Semiconductor detection equipment calibration system

Abstract

The invention relates to the technical field of equipment calibration, in particular to a semiconductor detection equipment calibration system, which comprises the following modules: an initialization module for initializing parameters of semiconductor detection equipment; the warping prediction module is used for establishing an edge warping prediction model and predicting edge warping conditions of wafers with different diameters; the diameter compensation module is used for establishing an edge compensation model and calculating a diameter compensation value according to the wafer edge warping degree predicted by the edge warping prediction model; the feedback adjustment module is used for automatically adjusting image generation parameters of the semiconductor detection equipment according to the diameter compensation value; the verification module is used for carrying out comparison verification on the compensated wafer image and the actual wafer diameter according to experimental analysis; and the application module performs automatic matching and diameter compensation. According to the method, the edge warping prediction model and the diameter compensation model are introduced, so that the diameter measurement error caused by wafer warping can be effectively predicted and compensated, and a remarkable effect on pre-calibration of semiconductor detection equipment is achieved.

Description

[0001] This application is a divisional application of the application filed on September 18, 2024, with application number 202411298032.4 and invention name “A semiconductor detection equipment calibration system”. Technical Field

[0002] The present invention relates to the technical field of equipment calibration, and in particular to a semiconductor testing equipment calibration system. Background Art

[0003] Semiconductor inspection equipment is used for wafer inspection, which includes the diameter inspection of the wafer to see if it is qualified. In the semiconductor manufacturing industry, the accuracy of the inspection equipment is crucial to ensuring the quality of the final product. As the basic material for semiconductor manufacturing, the dimensional accuracy of the wafer, especially the precise measurement of its diameter, is one of the key indicators for evaluating its quality. However, wafers often have edge warping due to their physical properties and different diameter sizes. This warping will directly affect the measurement accuracy of the wafer diameter.

[0004] There are many reasons for warping, including but not limited to internal stress of the material, temperature gradient, mechanical stress during processing, and stress mismatch between materials during thin film deposition. Different wafer diameters will also aggravate this problem. Wafers with larger diameters are more prone to edge warping. This warping not only affects the measurement of the diameter, but may also affect the microcircuit pattern on the wafer, thereby affecting the performance of semiconductor devices.

[0005] Although the design and manufacturing technology of semiconductor inspection equipment has made significant progress, the existing technology often fails to provide effective compensation solutions for the problem of wafer warpage. Traditional inspection equipment often does not take into account the impact of wafer warpage on measurement results during calibration, which limits the improvement of detection accuracy and increases the uncertainty in the manufacturing process. Therefore, there is an urgent need to develop a new semiconductor inspection equipment calibration system. Summary of the invention

[0006] Based on the above objectives, the present invention provides a semiconductor testing equipment calibration system.

[0007] A semiconductor testing equipment calibration system includes the following modules:

[0008] Initialization module: Initialize the parameters of semiconductor inspection equipment, including image processing parameters;

[0009] Warpage prediction module: Based on the different diameters and material properties of wafers, an edge warpage prediction model is established to predict the edge warpage of wafers with different diameters;

[0010] Diameter compensation module: establishes an edge compensation model and calculates the diameter compensation value based on the wafer edge warpage predicted by the edge warpage prediction model;

[0011] Feedback adjustment module: automatically adjusts the image generation parameters of the semiconductor inspection equipment according to the diameter compensation value, and then obtains the compensated wafer image through the semiconductor inspection equipment imaging to compensate for the influence of warpage on the inspection results;

[0012] Verification module: Based on experimental analysis, the compensated wafer image is compared and verified with the actual wafer diameter;

[0013] Application module: The wafer diameter is divided into multiple diameter range segments, and a range segment database is established. After the wafer image is actually detected and acquired, the diameter of the wafer image before compensation is imported into the range segment database, and the diameter compensation is automatically matched and performed according to the compensation values ​​of multiple range segments.

[0014] Furthermore, the image processing parameters include pixel distance. When diameter compensation is performed, the pixel distance in the image is first compensated to compensate for the diameter after the pixel distance is converted into the actual physical distance.

[0015] Furthermore, the material properties include elastic modulus, thermal expansion coefficient and material stress.

[0016] Furthermore, the edge warping prediction model specifically includes:

[0017] Parameter definition: D is the wafer diameter, E is the elastic modulus of the material, α is the thermal expansion coefficient of the material, ΔT is the temperature change, σ is the stress of the material, and δ is the warpage;

[0018] A physical model is used to describe the relationship between warpage and parameters: σ = EαΔT, where σ represents the internal stress of the material caused by the temperature change ΔT;

[0019] The warpage δ is estimated based on the stress σ and the wafer diameter D. The warpage is regarded as the bending of the circular plate, and its warpage is expressed as: Where f(D) is a function related to the wafer diameter, which is used to describe the geometric factor of the effect of diameter on warpage;

[0020] Elastic modulus E: reflects the material's ability to resist deformation. The larger E is, the harder the material is and the less likely it is to warp.

[0021] Thermal expansion coefficient α: describes the degree to which the unit length of a material changes due to temperature changes. The larger α and ΔT are, the greater the internal stress caused by temperature changes, resulting in greater warping.

[0022] Function f(D): describes how the wafer diameter affects the degree of warpage. The larger the wafer diameter, the greater the possibility of uneven force on the edge, and the degree of warpage may increase.

[0023] Furthermore, the specific function form of f(D) is:

[0024] f(D)=aD b +c, where D is the diameter of the wafer, a, b and c are model parameters, warpage data of wafers with different diameters are collected and determined by data fitting, aD b a represents the main trend of the warpage as a function of diameter, b represents the degree of nonlinearity of the relationship, and c represents the baseline warpage, which exists even at small diameters.

[0025] Furthermore, the data fitting adopts a nonlinear NLS method to minimize the difference between the model prediction value and the actual observation value, specifically including:

[0026] Collect a series of wafers with different diameters and the corresponding edge warpage measurements as input to the fitting process. Let D i is the diameter of the i-th wafer sample, W i is the corresponding actual edge warpage measurement value, where i = 1, 2, ..., N, N is the total number of samples;

[0027] The objective function (also called loss function or cost function) is defined to quantify the difference between the model predictions and the actual measurements, i.e. the sum of the squares of the differences over all samples:

[0028] Use numerical optimization methods to minimize the objective function S(a, b, c) to find the optimal parameters a, b and c, which is done through the gradient descent iterative algorithm;

[0029] By calculating the coefficient of determination R 2 Indicators to evaluate the quality and predictive ability of the fitted model, R 2 A value close to 1 indicates a good fit.

[0030] Furthermore, the edge compensation model calculates the diameter compensation value C based on the warpage δ to correct the diameter measurement error caused by warpage. The edge compensation model is expressed as: C=γδ+ζ, wherein C is the compensation value that needs to be applied to the measured diameter, γ is the compensation coefficient, which indicates the degree of influence of the warpage δ on the compensation value C, and ζ is the baseline compensation value, which is the compensation required when there is no warpage, i.e., δ=0.

[0031] Furthermore, the compensation coefficient γ and the baseline compensation value ζ are determined by applying the nonlinear NLS method, which is expressed as: Among them, C i and δ iare the actual compensation value of the i-th wafer sample and the warpage obtained by the edge warpage prediction model.

[0032] Once the model parameters γ and ζ are determined, the compensation model can be applied to any new wafer sample. First, the warpage of the wafer is calculated by the edge warpage prediction model, and then the corresponding diameter compensation value C is calculated using the compensation model.

[0033] Furthermore, the feedback adjustment module specifically includes:

[0034] Determine pixel scale: Pixel scale is defined as the actual physical length represented by one pixel in the image captured by the industrial camera, expressed as: For example, if a standard-sized object is known to occupy 100 pixels in an image, and the actual length of the object is 1 mm, then the pixel scale is 0.01 mm / pixel;

[0035] Calculate pixel compensation value: After determining the pixel scale, convert the compensation value C into the pixel distance in the image. 像素 Calculated by the following formula: C is the diameter compensation value calculated by the edge compensation model, C 像素 is the pixel compensation value corresponding to the diameter compensation value;

[0036] Apply pixel compensation value: calculated pixel compensation value C 像素 Directly used for image processing parameter adjustment.

[0037] The application modules are as follows:

[0038] 1. Define the diameter range segment:

[0039] According to the size distribution and detection requirements of the wafer, the wafer diameter is divided into several range segments. These range segments cover all possible wafer diameters, and the width of each range segment should be determined according to the accuracy requirements of the actual application. For example, if the wafer diameter ranges from 100 mm to 300 mm, it can be divided into several 20 mm wide range segments.

[0040] 2. Establish a range segment database:

[0041] Create a database to store the relevant information of each diameter range segment, including the range segment boundary and the corresponding compensation value. Each record in the database corresponds to a diameter range segment.

[0042] 3. Actual detection and diameter matching:

[0043] Automatic diameter compensation is performed when semiconductor inspection equipment acquires wafer images and measures their diameters.

[0044] 4. Automatic compensation process:

[0045] The process of automatically matching and performing diameter compensation can be achieved through automation.

[0046] Beneficial effects of the present invention:

[0047] The present invention, by introducing an edge warpage prediction model and a diameter compensation model, can effectively predict and compensate for the diameter measurement error caused by wafer warpage. This prediction and compensation mechanism takes into account the physical properties of the wafer (such as diameter and material properties) and its performance in the image acquisition process, thereby ensuring that the measurement result is closer to the actual physical size of the wafer. This is particularly important for quality control in the semiconductor manufacturing process, because even a tiny size error may lead to a significant decrease in chip performance. It has a significant effect on the pre-calibration of semiconductor detection equipment, making it more accurate to make diameter judgments when actually acquiring wafer images.

[0048] The present invention improves the automation level and adaptability of the detection process by establishing a range segment database and realizing the process of automatic matching and diameter compensation. It can automatically identify the diameter range of the wafer and select a suitable compensation value for diameter correction. This automated process not only reduces the need for manual intervention, but also improves the efficiency and consistency of the detection process, especially when processing a large number of wafers.

[0049] The present invention can update and expand the edge warpage prediction model and the diameter compensation model to adapt to new wafer material characteristics or changes in manufacturing processes. By updating the compensation values ​​in the database or adjusting the model parameters, the calibrated detection equipment can quickly adapt to new detection requirements and maintain its measurement accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0051] Figure 1 Schematic diagram of system function modules according to an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0053] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0054] like Figure 1 As shown, a semiconductor testing equipment calibration system includes the following modules:

[0055] Initialization module: Initialize the parameters of semiconductor inspection equipment, including image processing parameters;

[0056] Warpage prediction module: Based on the different diameters and material properties of wafers, an edge warpage prediction model is established to predict the edge warpage of wafers with different diameters;

[0057] Diameter compensation module: establishes an edge compensation model and calculates the diameter compensation value based on the wafer edge warpage predicted by the edge warpage prediction model;

[0058] Feedback adjustment module: automatically adjusts the image generation parameters of the semiconductor inspection equipment according to the diameter compensation value, and then obtains the compensated wafer image through the semiconductor inspection equipment imaging to compensate for the influence of warpage on the inspection results;

[0059] Verification module: Based on experimental analysis, the compensated wafer image is compared and verified with the actual wafer diameter;

[0060] Application module: The wafer diameter is divided into multiple diameter range segments, and a range segment database is established. After the wafer image is actually detected and acquired, the diameter of the wafer image before compensation is imported into the range segment database, and the diameter compensation is automatically matched and performed according to the compensation values ​​of multiple range segments.

[0061] The image processing parameters include pixel distance. When compensating the diameter, the pixel distance in the image is first compensated to compensate for the diameter after the pixel distance is converted into the actual physical distance.

[0062] Material properties include elastic modulus, coefficient of thermal expansion, and material stress.

[0063] The edge warpage prediction model specifically includes:

[0064] Parameter definition: D is the wafer diameter, E is the elastic modulus of the material, α is the thermal expansion coefficient of the material, ΔT is the temperature change, σ is the stress of the material, and δ is the warpage;

[0065] A physical model is used to describe the relationship between warpage and parameters: σ = EαΔT, where σ represents the internal stress of the material caused by the temperature change ΔT;

[0066] The warpage δ is estimated based on the stress σ and the wafer diameter D. The warpage is regarded as the bending of the circular plate, and its warpage is expressed as: Where f(D) is a function related to the wafer diameter, which is used to describe the geometric factor of the effect of diameter on warpage;

[0067] Elastic modulus E: reflects the material's ability to resist deformation. The larger the E, the harder the material and the smaller the warping.

[0068] Thermal expansion coefficient α: describes the degree to which the unit length of a material changes due to temperature changes. The larger α and ΔT are, the greater the internal stress caused by temperature changes, resulting in greater warping.

[0069] Function f(D): describes how the wafer diameter affects the degree of warpage. The larger the wafer diameter, the greater the possibility of uneven force on the edge, and the degree of warpage may increase.

[0070] Considering the physical relationship between wafer warpage and diameter, wafer warpage is usually caused by internal stress, which originates from material properties, heat treatment during processing, and stress mismatch factors of thin film deposition. As for the relationship between diameter and warpage, the degree of wafer warpage increases with the increase of diameter. The specific function form of f(D) is:

[0071] f(D)=aD b +c, where D is the diameter of the wafer, a, b and c are model parameters, warpage data of wafers with different diameters are collected and determined by data fitting, aD b a represents the main trend of the warpage as a function of diameter, b represents the degree of nonlinearity of the relationship, and c represents the baseline warpage, which exists even at small diameters.

[0072] The data fitting uses the nonlinear NLS method to minimize the difference between the model prediction value and the actual observation value, including:

[0073] Collect a series of wafers with different diameters and the corresponding edge warpage measurements as input to the fitting process. Let D i is the diameter of the i-th wafer sample, W iis the corresponding actual edge warpage measurement value, where i = 1, 2, ..., N, N is the total number of samples;

[0074] The objective function (also called loss function or cost function) is defined to quantify the difference between the model predictions and the actual measurements, i.e. the sum of the squares of the differences over all samples:

[0075] Use numerical optimization methods to minimize the objective function S(a, b, c) to find the optimal parameters a, b and c, which is done through the gradient descent iterative algorithm;

[0076] By calculating the coefficient of determination R 2 Indicators to evaluate the quality and predictive ability of the fitted model, R 2 A value close to 1 indicates a good fit.

[0077] The edge compensation model calculates the diameter compensation value C based on the warpage δ to correct the diameter measurement error caused by warpage. The edge compensation model is expressed as: C = γδ + ζ, where C is the compensation value that needs to be applied to the measured diameter, γ is the compensation coefficient, which indicates the degree of influence of the warpage δ on the compensation value C, and ζ is the baseline compensation value, which is the compensation required when there is no warpage, that is, δ = 0.

[0078] The compensation coefficient γ and the baseline compensation value ζ are determined by the nonlinear NLS method and are expressed as: Among them, C i and δ i are the actual compensation value of the i-th wafer sample and the warpage obtained by the edge warpage prediction model.

[0079] Once the model parameters γ and ζ are determined, the compensation model can be applied to any new wafer sample. First, the warpage of the wafer is calculated by the edge warpage prediction model, and then the corresponding diameter compensation value C is calculated using the compensation model.

[0080] The feedback adjustment module specifically includes:

[0081] Determine pixel scale: Pixel scale is defined as the actual physical length represented by one pixel in the image captured by the industrial camera, expressed as: For example, if a standard-sized object is known to occupy 100 pixels in an image, and the actual length of the object is 1 mm, then the pixel scale is 0.01 mm / pixel;

[0082] Calculate pixel compensation value: After determining the pixel scale, convert the compensation value C into the pixel distance in the image. 像素 Calculated by the following formula: C is the diameter compensation value calculated by the edge compensation model, C 像素is the pixel compensation value corresponding to the diameter compensation value;

[0083] Apply pixel compensation value: calculated pixel compensation value C 像素 Directly used for image processing parameter adjustment.

[0084] The application modules are as follows:

[0085] 1. Define the diameter range segment:

[0086] According to the size distribution and detection requirements of the wafer, the wafer diameter is divided into several range segments. These range segments cover all possible wafer diameters, and the width of each range segment should be determined according to the accuracy requirements of the actual application. For example, if the wafer diameter ranges from 100 mm to 300 mm, it can be divided into several 20 mm wide range segments:

[0087] Range 1: 100-119 mm;

[0088] Range 2: 120-139 mm;

[0089] Range segment 3: 140-159 mm; ...

[0091] The last range segment: 280-300 mm; in practical applications, the segmentation can be further refined to reduce the numerical span of the range segment and improve accuracy.

[0092] 2. Establish a range segment database:

[0093] Create a database to store the relevant information of each diameter range segment, including the range segment boundaries and corresponding compensation values. Each record in the database corresponds to a diameter range segment. The record structure is as follows:

[0094] Range segment ID;

[0095] minimum diameter;

[0096] Maximum diameter;

[0097] Compensation value;

[0098] The compensation value needs to be pre-calculated based on the experimental data and the prediction model and stored in the database.

[0099] 3. Actual detection and diameter matching:

[0100] When a wafer image is acquired and its diameter is measured by semiconductor inspection equipment, the following steps are performed for automatic diameter compensation:

[0101] Measuring wafer diameter: First, the diameter of the wafer image is measured (in pixels) using image processing technology, and then converted to the actual diameter (based on the pixel scale).

[0102] Range segment matching: Compare the measured wafer diameter with the range segments in the database to find the range segment that contains the diameter.

[0103] Apply compensation value: Compensate the measured wafer diameter according to the compensation value in the matched range segment record.

[0104] 4. Automatic compensation process:

[0105] The process of automatically matching and performing diameter compensation can be automated to achieve the following functions:

[0106] Read wafer image and perform diameter measurement.

[0107] Convert the measured diameter to actual size.

[0108] Query the database for the corresponding diameter range segment.

[0109] Read the compensation value for this range segment and apply it to the measured diameter.

[0110] In order to further explain the specific application of the solution of the present invention, the following example is given:

[0111] Semiconductor inspection equipment is being calibrated to measure wafer diameter and compensate for measurement errors caused by wafer edge warpage. The following is a complete example of the solution of the present invention:

[0112] The target wafer diameter range is 200mm to 300mm.

[0113] The wafer material is single crystal silicon, and the wafer edge warps due to temperature changes and chemical treatment during the production process. It is hoped that the detection equipment can automatically identify the wafer diameter and provide appropriate warp compensation for each diameter range to improve measurement accuracy.

[0114] Equipment calibration and testing:

[0115] Initialization and calibration: The device is calibrated before leaving the factory to set the basic parameters for image generation; the device is initially tested using a calibration wafer of standard size to determine the pixel scale (for example, 1 pixel = 0.001 mm).

[0116] Edge warpage prediction model establishment: Collect wafer samples of different diameters (200 mm to 300 mm) and measure their edge warpage; use the collected data to establish a warpage prediction model, for example, determine the model parameters through linear regression analysis.

[0117] Diameter compensation model establishment: For each diameter range (200-210mm, 210-220mm), measure the actual diameter of the warped wafer and calculate the required compensation value; establish a compensation model to convert the predicted warpage into a diameter compensation value.

[0118] Range segment database creation: Create a database record for each diameter range, including the minimum diameter, maximum diameter and corresponding compensation value.

[0119] Actual testing process:

[0120] Wafer diameter measurement: When a wafer to be tested is placed on the inspection platform, the equipment first automatically acquires the wafer image and measures its diameter. The measurement result is 250mm.

[0121] Warpage prediction and compensation value query: The equipment uses the edge warpage prediction model to predict the warpage of the wafer; based on the measured diameter (250mm), the equipment queries the compensation value of the corresponding range segment in the database.

[0122] Diameter correction: Based on the queried compensation value, the equipment automatically corrects the measured diameter to obtain the accurate diameter after compensation.

[0123] Result output: The equipment displays the compensated wafer diameter and records the data in the test report for subsequent quality control and analysis.

[0124] Example summary: Through the above process, the semiconductor inspection device of the present invention can automatically compensate for the measurement error caused by the wafer edge warping, thereby improving the accuracy and reliability of wafer diameter measurement. This automated compensation mechanism is particularly suitable for high-precision semiconductor manufacturing processes, and helps to improve product quality and production efficiency.

[0125] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present invention is limited to these examples. Under the concept of the present invention, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the present invention as described above, which are not provided in detail for the sake of simplicity.

[0126] The present invention is intended to cover all such substitutions, modifications and variations that fall within the broad scope of the claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A semiconductor testing equipment calibration system, characterized in that: Includes the following modules: Initialization module: Initialize the parameters of semiconductor inspection equipment, including image processing parameters; Warpage prediction module: Based on the different diameters and material properties of wafers, an edge warpage prediction model is established to predict the edge warpage of wafers with different diameters; Diameter compensation module: establish an edge compensation model, calculate the diameter compensation value according to the wafer edge warpage predicted by the edge warpage prediction model, and the edge warpage prediction model specifically includes: Parameter definition: D is the wafer diameter, E is the elastic modulus of the material, α is the thermal expansion coefficient of the material, ΔT is the temperature change, τ is the stress of the material, and δ is the warpage; A physical model is used to describe the relationship between the warpage degree and the parameters: τ = EαΔT, where σ represents the internal stress of the material caused by the temperature change ΔT; The warpage δ is estimated based on the stress σ and the wafer diameter D. The warpage is regarded as the bending of the circular plate, and its warpage is expressed as: Where f(D) is a function related to the wafer diameter, which is used to describe the geometric factor of the effect of diameter on warpage; The specific function form of f(D) is: f(D)=aD b +c, where D is the diameter of the wafer, a, b and c are model parameters, warpage data of wafers with different diameters are collected and determined by data fitting, aD b a represents the main trend of the warpage with diameter, b represents the degree of nonlinearity of the relationship, and c represents the baseline warpage, which is the warpage that exists even at small diameters; The data fitting adopts the nonlinear NLS method to minimize the difference between the model prediction value and the actual observation value, specifically including: Collect a series of wafers with different diameters and the corresponding edge warpage measurements as input to the fitting process. Let D i is the diameter of the i-th wafer sample, W i is the corresponding actual edge warpage measurement value, where i = 1, 2, …, N, N is the total number of samples; The objective function is defined to quantify the difference between the model predictions and the actual measurements, i.e. the sum of the squares of all sample differences: Use numerical optimization methods to minimize the objective function S(a'b'c) to find the optimal parameters a, b and c, which is done through a gradient descent iterative algorithm; By calculating the coefficient of determination R 2 Indicators to evaluate the quality and predictive ability of the fitted model, R 2 A value close to 1 indicates a good fit; Feedback adjustment module: automatically adjusts the image generation parameters of the semiconductor inspection equipment according to the diameter compensation value, and then obtains the compensated wafer image through the semiconductor inspection equipment imaging to compensate for the impact of warpage on the inspection results; Verification module: Based on experimental analysis, the compensated wafer image is compared and verified with the actual wafer diameter.

2. A semiconductor testing equipment calibration system according to claim 1, characterized in that: It also includes an application module: dividing the wafer diameter into multiple diameter range segments, and establishing a range segment database. After actually detecting and acquiring the wafer image, the diameter of the wafer image before compensation is imported into the range segment database, and the diameter compensation is automatically matched and performed based on the compensation values ​​of multiple range segments.

3. A semiconductor testing equipment calibration system according to claim 1, characterized in that: The image processing parameters include pixel distance. When compensating the diameter, the pixel distance in the image is first compensated to compensate for the diameter after the pixel distance is converted into the actual physical distance.

4. A semiconductor testing equipment calibration system according to claim 2, characterized in that: The material properties include elastic modulus, thermal expansion coefficient, and material stress.

5. A semiconductor testing equipment calibration system according to claim 1, characterized in that: The edge compensation model calculates the diameter compensation value C based on the warpage δ to correct the diameter measurement error caused by warpage. The edge compensation model is expressed as: C=γδ+ζ, wherein C is the compensation value that needs to be applied to the measured diameter, γ is the compensation coefficient, which indicates the degree of influence of the warpage δ on the compensation value C, and ζ is the baseline compensation value, which is the compensation required when there is no warpage, i.e., δ=0.

6. A semiconductor testing equipment calibration system according to claim 5, characterized in that: The compensation coefficient γ and the baseline compensation value ζ are determined by the nonlinear NLS method and are expressed as: Among them, C i and δ i are the actual compensation value of the i-th wafer sample and the warpage obtained by the edge warpage prediction model.

7. A semiconductor testing equipment calibration system according to claim 6, characterized in that: The feedback adjustment module specifically includes: Determine pixel scale: Pixel scale is defined as the actual physical length represented by one pixel in the image captured by the industrial camera, expressed as: Calculate pixel compensation value: After determining the pixel scale, convert the compensation value C into the pixel distance in the image. 像素 Calculated by the following formula: C is the diameter compensation value calculated by the edge compensation model, C 像素 is the pixel compensation value corresponding to the diameter compensation value; Apply pixel compensation value: calculated pixel compensation value C 像素 Directly used for image processing parameter adjustment.

Images