influence factor corresponding to the current coordinates.

[0096] In the present embodiment, the defect probability of the current coordinates represents the correlation degree between the historical wafer defect coordinates and the current wafer in distance, that is, the closer the current coordinates are to the historical wafer defect coordinates in space, the greater the probability of defects in the current coordinates. The defect influence factor corresponding to the current coordinates represents the correlation degree between the historical wafer defect coordinates and the current wafer in time, that is, the closer the current coordinates are to the historical wafer defect coordinates in time, the deeper the degree of influence of defects on the current coordinates, and the defect dynamic weight combines the defect probability of the current coordinates and the defect influence factor corresponding to the current coordinates.

[0097] Further, step S1021 specifically includes:

[0098] determining a defect influence weight between the defect coordinates and the current coordinates based on the first distance between the defect coordinates and the current coordinates;

[0099] normalizing the defect influence weight between all defect coordinates and the current coordinates to determine the defect probability of the current coordinates.

[0100] In this embodiment, step S1021 includes:

[0101] For each coordinate on the current wafer, calculate the defect probability density at that coordinate: using the current coordinate Centered on the coordinates of each defect in the historical defect data To current coordinates The distance between the points in the defect influences a defect weight; understandably, the closer the distance, the greater the contribution of the defect influence weight. In this embodiment, the weights are determined by a kernel function. The decision is made, i.e., the weight value is the kernel function value.

[0102] Normalization: Sum the defect impact weights of all historical defect coordinates and divide by the normalization factor to obtain the defect probability of that coordinate.

[0103] For example, for each coordinate on the k-th wafer in the current batch Calculate the defect probability at that coordinate (i.e., the probability that a die located at that coordinate will have a defect):

[0104] (1);

[0105] In equation (1), Current coordinates The probability of defects, For kernel function, This represents the total number of defect coordinates in the historical defect data.

[0106] It is understandable that historical defect data may involve defect data from multiple wafers, and the total number of defect coordinates in the historical defect data is [missing information]. The same defect coordinates are not counted repeatedly. For example, the historical defect data includes defect data from three wafers, where the defect coordinates on the first wafer include... The defect coordinates on the second wafer include , and The defect coordinates on the third wafer include and The defect coordinates can be determined. If an item appears repeatedly, it will not be counted as a duplicate. Therefore, in this example... The value is 5.

[0107] In the above formula, The smoothness of the kernel function K can be determined by adjusting the value of h based on the distribution of historical defect data. For example, Scott's rule or Silverman's rule can be used to determine the value of h, or cross-validation can be used to set h.

[0108] In this embodiment, the Silverman rule is used to set the parameters: where d is the dimensionality (in this embodiment, d = 2), and d can take the value 1 or other values in alternative embodiments, and σ is the standard deviation of the data.

[0109] In other alternative embodiments, where, σ is the average of the standard deviations of the data.

[0110] It will be appreciated that any known or unknown kernel function can be used without limitation, for example, the kernel function The Gaussian kernel function or Epanechnikov kernel can be used.

[0111] It should be noted that this embodiment considers the two-dimensional case, and the normalization factor is set to In other alternative embodiments, the normalization factor can be set to i.e.,

[0112] .

[0113] Exemplarily,

[0114] The defect coordinates in the historical defect data include , and , the total number of defect coordinates in the historical defect data N is 3, and the current coordinates , ;

[0115] The Gaussian kernel function is ;

[0116] The Euclidean distance between each defect coordinate and the current coordinates is calculated respectively:

[0117] The Euclidean distance from to ;

[0118] The Euclidean distance from to ;

[0119] The Euclidean distance from to .

[0120] The Gaussian kernel function value corresponding to the Euclidean distance between each historical defect coordinate and the current coordinates is calculated:

[0121] ;

[0122] ;

[0123] .

[0124] Sum the weights of all historical defect coordinates:

[0125] Normalization: , the defect probability at the current wafer coordinate is about . In turn, traverse each coordinate on the wafer and obtain the defect probability of each coordinate on the current wafer.

[0126] In other alternative embodiments, step S1021 comprises:

[0127] Pre-compute a probability density map (heat map) to optimize the calculation process, including: divide the wafer into several grids (for example, 200x200), calculate for the center point of each grid, use fast Fourier transform (FFT) to speed up convolution operation, or use KD tree for nearest neighbor search to optimize calculation efficiency.

[0128] Step S1022 in this embodiment specifically comprises:

[0129] Determine the first adjustment factor of each wafer historical wafer based on the wafer difference corresponding to the defect coordinates and the current coordinates; the first adjustment factor is inversely related to the value of the wafer difference;

[0130] Determine the defect influence factor corresponding to the current coordinate based on the defect level corresponding to each defect coordinate and the first adjustment factor of the corresponding wafer.

[0131] For example, step S1022 determines the defect influence factor corresponding to the current coordinate by formula (2):

[0132] (2);

[0133] In formula (2), k is the wafer wafer, ;

[0134] is the defect influence factor corresponding to the coordinate on the kth wafer in the current batch of wafers (here, the kth wafer is the current wafer);

[0135] is the defect coordinate on the wafer before the kth wafer in the current batch of wafers The corresponding defect level factor, the defect level factor in the embodiment including 1.0, 0.7 or 0.2.

[0136] In formula (2) is the first adjustment factor, it can be understood that when is a positive number less than 1, the historical wafer is closer in time to the current wafer, and the defect coordinates on the historical wafer have a deeper impact on the current coordinates.

[0137] Exemplarily, k is 3, that is, the current wafer is the kth wafer, and the previous two wafers have obtained clear defect coordinates and corresponding defect level factors through sampling detection, for example , , , That is, among the current batch of wafers, the defect coordinates on the k-1th wafer include and , wherein the defect level factor of is 1.0, the defect level factor of is 0.7, and the defect coordinates on the k-2th wafer include and , the defect level factor of is 0.2, the defect level factor of is 0.2. Then in the example

[0138] = , .

[0139] Exemplarily, It should be noted that those skilled in the art can adjust the value range of as needed, for example, in other alternative embodiments, or .

[0140] The defect impact factor in the embodiment characterizes the impact of the defect coordinates on the adjacent wafers in the current batch of wafers on the defect distribution of the current wafer, and the wafer that is closer in time has a higher impact on the defect of the current wafer.

[0141] It can be understood that some defects on the wafer are "killer defects", such as defects that cause short circuit or open circuit of the circuit, which directly cause the chip to fail. Some defects are "benign defects", such as small particles in non-critical areas, which may not affect the function of the chip, and the traditional uniform sampling cannot distinguish the two. The embodiment sets the level of the defect degree of the defect coordinates, and represents the severity of the defect level through the defect level factor, so as to ensure that serious defects can be preferentially detected, and the risk of yield loss of control is reduced as a whole.

[0142] In step S1023 of the embodiment, the defect probability of the current coordinate and the defect influence factor corresponding to the current coordinate are respectively weighted and linearly combined to obtain the defect dynamic weight of each current coordinate.

[0143] Exemplarily, (3);

[0144] In formula (3), k is a wafer piece (in the embodiment, it is the current wafer piece to be measured), is the defect dynamic weight of the current coordinate on the kth wafer;

[0145] is the defect probability of the current coordinate ;

[0146] is the defect influence factor corresponding to the current coordinate ;

[0147] The coefficient corresponding to the defect probability of the current coordinate is greater than or equal to the coefficient corresponding to the defect influence factor corresponding to the current coordinate. The sum of the coefficient corresponding to the defect probability of the current coordinate and the coefficient corresponding to the defect influence factor corresponding to the current coordinate is 1.

[0148] Exemplarily, , and may be adaptively valued within the above range. In other alternative embodiments, , and may be adaptively valued within the above range.

[0149] It represents the degree of spatial correlation between the current coordinates and the historical defect coordinates. The closer the current coordinates are to the defect coordinates, the higher the risk of a defect occurring at that location. Conversely, the farther the current coordinates are from the defect coordinates, the lower the risk of a defect occurring at that location.

[0150] It characterizes the temporal correlation between the current coordinates and historical defect coordinates within the current batch of wafers. The closer the time interval between the current coordinates and the defect coordinates on the previous few wafers in the batch, the higher the risk of a defect occurring at that location. Conversely, the relatively far time interval between the current coordinates and the defect coordinates on the previous few wafers in the batch indicates a relatively low risk of a defect occurring at that location.

[0151] Defect dynamic weight in this embodiment Taking into account and And through coefficient and adjust and This increased proportion further improved the accuracy of defect detection.

[0152] For example, in this embodiment, step S103 specifically includes:

[0153] When the dynamic weight of the defect corresponding to the current coordinate is greater than or equal to the third set threshold, the detection area to which the current coordinate belongs is the first area; or when the defect influence factor corresponding to the current coordinate is greater than or equal to the fourth set threshold, the detection area to which the current coordinate belongs is the first area.

[0154] When the dynamic weight of the defect corresponding to the current coordinate is greater than or equal to the fifth set threshold and less than the third set threshold, the detection area to which the current coordinate belongs is the second area;

[0155] When the dynamic weight of the defect corresponding to the current coordinate is less than the third set threshold, the detection area to which the current coordinate belongs is the third area.

[0156] It should be noted that those skilled in the art can adaptively adjust the number of detection areas and the values ​​of the third, fourth, or fifth set thresholds based on factors such as wafer size, die size, or historical defect data.

[0157] For example, in this embodiment, the third threshold is set to 0.8, the fourth threshold is set to 0.9, and the fifth threshold is set to 0.6. .

[0158] Based on the above dynamic weighting of defects Analysis shows that the dynamic weight of the defect predicted by the current coordinates The higher the value, the greater the probability of a defect at the current coordinate of the die. Therefore, based on the dynamic weight of defects... Dividing all coordinates on the current wafer into regions carries a lower risk (i.e., dynamic defect weighting). Coordinates with lower risk (i.e., higher risk, i.e., defect dynamic weights) are assigned to the third region, while those with higher risk (i.e., higher risk, i ... Coordinates with higher or higher defect impact factors are assigned to the first region.

[0159] Furthermore, this embodiment sets different sampling rates according to the degree of risk. For example, when the detection area to which the current coordinate belongs is the first area, the first sampling detection is performed on the first area, and the sampling rate of the first sampling detection is greater than or equal to 100%. When the detection area to which the current coordinate belongs is the second area, the second sampling detection is performed on the second area. When the detection area to which the current coordinate belongs is the third area, the third sampling detection is performed on the third area, wherein the sampling rate of the third sampling detection is less than or equal to the sampling rate of the second sampling detection.

[0160] For example, the first sampling detection includes: using all coordinates within the current detection area as sampling points.

[0161] In other alternative embodiments, the first sampling detection includes: using all coordinates and adjacent coordinates of all coordinates within the current detection area as sampling points. Specifically, taking a coordinate within the current detection area as the center, and using that coordinate and its adjacent coordinates as sampling points, traversing each coordinate within the current detection area to determine all sampling points.

[0162] by Figure 2 The coordinates shown For example, the sampling points include an array: and Similarly, sampling is performed on all other coordinates and their adjacent coordinates within the first region. It is understood that for high-risk areas, the defect in a coordinate may be fatal, meaning the defect could affect the area surrounding that coordinate. This embodiment avoids missing defect coordinates by "oversampling" high-risk detection areas.

[0163] Furthermore, in this embodiment, the second sampling detection includes:

[0164] The number of sampling points corresponding to the current detection area is calculated based on the total number of coordinates in the current detection area and the average value of the dynamic defect weights of all coordinates in the current detection area.

[0165] All coordinates within the current detection area are sorted in descending order of their defect dynamic weights, and several coordinates from the previous sampling points are extracted as sampling points for the current detection area.

[0166] In the embodiment, the number of sampling points in the current detection area is calculated based on the following formula :

[0167] wherein k is the wafer number, is the average of the defect dynamic weights of all coordinates in the current detection area of the kth wafer, is the basic sampling rate, is the total number of coordinates in the current detection area, and the basic sampling rate is 0.1, 0.2 or 0.3;

[0168] All coordinates in the current detection area are sorted in descending order of their defect dynamic weights, and the first coordinates are extracted as the sampling points for the second sampling detection.

[0169] For example, the total number of coordinates in the current detection area is 100 (for example, the total number of Dies with defect dynamic weights between 0.6 and 0.8 is 100), and the average of the defect dynamic weights of all coordinates in the current detection area is 0.7, then the number of sampling points is 14, and the first 14 coordinates in the sequence are extracted as the sampling points for the second sampling detection.

[0170] In other alternative embodiments, the number of sampling points in the current detection area is calculated based on the following formula :

[0171] .

[0172] For example, the third sampling detection includes:

[0173] The number of sampling points corresponding to the current detection area is calculated based on the total number of coordinates in the current detection area, and a number of coordinates in the current detection area are randomly extracted for detection.

[0174] For example, the number of sampling points in the current detection area is calculated based on the following formula :

[0175] wherein is the total number of coordinates in the current detection area.

[0176] The values of and may be adjusted based on the wafer size, For example, for a wafer with a size of 300 mm, For example, for a wafer with a size of 150 mm, .

[0177] In other alternative embodiments, the third sampling detection comprises:

[0178] calculating the number of sampling points corresponding to the current detection area based on the total number of coordinates within the current detection area, dividing the current detection area into a plurality of grids based on the number of sampling points corresponding to the current detection area, calculating the number of secondary sampling points based on the number of grids and the number of sampling points, and extracting the number of secondary sampling points of coordinates within each grid for detection.

[0179] Further, at least one coordinate within each grid is extracted for detection.

[0180] For example, the current detection area is divided into grids, , , the number of secondary sampling points in each grid is set to , and coordinates within each grid are extracted for detection.

[0181] Further, the sampling detection method for wafers in the embodiment further comprises:

[0182] Based on the above sampling detection method, after sampling detection of the current wafer, new defect data is recorded and updated to the historical defect data, including defect coordinates and defect level factors corresponding to the defect coordinates.

[0183] The present embodiment obtains the defect dynamic weight of each coordinate on the current wafer based on the defect coordinates and the defect level factors corresponding to the defect coordinates, to ensure that serious defects are detected first, improve the representativeness of the sampling result, and effectively reduce the yield risk. By dividing each coordinate on the wafer to be detected into different detection areas and setting the corresponding sampling rate, the representativeness of the sampling detection points is improved, and over-sampling or insufficient sampling in some areas is avoided.

[0184] Embodiment 2:

[0185] It can be understood that the sampling strategy in the prior art is static and fixed, and has no learning ability within a batch. Based on the in-depth analysis of the above technical problems, the present embodiment proposes to dynamically update the defect dynamic weight based on the sampling detection result, to further reduce the yield risk by optimizing the subsequent sampling strategy in real time according to the sampling detection result of the previous wafers in the same batch.

[0186] The sampling detection method for wafers in the embodiment specifically comprises

[0187] S101, obtaining historical defect data of a wafer, the historical defect data including defect coordinates and defect level factors corresponding to the defect coordinates; ​

[0188] S102, obtaining a defect dynamic weight of each coordinate on the current wafer based on the defect coordinate and a defect level factor corresponding to the defect coordinate;

[0189] S103, dividing each coordinate on the wafer to be detected into at least two detection regions for sampling detection based on the defect dynamic weight, wherein the sampling rates of the at least two detection regions are different;

[0190] S104, dynamically updating the defect dynamic weight based on the sampling detection result, including updating a defect probability of the current coordinate in the defect dynamic weight, a defect influence factor corresponding to the current coordinate, a coefficient corresponding to the defect probability of the current coordinate, or a coefficient corresponding to the defect influence factor corresponding to the current coordinate.

[0191] It can be understood that steps S101-S103 in the embodiment can refer to embodiment 1, and step S104 in the embodiment includes: dynamically updating the defect dynamic weight based on the sampling detection result after detecting L wafers, wherein L is 3, and in other alternative embodiments, L can be adaptively valued, for example, L is 4 or 5.

[0192] In an exemplary embodiment, step S104 includes: updating the defect probability of the current coordinate in the defect dynamic weight based on the sampling detection result, specifically:

[0193] Taking the ratio between the total number of defect coordinates detected in the historical wafers and the total number of sampling points in the historical wafers as a normalization factor, and the historical wafers are the previous L wafers of the current wafer;

[0194] Updating the defect probability of the current coordinate based on the normalization factor and the number of wafers of the historical wafers to obtain the defect probability of the current coordinate after updating.

[0195] Exemplarily, ;

[0196] the defect probability of the current coordinate after updating the defect dynamic weight; the defect probability of the current coordinate before updating the defect dynamic weight;

[0197] the defect probability of the current coordinate before updating the defect dynamic weight; the defect probability of the current coordinate before updating the defect dynamic weight;

[0198] the total number of defect coordinates detected in the previous L wafers of the current wafer in the batch of wafers;

[0199] the total number of sampling points in the previous L wafers of the current wafer in the batch of wafers, that is, the sum of the sampling points in all detection regions. For example, in the embodiment, the defect dynamic weight is dynamically updated based on the sampling detection result after detecting three wafers.

[0200] For example, L is 3, i.e. the defect probability of the current coordinate on the current wafer is updated after detecting three wafer pieces. When the current wafer piece is k, the number of detected defect coordinates and the number of sampling points on the k-1th, k-2th and k-3th wafer are counted. The number of sampling points of the k-1th wafer is 20, and the number of actually measured defect coordinates is 18. The number of sampling points of the k-2th wafer is 10, and the number of actually measured defect coordinates is 7. The number of sampling points of the k-1th wafer is 18, and the number of actually measured defect coordinates is 18. Then

[0201] The updated defect dynamic weight of the current coordinate is

[0202] In another exemplary embodiment, step S104 comprises: when a wafer defect in a batch is missed, the dependence of the defect dynamic weight of each current coordinate on all historical defect data is enhanced by updating the coefficient corresponding to the defect probability of the current coordinate or the coefficient corresponding to the defect influence factor corresponding to the current coordinate.

[0203] It can be understood that if at least one Die in the Die of the same batch wafer is found to have a defect and the defect level is the highest level (for example, the defect level factor is 1.0) except the Die on the coordinates detected by sampling in the detection process or actual use process, the dependence of the defect dynamic weight of each current coordinate on all historical defect data can be enhanced by adjusting the coefficient.

[0204] For example, the is modified to , the is modified to . It should be noted that at this time, the modified and still satisfy the initial defined value range, for example, the initial limit The modified and still satisfy the above range. For example, in the sampling detection of the previous wafer, the value is 0.57, the value is 0.85, then for the current wafer, the value is modified to 0.62, the value is 0.83.

[0205] ​​​In another example embodiment, the step S104 specifically comprises: when the growth rate of the defect density per unit area of the wafer in the batch is greater than a first set threshold, the influence of the recent historical defect data on the dynamic weight of the defect of each current coordinate is strengthened by updating the defect influence factor corresponding to the current coordinate.

[0206] For example, the first set threshold is 30%, the defect density per unit area on the last two wafers is obtained based on the sampling detection result, the defect density growth rate is calculated based on the defect density per unit area on the last two wafers, and when the defect density growth rate is greater than or equal to 30%, the defect influence factor corresponding to the current coordinate is updated. is corrected to .

[0207] Further, the corrected still satisfies the initial defined value range, for example, the initial defined , the corrected still satisfies .

[0208] Further, the sampling detection method for the wafer in the embodiment further comprises:

[0209] When the dynamic weight of the defect of the same coordinate on the continuous multiple wafers in the same batch increases, the coordinate on the current wafer is directly divided into the detection area with the highest sampling rate.

[0210] For example, when , the coordinate on the current wafer is directly divided into the first area, is the dynamic weight of the defect of the current coordinate on the kth wafer, is the dynamic weight of the defect of the current coordinate on the k-3th wafer.

[0211] Further, the sampling detection method for the wafer in the embodiment further comprises:

[0212] When the growth rate of the number of coordinates in the detection area with the highest sampling rate on the adjacent two wafers is greater than a second set threshold, the process is stopped for inspection.

[0213] For example, the second set threshold in the embodiment is , and when , the process is stopped for inspection. is the number of coordinates in the detection area with the highest sampling rate on the kth wafer, is the number of coordinates in the detection area with the highest sampling rate on the k-1th wafer.

[0214] The embodiment realizes real-time acquisition and analysis of detection data, dynamically updates the dynamic weight of the defect based on the sampling detection result, optimizes the subsequent sampling strategy in real time, and further improves the representativeness of the sampling result.

[0215] As Figure 3 shown, the embodiments of the present disclosure further provide a sampling detection device for a wafer, the device comprising:

[0216] a data acquisition module 31 configured to acquire historical defect data of the wafer, the historical defect data comprising defect coordinates and defect level factors corresponding to the defect coordinates;

[0217] a data processing module 32 configured to acquire defect dynamic weights of each coordinate on a current wafer based on the defect coordinates and the defect level factors corresponding to the defect coordinates;

[0218] a sampling detection module 33 configured to divide each coordinate on a wafer to be detected into at least two detection regions for sampling detection based on the defect dynamic weights, wherein sampling rates of the at least two detection regions are different.

[0219] The embodiments of the present disclosure further provide an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the program to implement the sampling detection method for a wafer described in various embodiments of the present disclosure.

[0220] Figure 4 A hardware structure diagram of an electronic device according to an embodiment of the present disclosure is shown. As Figure 4 shown, the electronic device 40 can include at least one processor 41, a memory 42 (e.g., a non-volatile memory), an internal memory 43, and a communication interface 44, and the at least one processor 41, the memory 42, the internal memory 43, and the communication interface 44 are connected together via a bus 45. The at least one processor 41 executes at least one computer-readable instruction stored or encoded in the memory 42.

[0221] It should be understood that the computer-executable instructions stored in the memory 42, when executed, cause the at least one processor 41 to perform various operations and functions described above in conjunction with Figure 1 the various embodiments of the present disclosure.

[0222] In the embodiments of the present disclosure, the electronic device 40 can include, but is not limited to, a personal computer, a server computer, a workstation, a desktop computer, a laptop computer, a notebook computer, a mobile electronic device, a smart phone, a tablet computer, a cellular phone, a personal digital assistant (PDA), a handheld device, a messaging device, a wearable electronic device, a consumer electronic device, and the like.

[0223] The embodiment of the present disclosure further provides a computer readable medium, and computer execution instructions are carried on the computer readable medium. When the computer execution instructions are executed by a processor, various operations and functions of the sampling detection method for a wafer described in various embodiments of the present disclosure can be implemented.

[0224] The computer readable medium in the present disclosure can be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium may, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or apparatus.

[0225] In the present disclosure, the computer readable signal medium can include a data signal propagating in a baseband or as a carrier wave in a propagated data signal, which carries the computer readable program code. Such a propagated data signal can take various forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium can also be any computer readable medium other than the computer readable storage medium that can send, propagate or transmit the program for use by or in connection with an instruction execution system, device or apparatus. The program code contained on the computer readable medium can be transmitted using any suitable medium, including but not limited to wireless, wireline, optical fiber, RF, etc., or any suitable combination of the above.

[0226] Those skilled in the art will appreciate that embodiments of the present disclosure can be provided as a method, a system, or a computer program product. Therefore, the present disclosure can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present disclosure can take the form of a computer program product implemented on one or more computer usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0227] The computer program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other processing device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other processing device to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions specified in the flowchart block or blocks. Figure 1 The flowchart and / or block diagram in the drawings illustrate the architecture, functionality, and operation of possible implementations of apparatuses and computer program products according to various embodiments of the present disclosure. Figure 1 In this manner, computer(s) can be configured to implement a process for operating on or with one or more devices or apparatuses to perform the functions described in the flowchart and / or block diagram block(s).

[0228] It will be apparent to those skilled in the art that the present disclosure is not limited to the above-mentioned specific embodiments and can be implemented in other concrete forms without departing from the spirit or essential characteristics of the present disclosure. Therefore, the embodiments should be considered in a descriptive sense only and not for purposes of limitation. The scope of the disclosure should be defined by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalents of the claims are intended to be embraced therein. No claim element is to be construed as a disclaimer, unless the claim element is expressly recited as being "essential" to at least one claim.

[0229] Furthermore, it should be understood that, although the present disclosure is described in relation to the embodiments, not every implementation includes every embodiment, and the inclusion of an embodiment in a particular implementation will not cause the implementation to necessarily include all other embodiments. Rather, each embodiment will be considered in relation to the other embodiments, and the inclusion of an embodiment in the description of the disclosure should be understood as an additional embodiment that can be included in the implementation of the disclosure.

Claims

1. A sampling inspection method for wafers, characterized by, The method comprises the following steps: obtaining historical defect data of a wafer, wherein the historical defect data comprises defect coordinates and defect level factors corresponding to the defect coordinates; obtaining a defect dynamic weight of each coordinate on a current wafer based on the defect coordinates and the defect level factors corresponding to the defect coordinates, specifically comprising: obtaining a defect probability of a current coordinate based on a positional relationship between the current coordinate and each of the defect coordinates, obtaining a defect influence factor corresponding to the current coordinate based on the defect level factors corresponding to the defect coordinates, and obtaining the defect dynamic weight of each coordinate on the current wafer based on the defect probability of the current coordinate and the defect influence factor corresponding to the current coordinate; dividing each coordinate on a wafer to be detected into at least two detection regions for sampling detection based on the defect dynamic weight, wherein the sampling rates of the at least two detection regions are different; wherein the defect influence factor corresponding to the current coordinate is obtained based on the defect level factors corresponding to the defect coordinates, specifically comprising: determining a first adjustment factor of each wafer historical wafer based on wafer differences of wafer slices corresponding to the defect coordinates and the current coordinate, wherein the first adjustment factor is inversely related to the value of the wafer difference, and determining the defect influence factor corresponding to the current coordinate based on the defect level corresponding to each of the defect coordinates and the first adjustment factor of the corresponding wafer slice.

2. The method for sampling inspection of a wafer according to claim 1, wherein The defect probability of the current coordinate is obtained based on the positional relationship between the current coordinate and each of the defect coordinates, comprising: determining a defect influence weight between the defect coordinates and the current coordinate based on a first distance between the defect coordinates and the current coordinate; normalizing the defect influence weights between all defect coordinates and the current coordinate to determine the defect probability of the current coordinate.

3. The method for sampling inspection of a wafer according to claim 1, wherein The defect dynamic weight of each current coordinate is obtained based on the defect probability of the current coordinate and the defect influence factor corresponding to the current coordinate, comprising: linearly combining the defect probability of the current coordinate and the defect influence factor corresponding to the current coordinate respectively to obtain the defect dynamic weight of each current coordinate, wherein the coefficient corresponding to the defect probability of the current coordinate is greater than or equal to the coefficient corresponding to the defect influence factor of the current coordinate; and / or, the sum of the coefficient corresponding to the defect probability of the current coordinate and the coefficient corresponding to the defect influence factor of the current coordinate is 1.

4. The method for sampling inspection of a wafer according to claim 1, wherein The sampling detection method further comprises: dividing the current coordinate into a corresponding detection region for sampling detection based on the defect influence factor corresponding to the current coordinate.

5. The method for sampling inspection of a wafer according to claim 1, wherein The sampling detection comprises first sampling detection, and the first sampling detection comprises: taking all coordinates in the current detection region as sampling points; and / or, taking all coordinates in the current detection region and neighboring coordinates of all coordinates as sampling points.

6. The method for sampling inspection of a wafer according to claim 1, wherein The sampling detection comprises second sampling detection, and the second sampling detection comprises: calculating a number of sampling points corresponding to the current detection region based on the total number of coordinates in the current detection region and the average value of the defect dynamic weights of all coordinates in the current detection region; sorting all coordinates in the current detection region in order from high to low according to their defect dynamic weights, and extracting the first number of coordinates as sampling points of the current detection region.

7. The method for sampling inspection of a wafer according to claim 1, wherein The sampling detection includes a third sampling detection, and the third sampling detection includes: calculating a number of sampling points corresponding to the current detection area based on a total number of coordinates in the current detection area; randomly extracting a number of coordinates in the current detection area for detection; and / or dividing the current detection area into a plurality of grids based on the number of sampling points corresponding to the current detection area, calculating a number of secondary sampling points based on the number of grids and the number of sampling points, and extracting a number of coordinates in each grid for detection.

8. The method for sampling inspection of a wafer according to claim 1, wherein The defect dynamic weight includes a defect probability of a current coordinate, and the sampling detection method further includes: taking a ratio between a total number of defect coordinates detected in a historical wafer and a total number of sampling points of the historical wafer as a normalization factor; updating the defect probability of the current coordinate based on the normalization factor and a number of wafers before updating of the historical wafer, to obtain an updated defect probability of the current coordinate.

9. The method for sampling inspection of a wafer according to claim 8, wherein The defect dynamic weight includes a defect probability of a current coordinate, and the sampling detection method further includes: ； the current coordinates after dynamic weight update for the defect probability of the defect; the current coordinates before the dynamic weight update for the defect probability of the defect; Total number of defect coordinates detected in the current wafer among the wafer batch; Total number of sampling points in the current wafer for the batch of wafers.

10. The method for sampling inspection of a wafer according to Claim 1, wherein The sampling detection method further includes: when a defect in a wafer in a batch is missed, enhancing a dependence of a defect dynamic weight of each current coordinate on all historical defect data; and / or when a growth rate of a defect density per unit area in the wafer in the batch is greater than a first set threshold, strengthening an influence of recent historical defect data on the defect dynamic weight of each current coordinate.

11. The method for sampling inspection of a wafer according to Claim 1, wherein The sampling detection method further includes: when a defect dynamic weight of a same coordinate on a plurality of wafers in a same batch increases, directly dividing the coordinate on a current wafer into a detection area with a highest sampling rate; and / or when a growth rate of a number of coordinates in a detection area with a highest sampling rate on two adjacent wafers is greater than a second set threshold, stopping a process.

12. A sampling inspection apparatus for a wafer, for implementing the sampling inspection method according to any one of claims 1 to 11, characterized by, The device includes: a data acquisition module configured to acquire historical defect data of a wafer, the historical defect data including defect coordinates and defect level factors corresponding to the defect coordinates; a data processing module configured to acquire a defect dynamic weight of each coordinate on a current wafer based on the defect coordinates and the defect level factors corresponding to the defect coordinates; a sampling detection module configured to divide each coordinate on a wafer to be detected into at least two detection areas based on the defect dynamic weight, to perform sampling detection, wherein sampling rates of the at least two detection areas are different.

13. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor implements the sampling detection method for a wafer according to any one of claims 1-11 when executing the program.

14. A computer readable medium characterized by The computer readable medium carries computer execution instructions, and the computer execution instructions are used to implement the sampling detection method for a wafer according to any one of claims 1-11 when executed by the processor.

Images