Method for measuring critical dimension of metal mask

Abstract

The invention provides a method for measuring the critical dimension of a metal mask. The method comprises the following steps: acquiring a first measurement image in a state that a backlight source is turned on and no test sample exists on a glass carrier; the inherent compensation value of the measurement equipment is calculated according to the first measurement image, that is, the illumination non-uniformity of the backlight source and the glass carrying table is quantified through the inherent compensation value, and basic data are provided for subsequent compensation based on the inherent error of the equipment. In the application, according to the relational expression between the gray-scale value and the spacing dimension, the key dimension measurement value of the to-be-tested sample is compensated, the influence of measurement equipment is eliminated, and the measurement precision and reliability are improved.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor technology, and particularly to a method for measuring the critical dimension of a metal mask. Background Art

[0002] A mask (MASK) is mainly used as a master for pattern transfer and has typical customization characteristics. Masks are divided into photomasks (reticles) and metal masks (shadow masks). Photomasks are mainly used in the lithography process and are made of high-purity quartz glass for making circuits. A fine metal mask (FMM) is a key component used in the manufacture of OLED displays, mainly for precisely depositing organic materials to form pixel patterns.

[0003] A total pitch and critical dimension (TPCD) measurement device is a device used for measuring the critical dimension during the manufacture of FMM. The light source system of the TPCD measurement device adopts a strip-shaped backlight design, and the light-emitting source is an LED array light source. Due to the physical characteristics of the LED array light source, there are differences in the light-emitting intensity of LEDs in different regions, resulting in poor overall brightness uniformity.

[0004] In addition, there are flatness differences in the glass stage of the TPCD measurement device during actual use, which further exacerbates the uneven illuminance during the measurement of products in different regions.

[0005] The above two kinds of uneven illuminance directly affect the measurement result of the CD (Critical Dimension) by the TPCD measurement device, resulting in inherent differences in the CD measurement process and having a huge impact on the measurement accuracy of the critical dimension during the FMM manufacturing process. Summary of the Invention

[0006] The present application provides a method for measuring the critical dimension of a metal mask. By analyzing the measurement image without a sample to be tested, extracting the brightness distribution information of the backlight and the glass stage, and establishing a relationship between the gray-scale value and the pitch size, the compensation for the critical dimension of the sample to be tested is realized, and then the correction value of the critical dimension of the metal mask at different test points is obtained.

[0007] The object of the present invention is to provide a method for measuring the critical dimension of a metal mask, which is applied to a device for measuring the critical dimension of a metal mask. The metal mask has a first surface and a second surface, and a plurality of first holes and a plurality of second holes are respectively arranged at intervals on the above surfaces. The critical dimension is the minimum opening size at the intersection of the first hole and the second hole, and the method includes the following steps: S01. When the backlight is on and there is no test sample on the glass stage, obtain a first measurement image, and calculate the inherent compensation value of the key dimension of the measuring device at different preset test points in combination with the relationship formula between the gray-scale value and the pitch size. , where A is the total number of all preset test points, a ≤ A, and both A and a are positive integers; S02. When the backlight is on and a test sample to be tested is placed on the glass stage, obtain a second measurement image, and the position coordinates of the reference point and the b-th test point. B is the total number of all test points, b ≤ B, and both B and b are positive integers, where B < A; S03. Determine whether the shape and size of the holes in the metal mask plate reflected by the relationship formula between the gray-scale value and the pitch size adopted in S01 are consistent with the shape and size of the holes in the test sample to be tested in S02; S04. When the shape and size of the holes are both consistent, or when the shape of the holes is inconsistent or both the shape and size of the holes are inconsistent, use the first correction method for measurement; when the shape of the holes is consistent but the size of the holes is inconsistent, use a second correction method different from the first correction method for measurement.

[0008] In one embodiment, the first correction method is: find the inherent compensation value of the key dimension of the preset test point corresponding to the position coordinate of the b-th test point in the first measurement image , and according to generate the correction value of the key dimension of the metal mask plate at the b-th test point , where is the key dimension value of the standard position obtained through standard image testing.

[0009] In one embodiment, the second correction method is: If the shape of the holes is consistent but the size of the holes is inconsistent, then find the inherent compensation value of the key dimension of the preset test point corresponding to the position coordinate of the reference point in the first measurement image ; Find the inherent compensation value of the key dimension of the preset test point corresponding to the position coordinate of the b-th test point in the first measurement image ; Calculate the difference between the inherent compensation values of the key dimensions of the b-th test point and the reference point in the second measurement image , , and according to generate the correction value of the key dimension of the metal mask plate at the b-th test point .

[0010] In one embodiment, in S01, a first measurement image is acquired, and an inherent compensation value of a key dimension of the measurement device at different preset test points is calculated by combining the relationship between the gray scale value and the spacing dimension. , the steps include: S011, each time the backlight moves, one first measurement image is taken, and a total of A first measurement images are acquired. Image information of a preset test point is obtained for each first measurement image; S012, the average gray scale value of the a-th first measurement image is obtained , and the average gray scale value of the a-th first measurement image is substituted into the relationship between the gray scale value and the spacing dimension to determine the inherent compensation value of the preset test point where the a-th first measurement image is located , and the value of a is 1, 2, 3... A.

[0011] In one embodiment, in S01, a first measurement image is acquired, and an inherent compensation value of a key dimension of the measurement device at different preset test points is calculated by combining the relationship between the gray scale value and the spacing dimension. , the steps include: S0111, each time the backlight moves, one first measurement image is taken, and a total of A first measurement images are acquired. Image information of a preset test point is obtained for each first measurement image; S0112, the average gray scale values of the A first measurement images are respectively obtained , , ...... , and the mode value of the average gray scale values is selected therefrom as the reference value; S0113, the difference between the average gray scale value of the a-th first measurement image and the mode value is substituted into the relationship between the gray scale value and the spacing dimension to determine the inherent compensation value of the preset test point where the a-th first measurement image is located , and the value of a is 1, 2, 3... A.

[0012] In one embodiment, the calculation method of the average gray scale value of the a-th first measurement image includes: S01a. Obtain a plurality of pixel regions based on the critical dimension of the test sample to be compensated and the designed hole pitch. Among them, the first pixel region is a first circular region centered at the origin of the first measurement image with a radius equal to the sum of 1.5 times the critical dimension and 2 times the designed hole pitch; the second pixel region is an annular region obtained by subtracting the first circular region from a second circular region centered at the origin of the first measurement image with a radius equal to the sum of 2.5 times the critical dimension and 3 times the designed hole pitch; the third pixel region is the remaining pixel region after subtracting the first pixel region and the second pixel region from all regions of the a-th first measurement image. S01b. The gray-scale weight of the first pixel region is 100%, the gray-scale weight of the second pixel region is 95%, and the gray-scale weight of the third pixel region is 80%. S01c. Calculate the average gray-scale calculated value of the a-th first measurement image according to the following formula : , where is the average gray-scale value of the first pixel region, is the average gray-scale value of the second pixel region, is the average gray-scale value of the third pixel region.

[0013] In one embodiment, the steps of establishing the relationship formula between the gray-scale value and the spacing dimension include: S10. With the backlight turned on and fixedly set at the center of the glass stage, and a standard sample placed on the glass stage, determine the reference point (0, 0) and obtain the standard image of the reference point. S20. Select a plurality of processing ranges that are consistent with the standard hole size from the standard image, and perform image processing on the image information within each processing range to obtain M processed images, forming a total of M data groups (Xi, Yi); Among them, the process of image processing includes setting different gray-scale thresholds between 0 and 255, observing whether the boundary contour of the standard hole can be formed within the processing range. If the boundary contour of the standard hole is formed at a certain gray-scale threshold, record the gray-scale value Xi of the processing range and the measured value Yi of the critical dimension of the standard hole at this gray-scale threshold. The measured value Yi of the critical dimension of the standard hole is calculated through the boundary contour.

[0014] S30. Perform data fitting on the M data groups (Xi, Yi) to obtain the relationship formula between the gray-scale value X and the critical dimension Y of the standard hole.

[0015] In one embodiment, in step S30, the least squares method is used to perform data fitting on the multiple data groups (Xi, Yi), including the following steps: S301, provide M of the data groups (Xi, Yi), and assume that the values in the M data groups (Xi, Yi) satisfy the linear model: , where k is the slope and b is the intercept, is the random error; S302, through the following formulas (1), (2) and (3), calculate the slope k and the intercept b on the premise of minimizing the residual sum of squares RSS: Formula (1); Formula (2); Formula (3); S303, calculate the coefficient R through the following formulas (4) and (5) 2 , if , then determine that the relationship between the grayscale value and the pitch size satisfies the linear model Y = kX + b, Formula (4); Formula (5).

[0016] In one embodiment, in step S303, if , then assume that the values in the M data groups (Xi, Yi) satisfy the quadratic model: , where d1 is the quadratic term coefficient, d2 is the linear term coefficient, and e is a constant, is the random error; Solve for d1, d2, and e through operations and perform model verification.

[0017] In one embodiment, the second measurement image obtained in S02 is an image with the first surface as the upper surface; The standard image obtained in S10 is an image with the first surface as the upper surface; The method further includes: a step of verifying the relationship between the grayscale value and the pitch size, including: S100, in the state where the backlight is turned on and fixedly arranged at the center of the glass stage, a standard sample is placed on the glass stage and the second surface is the upper surface, determine the reference verification point (0, 0), and obtain the standard verification image of the reference verification point; S200. Select multiple verification processing ranges that match the minimum opening size from the standard verification image, and perform image processing on the image information within each verification processing range to obtain N processed images, thus forming N data groups (Xj, Yj) in total. Among them, the process of the image processing includes setting different grayscale thresholds (threshold parameters for binarization) between 0 and 255, observing whether the boundary contour of the minimum opening size can be formed within the verification processing range. If, at a certain grayscale threshold, the boundary contour of the minimum opening size is formed, record the grayscale value Xj of the verification processing range and the measured value Yj of the minimum opening size at this grayscale threshold. Among them, the measured value Yj of the minimum opening size is calculated through the boundary contour.

[0018] S300. Perform data fitting on the N data groups (Xj, Yj) to obtain the relationship formula between the grayscale value X and the minimum opening size Y. S400. If the "relationship formula between the grayscale value X and the minimum opening size Y" and the "relationship formula between the grayscale value and the spacing size" are of the same function model, and the difference in the grayscale value coefficients is less than 0.01, then the verification of the relationship formula between the grayscale value and the spacing size is completed.

[0019] In one embodiment, the width of the backlight source is and the backlight source moves at a frequency of mm / time, and the center of the measured image obtained each time coincides with the center of the backlight source.

[0020] The present invention has at least the following advantages or beneficial effects: In this application, in the state where the backlight source is turned on and there is no test sample on the glass stage, the first measured image is obtained. Calculate the inherent compensation value of the measuring device according to the first measured image, which is equivalent to quantifying the illumination non-uniformity of the backlight source and the glass stage through the inherent compensation value, providing basic data for subsequent compensation based on the inherent error of the device. According to the relationship formula between the grayscale value and the spacing size, compensate the measured value of the key dimension of the test sample to eliminate the influence of the measuring device and improve the measuring accuracy and reliability.

[0021] In this application, by analyzing the measured image without the product, extract the brightness distribution information of the light source and the glass stage, accurately calculate the influence of the measuring device on the measured value of the key dimension, so as to eliminate the measurement deviation caused by illumination non-uniformity. In this application, use image processing technology to quickly obtain the brightness distribution information of each region, without additional measuring devices, reducing the measurement cost and time.

[0022] In this application, the compensation of the measurement results is achieved through the relationship between the grayscale value and the spacing size, which significantly improves the reliability and consistency of the measurement results. By determining the inherent compensation values ​​of different preset test points , which is absolute and can clearly define the inherent compensation value of each preset test point, facilitating its application in subsequent compensation processes.

[0023] In this application, due to the special structural design of the metal mask (there are the first hole and the second hole), when the backlight source irradiates the second surface (back side) of the metal mask, the propagation of light will change, in order to avoid the inhomogeneity of light when the backlight source irradiates the metal mask and affect the accuracy of the grayscale value. It has been proved through practice that according to this embodiment, the first measurement image is set to multiple pixel areas, and different pixel areas are set with different grayscale weights, so that the grayscale calculation value of the a-th preset test point is closer to the actual shooting value of the second measurement image (the second measurement image is the shot image after being affected by the hole structure of the metal mask).

[0024] In this application, for different types of metal masks (with different hole types / different sizes), different hole types have different critical dimensions, and different grayscale values ​​and spacing dimensions are fitted. Based on algorithms and a large amount of data measurement and processing, the grayscale value and spacing dimension relationship is established to achieve accurate and quantitative evaluation of the relationship between grayscale value and critical dimension. Coefficient R 2 The closer it is to 1, the more accurate the fitting model is. , then assume that the values ​​in the M data groups (Xi, Yi) satisfy the quadratic model, or combine the data characteristics to determine whether it is necessary to verify the polynomial fitting model, fractional fitting model, piecewise fitting model, etc.

[0025] In this application, a verification method for the relationship between grayscale value and spacing size is provided, and the verified relationship is more accurate. The value of the critical size calculated by the relationship is also more real and reliable. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0027] Figure 1 A schematic diagram of a process for measuring a critical dimension of a metal mask provided by an embodiment of the present invention; Figure 2Schematic cross-sectional view of a partial metal mask provided by an embodiment of the present invention; Figure 3 Schematic diagram of a first measurement image provided by an embodiment of the present invention; Figure 4 Schematic diagram of a second measurement image provided by an embodiment of the present invention; Figure 5a 、 Figure 5b 、 Figure 5c 、 Figure 5d 、 Figure 5e Shapes of different types of holes in different metal masks in embodiments of the present application; Figure 6a Schematic structure diagram when taking an image of a preset test point 1 in an embodiment of the present application; Figure 6b Schematic structure diagram when taking an image of a preset test point 2 in an embodiment of the present application; Figure 6c Schematic structure diagram when taking an image of a preset test point A in an embodiment of the present application; Figure 7a First measurement image with position coordinates (0, 159) in an embodiment of the present application; Figure 7b Gray scale value distribution diagram of the first measurement image in an embodiment of the present application; Figure 8a Schematic diagram of dividing the a-th first measurement image into three pixel regions in an embodiment of the present application; Figure 8b Schematic diagram showing the division logic of three pixel regions in the second measurement image in an embodiment of the present application; Figure 8c Schematic diagram showing the division logic of three pixel regions through cross-sectional views during the image shooting process on both sides in an embodiment of the present application; Figure 9a Gray scale image of a standard hole formed when the gray scale threshold is equal to 92 in an embodiment of the present application; Figure 9b Gray scale image of a standard hole formed when the gray scale threshold is equal to 112 in an embodiment of the present application; Figure 9c Gray scale image of a standard hole formed when the gray scale threshold is equal to 122 in an embodiment of the present application; Figure 9d Gray scale image of a standard hole formed when the gray scale threshold is equal to 132 in an embodiment of the present application; Figure 10a The second measurement image obtained in S02 in an embodiment of the present application is an image with the first surface as the upper surface; Figure 10bIn the embodiment of the present application, the image obtained in S100 is an image with the second surface as the upper surface; Figure 11 This is a schematic diagram of the critical dimension measurement device of the metal mask in the embodiment of the present application; Figure 12 This is the specific structure diagram of the backlight source in the embodiment of the present application; Figure 13 This is the grayscale data of the flatness of the glass stage in the embodiment of the present application; Figure 14 This is the grayscale distribution trend diagram of all preset test points in the embodiment of the present application; Figure 15 This is the distribution diagram of the actual test points in the second side images in the embodiment of the present application; Figure 16a This is the relational expression of the two grayscale values and the spacing dimension fitted by the first hole type in the embodiment of the present application; Figure 16b This is the relational expression of the two grayscale values and the spacing dimension fitted by the second hole type in the embodiment of the present application; Figure 16c This is the relational expression of the two grayscale values and the spacing dimension fitted by the third hole type in the embodiment of the present application; Figure 16d This is the relational expression of the two grayscale values and the spacing dimension fitted by the fourth hole type in the embodiment of the present application; Figure 16e This is the relational expression of the two grayscale values and the spacing dimension fitted by the fifth hole type in the embodiment of the present application.

[0028] Icon: Light source 10, glass stage 20, metal mask 30: first surface 31, small hole 311, second surface 32, large hole 321. Detailed implementation manners

[0029] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some but not all of the embodiments of the present invention. Generally, the components of the embodiments of the present invention described and illustrated herein can be arranged and designed in various different configurations.

[0030] Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed present invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0031] It should be noted that like reference numerals and letters refer to like items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0032] In the description of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the inventive product is customarily placed during use. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. In addition, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and should not be construed as indicating or implying relative importance.

[0033] In addition, terms such as "horizontal" and "vertical" do not mean that the components are required to be absolutely horizontal or hanging vertically, but can be slightly inclined. For example, "horizontal" only means that its direction is more horizontal relative to "vertical", and does not mean that the structure must be completely horizontal, but can be slightly inclined.

[0034] In the description of the present invention, it should also be noted that unless otherwise clearly specified and limited, the terms "set", "install", "connect", and "couple" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0035] In the prior art, since the backlight source of the measurement device generally uses an LED array light source, the brightness non-uniformity of the LED array light source and the flatness difference of the glass stage result in an inherent measurement difference in the TPCD measurement device when measuring CD. The specific manifestations are as follows: the illuminance non-uniformity causes deviations in the CD measurement values in different regions, affecting the measurement accuracy. In the prior art, there is a lack of effective means to measure the illuminance magnitude in each region, and it is impossible to accurately determine the actual influence of the measurement device on the CD measurement result. Due to the inability to accurately determine the influence of the measurement device, it is difficult to effectively compensate and correct the measurement result, resulting in a low reliability of the measurement result.

[0036] The present application provides a method for measuring the critical dimension of a metal mask. By analyzing the measurement image without the sample to be tested, the brightness distribution information of the backlight source and the glass stage is extracted, and the relationship between the gray scale value and the spacing dimension is established, so as to realize the compensation of the critical dimension of the sample to be tested, and then obtain the correction value of the critical dimension of the metal mask at the b-th test point.

[0037] Please refer to Figure 1 , the present application provides a method for measuring the critical dimension of a metal mask, which is applied to the measuring equipment for the critical dimension of the metal mask. As Figure 2 shown, a cross-sectional schematic diagram of a part of the metal mask 30 is intercepted. The metal mask 30 includes a first surface 31 and a second surface 32. There is a first hole 311 (small hole) on the first surface 31, and a second hole 321 (large hole) on the second surface 32. The critical dimension is the minimum opening dimension at the intersection of the first hole 311 and the second hole 321 (or called the critical dimension, minimum feature size), such as the CD value shown in Figure 2 . The minimum opening dimension is the minimum feature size that needs to be strictly controlled during the manufacturing process of the FMM, which directly affects the performance of the OLED device.

[0038] The method for measuring the critical dimension of the metal mask involved in the present application includes the following steps: S01, in the state where the backlight source is turned on and there is no test sample on the glass stage, obtain the first measurement image as shown in Figure 3 , and calculate the inherent compensation value of the critical dimension of the measuring equipment at different preset test points in combination with the relationship between the gray scale value and the spacing dimension , where A is the total number of all preset test points, a ≤ A, and both A and a are positive integers. Among them, the spacing dimension at different preset test points obtained by substituting the gray scale values of the first measurement image at different preset test points into the relationship between the gray scale value and the spacing dimension is the inherent compensation value of the critical dimension of the measuring equipment at different preset test points . In this step, a compensation database for the inherent error of the measuring equipment can be calculated, or only the inherent error of the measuring equipment at the corresponding position can be obtained according to the detection requirements for subsequent compensation.

[0039] S02, in the state where the backlight source is turned on and a test sample is placed on the glass stage, obtain the second measurement image as shown in Figure 4 , and the position coordinates of the reference point and the b-th test point, where B is the total number of all test points, b ≤ B, and both B and b are positive integers, and among them, B < A. In this step, the field of view of the above first measurement image is a preset test point. The field of view of the second measurement image is a test point. The test point in the second measurement image can be determined according to the requirements of the OLED manufacturer.

[0040] S03. Determine whether the shape and size of the holes in the metal mask plate reflected by the relationship formula of the gray scale value and the pitch size used in S01 are consistent with the shape and size of the holes in the sample to be tested in S02. In this step, the FMM manufacturer can store multiple different relationship formulas. For example, if the relationship formula of the gray scale value and the pitch size stored by the FMM manufacturer satisfies a linear relationship, the shape of the holes in the sample to be tested is a rounded rectangle (or rectangle) as shown in Figure 5a and Figure 5b shown, and a rhombus as shown in Figure 5c shown; if it satisfies a quadratic relationship, the shape of the holes in the sample to be tested is a circle (or ellipse) as shown in Figure 5d shown; if it satisfies an inverse proportional relationship, the shape of the holes in the sample to be tested is a hourglass-like shape as shown in Figure 5e shown. Specifically, as shown in Figure 5a and Figure 5b and Figure 5c and Figure 5d and Figure 5e shown, they are the shapes of different types of holes in different metal mask plates respectively. Different hole types have different critical dimensions, and different relationship formulas of gray scale values and pitch sizes are fitted. Figure 5a and Figure 5b have the same hole shape but different hole sizes. Figure 5c and Figure 5d and Figure 5e have different hole shapes and different hole sizes. In addition, the shape and size of the holes in the sample to be tested in S02 can be obtained in two ways: One is the product description of the metal mask plate. Generally, before measuring the critical dimensions, it will describe the shape and size of the holes in the metal mask plate product to be tested, as well as other parameters of the metal mask plate product to be tested. The other is to observe the shape of the holes in the metal mask plate to be tested through the second measurement image obtained in S02, and measure the size of the holes in the metal mask plate to be tested through the second measurement image obtained in S02.

[0041] S04. When the shape and size of the holes are both consistent, or when the hole shapes are inconsistent, use the first correction method for measurement. When the hole shapes are consistent but the hole sizes are inconsistent, use a second correction method different from the first correction method for measurement. In this step, "the hole shapes are inconsistent" includes the following two situations: 1. The shape and size of the holes are both inconsistent; 2. The hole shapes are inconsistent but the hole sizes are consistent.

[0042] In this embodiment, in the state where the backlight is turned on and there is no test sample on the glass stage, a first measurement image is acquired. The inherent compensation value of the measuring device is calculated according to the first measurement image, which is equivalent to quantifying the illumination non-uniformity of the backlight and the glass stage through the inherent compensation value, providing basic data for subsequent compensation based on the inherent error of the device. On the one hand, according to the relational formula between the gray scale value and the pitch size, the measured value of the critical dimension of the test sample to be measured is compensated to eliminate the influence of the measuring device and improve the measurement accuracy and reliability. On the one hand, by analyzing the measurement image without a product, the brightness distribution information of the light source and the glass stage is extracted, and the influence of the measuring device on the measured value of the critical dimension is accurately calculated, thereby eliminating the measurement deviation caused by illumination non-uniformity. On the one hand, the image processing technology is used to quickly obtain the brightness distribution information of each region without additional measuring devices, reducing the measurement cost and time. On the one hand, through the relational formula between the gray scale value and the pitch size, the compensation of the measurement result is realized, significantly improving the reliability and consistency of the measurement result.

[0043] In one embodiment, the first correction method is as follows: Find the inherent compensation value of the critical dimension of the preset test point corresponding to the position coordinate of the b-th test point in the first measurement image , and according to Generate the correction value of the critical dimension of the metal mask at the b-th test point , where is the critical dimension value of the standard position obtained by testing with a standard image. In this step, the correction value of the critical dimension is calculated by an absolute calculation method The calculation process of, this calculation process can calculate the inherent compensation value of the critical dimension of the preset test point more accurately.

[0044] In one embodiment, the second correction method is as follows: Find the inherent compensation value of the critical dimension of the preset test point corresponding to the position coordinate of the reference point in the first measurement image ; Find the inherent compensation value of the critical dimension of the preset test point corresponding to the position coordinate of the b-th test point in the first measurement image ; Calculate the difference between the inherent compensation values of the critical dimensions of the b-th test point and the reference point in the second measurement image , , and according to Generate the correction value of the critical dimension of the metal mask at the b-th test point .

[0045] In this step, the first correction method and the second correction method are parallel solutions. When confirming the relationship between different gray-scale values and pitch sizes, different situations are classified by combining the shape and size of the holes in the metal mask, which can simplify the calculation process and improve the calculation efficiency.

[0046] In one embodiment, S01, obtain a first measurement image, and calculate the inherent compensation value of the key dimension of the measurement device at different preset test points by combining the relationship between the gray-scale value and the pitch size , the steps include: S011, each time the backlight moves, take a first measurement image, and a total of A first measurement images are obtained. Each first measurement image obtains the image information of a preset test point. In this step, the image information of this preset test point reflects the information of the backlight brightness and the glass stage at this preset test point. As Figure 6a is a structural schematic diagram when taking an image of the preset test point 1. As Figure 6b is a structural schematic diagram when taking an image of the preset test point 2. As Figure 6c is shown as a structural schematic diagram when taking an image of the preset test point A.

[0047] S012, obtain the average gray-scale value of the a-th first measurement image , and use the average gray-scale value of the a-th first measurement image as the gray-scale value, substitute it into the relationship between the gray-scale value and the pitch size, and determine that the pitch size of the preset test point where the a-th first measurement image is located is the inherent compensation value of the preset test point where the a-th first measurement image is located , and the value of a is 1, 2, 3... A. As Figure 7a shown is the first measurement image with the position coordinates (0, 159), Figure 7b is the gray-scale value distribution diagram of this first measurement image. When calculating the average gray-scale value of this first measurement image , obvious noises need to be removed. As Figure 7b the points with gray-scale within 0 - 100 in are noises, and these noises are Figure 7a the "cross" calibration lines in.

[0048] In this embodiment, by separately obtaining the average gray-scale values of different preset test points, substituting them into the relationship between the gray-scale value and the pitch size respectively, and respectively determining the inherent compensation values of different preset test points , it has absoluteness, can clarify the inherent compensation value of each preset test point, and is convenient for application in the subsequent compensation process.

[0049] In one embodiment, S01, obtain a first measurement image, and calculate the inherent compensation value of the key dimension of the measurement device at different preset test points by combining the relationship between the gray-scale value and the pitch size , the steps include: S0111, take a first measurement image each time the backlight moves once, and obtain a total of A first measurement images. Each first measurement image obtains the image information of a preset test point. In this step, the image information of this preset test point reflects the information of the backlight brightness and the glass stage at this preset test point. In this step, reference can be made to Figure 6a , Figure 6b and Figure 6c , when taking pictures of each preset test point, the center of the backlight coincides with the center of this preset test point.

[0050] S0112, obtain the average gray scale value of A first measurement images respectively , , ...... , and select the mode value of the average gray scale value from them as the reference value. In this step, in specific practical operations, the mode of the average gray scale values of A first measurement images can be set as the reference value, which can find the test points that need to be compensated more quickly.

[0051] S0113, use the difference between the average gray scale value of the a-th first measurement image and the mode value as the gray scale value and substitute it into the relational formula between the gray scale value and the pitch size to determine that the pitch size of the preset test point where the a-th first measurement image is located is the inherent compensation value of the preset test point where the a-th first measurement image is located , and the value of a is 1, 2, 3... A.

[0052] In this step, the difference in the key size of the standard hole is calculated through the difference between the average gray scale value of the a-th first measurement image and the mode value , which is the inherent compensation value of the preset test point where the a-th first measurement image is located .

[0053] In the above two embodiments of S011 - S012 and S0111 - S0113, when the backlight is turned on and there is no test sample on the glass stage, the first measurement image is taken / tested. Specifically, the width of the backlight is d, where 100mm - 20mm ≤ d ≤ 100mm + 20mm. During actual measurement, the backlight moves. The backlight moves for testing at a position interval of 10mm, that is, an image is collected every time the backlight moves 10mm. During normal testing, the gap between hole positions is very large, about 120mm. During actual measurement, the distance between the key dimensions of the standard holes is greater than 100mm. In a specific embodiment, the test points of the second measurement image may include: (420, 0), (300, 0), (180, 0), (0, 0), (-180, 0), (-300, 0), (-420, 0). (420, 32), (300, 32), (180, 32), (0, 32), (-180, 32), (-300, 32), (-420, 32). (420, -32), (300, -32), (180, -32), (0, -32), (-180, -32), (-300, -32), (-420, -32).

[0054] In one embodiment, the average gray scale value of the a-th first measurement image is calculated as follows: S01a, obtaining a plurality of pixel regions based on the key dimensions of the test sample to be compensated and the designed hole spacing. Among them, as Figure 8c shown, the key dimension (or CD value) is the minimum opening size at the intersection of the first hole 311 and the second hole 321. The designed hole spacing is the minimum distance between two second holes 321. As Figure 8a shown, the a-th first measurement image is divided into three pixel regions. Figure 8b shown, the division logic of the three pixel regions is illustrated in the second measurement image. Figure 8c shown, the division logic of the three pixel regions is illustrated through a cross-sectional view during the measurement image shooting process (note: Figure 8c only illustrates the setting logic, and the specific proportional relationship is not limited). Figure 8c illustrates the light source 10, the glass stage 20, and the metal mask 30, where the total width of the light source 10 is d, and the light source 10 includes 3 columns of LED dot matrices.

[0055] Specifically, the first pixel region is a first circular region centered at the origin of the first measurement image and with a radius equal to the sum of 1.5 times the critical dimension of the sample to be tested and 2 times the designed hole pitch. The second pixel region is an annular region formed by subtracting the first circular region from a second circular region centered at the origin of the first measurement image and with a radius equal to the sum of 2.5 times the critical dimension of the sample to be tested and 3 times the designed hole pitch. The third pixel region is the remaining pixel region after subtracting the first pixel region and the second pixel region from all regions of the a-th first measurement image.

[0056] S01b, the gray-scale weight of the first pixel region is 100%, the gray-scale weight of the second pixel region is 95%, and the gray-scale weight of the third pixel region is 80%.

[0057] S01c, calculate the average gray-scale calculated value of the a-th first measurement image according to the following formula : , where is the average gray-scale value of the first pixel region, is the average gray-scale value of the second pixel region, is the average gray-scale value of the third pixel region.

[0058] In this embodiment, due to the special structural design of the metal mask (there are the first hole and the second hole), when the backlight irradiates the second surface (back surface) of the metal mask, the propagation of light will change. In order to avoid the non-uniformity of light when the backlight irradiates the metal mask and affect the accuracy of the gray-scale value. It has been proved by practice that by setting the first measurement image as multiple pixel regions in this embodiment and setting different gray-scale weights for different pixel regions, the result of the gray-scale calculated value at the a-th preset test point is closer to the actual captured value of the second measurement image (the second measurement image is the captured image affected by the hole structure of the metal mask).

[0059] In one embodiment, the steps for establishing the relationship formula between the gray-scale value and the spacing dimension include: S10, in the state where the backlight is turned on and fixedly set at the center of the glass stage and a standard sample is placed on the glass stage, determine the reference point (0, 0) and obtain the standard image of the reference point. In this step, for example, if 35 standard holes can be observed in each shot, then select the most central standard hole within the field of view as the reference point.

[0060] S20. Select multiple processing ranges that match the standard hole size from the standard image, and perform image processing on the image information within each processing range to obtain M processed images, thus forming M data groups (Xi, Yi). In this step, one processing range is the range of one standard hole.

[0061] Among them, the process of image processing includes setting multiple different grayscale thresholds (threshold parameters for binarization) with a step size of 1 between 0 and 255, observing whether the boundary contour of the standard hole can be formed within the processing range. If the boundary contour of the standard hole is formed at a certain grayscale threshold, record the grayscale value Xi of the processing range and the measured value Yi of the key dimension of the standard hole at this grayscale threshold. Among them, the measured value Yi of the key dimension of the standard hole is calculated through the boundary contour. As Figure 9a 、 Figure 9b 、 Figure 9c and Figure 9d shown is the grayscale image of the standard hole formed at a certain grayscale threshold. Figure 9a is the grayscale image of the standard hole formed when the grayscale threshold is equal to 92. It can be clearly seen from the figure the boundary contour of the standard hole. Figure 9b is the grayscale image of the standard hole formed when the grayscale threshold is equal to 112. It can be clearly seen from the figure the boundary contour of the standard hole. Figure 9c is the grayscale image of the standard hole formed when the grayscale threshold is equal to 122. It can be clearly seen from the figure the boundary contour of the standard hole. Figure 9d is the grayscale image of the standard hole formed when the grayscale threshold is equal to 132. The boundary contour of the standard hole cannot be formed in the figure.

[0062] S30. Perform data fitting on the M data groups (Xi, Yi) to obtain the relationship between the grayscale value X and the key dimension Y of the standard hole.

[0063] In this embodiment, for metal mask plates of different models (with different hole shapes / different sizes), different hole shapes have different key dimensions, and the relationships between different grayscale values and pitch dimensions are fitted.

[0064] The fitting relationships of metal mask plates with different hole shapes are different. For example, the relationship between the grayscale value and the key dimension value of a non-straight-edge hole shape satisfies: , 。Among them and are respectively Figure 5a ( Figure 5b or Figure 5c )shown key dimensions in two directions.

[0065] In this embodiment, based on algorithms and a large amount of data measurement and processing, a relationship between grayscale values and pitch dimensions is established to achieve an accurate and quantitative assessment of the relationship between grayscale values and critical dimensions. In one embodiment, in step S30, the least squares method is used to perform data fitting on multiple data groups (Xi, Yi), including the following steps: S301, provide M data groups (Xi, Yi), and assume that the values in the M data groups (Xi, Yi) satisfy the linear model: , is the slope, is the intercept, is the random error.

[0066] S302, through the following formulas (1), (2), and (3), calculate the slope k and the intercept b on the premise of minimizing the residual sum of squares RSS: Formula (1).

[0067] Formula (2).

[0068] Formula (3).

[0069] S303, calculate the coefficient R through the following formulas (4) and (5) 2 , if , then determine that the relationship between the grayscale value and the pitch dimension satisfies the linear model , Formula (4).

[0070] Formula (5).

[0071] In this embodiment, the closer the coefficient R 2 is to 1, the more accurate the fitting model. If the relationship between the grayscale value and the pitch dimension satisfies the linear model , then substitute the grayscale value X into the above linear model, and output the critical dimension Y of the standard hole for verification. As long as the verification result meets the preset error.

[0072] In one embodiment, in step S303, if , then assume that the values in the M data groups (Xi, Yi) satisfy the quadratic model: , where d1 is the quadratic term coefficient, d2 is the linear term coefficient, e is a constant, is the random error.

[0073] Solve for d1, d2, and e through operations and perform model verification.

[0074] In this embodiment, in combination with data characteristics, it is determined whether it is necessary to verify models such as a polynomial fitting model (quadratic fitting model), a fractional fitting model, a piecewise fitting model, and an inverse proportional fitting model.

[0075] In any of the above embodiments of the present application, as Figure 10a shown, the second measurement image obtained in S02 is an image with the first surface as the upper surface. The standard image obtained in S10 is an image with the first surface as the upper surface. Obtaining a measurement image with the first surface as the upper surface in any of the above embodiments of the present application can avoid the influence of the second hole on the critical dimension, establish a more accurate relationship between the gray scale value and the pitch dimension, and the calculated critical dimension value of the metal mask is also more accurate.

[0076] In one embodiment, the method further includes: a step of verifying the relationship between the gray scale value and the pitch dimension, including: S100, in a state where the backlight is turned on and fixedly arranged at the center of the glass stage, a standard sample is placed on the glass stage and the second surface is the upper surface, determine the reference verification point (0, 0), and obtain a standard verification image of the reference verification point. As Figure 10b shown, the image obtained in S100 is an image with the second surface as the upper surface.

[0077] S200, select a plurality of verification processing ranges consistent with the minimum opening size from the standard verification image, and perform image processing on the image information within each verification processing range to obtain N processed images, and a total of N data groups (Xj, Yj) are formed.

[0078] Among them, the process of image processing includes setting different gray scale thresholds (threshold parameters for binarization) between 0 and 255, observing whether the boundary contour of the minimum opening size can be formed within the verification processing range. If the boundary contour of the minimum opening size is formed at a certain gray scale threshold, record the gray scale value Xj of the verification processing range and the measured value Yj of the minimum opening size at this gray scale threshold. Among them, the measured value Yj of the minimum opening size is calculated through the boundary contour.

[0079] S300, perform data fitting on the N data groups (Xj, Yj) to obtain the relationship between the gray scale value X and the minimum opening size Y.

[0080] S400, if the "relationship between the gray scale value X and the minimum opening size Y" and the "relationship between the gray scale value and the pitch dimension" are the same function model, and the difference between the gray scale value coefficients is less than 0.01, then the verification of the relationship between the gray scale value and the pitch dimension is completed. If it is greater than 0.01, analyze the reason and re-form the relationship between the gray scale value and the pitch dimension.

[0081] In this embodiment, a verification method for the relational expression between the grayscale value and the pitch size is given, and the relational expression after verification is more accurate. The numerical values of the critical dimensions calculated through this relational expression are also more authentic and reliable.

[0082] In one embodiment, the width of the backlight source is , and the backlight source moves at a frequency of mm / time, and the center of the measurement image obtained each time coincides with the center of the backlight source. In this step, as Figure 6a shown, when taking a picture of the preset test point 1, the backlight source is at the leftmost side. At this time, the light emitted by the 11th - 26th LED beads of the backlight source has the greatest influence on the preset test point 1. When taking a picture of the preset test point 11 (not shown in the figure), the backlight source is also at the leftmost side. At this time, the light emitted by the 27th - 43rd LED beads of the backlight source has the greatest influence on the preset test point 11. When taking pictures of the preset test point 1 and the preset test point 11, although the position of the backlight source is the same, the main light-emitting areas of the backlight source LEDs are different.

[0083] In a specific embodiment, the width of the backlight source is 100mm ± 20mm, and the backlight source moves at a frequency of 10mm / time. The sizes of the metal mask plates are 850mm * 70mm, 850mm * 150mm, 1200mm * 70mm, 1200mm * 150mm, 1200mm * 220mm, 1200mm * 310mm.

[0084] In one embodiment, after the number of times the measurement device is used exceeds the threshold number of times or the usage time exceeds the threshold time, the steps of S01 are executed once to re-obtain the inherent compensation values of the measurement device at different test points. Regularly updating the data in this step can avoid the problem that the brightness decays due to the degradation of the backlight source, which in turn causes inaccurate inherent compensation values. Therefore, the first measurement image is not static, and data will be re-obtained regularly. For example, it can be re-shot once a month.

[0085] In a specific embodiment, the above-mentioned method for measuring the critical dimensions of the metal mask plate is adopted, and a device for measuring the critical dimensions of the metal mask plate as shown in Figure 11 is used. The control module in this device is not shown in the figure. Figure 12 As shown, it is the specific structure diagram of the backlight source 10, and the backlight source 10 includes a plurality of LED chips arranged in a dot matrix. Figure 13 As shown, it is the grayscale data of the flatness of the glass stage. Through the Figure 13 color change, it can be clearly seen that the surface of the glass stage is not flat. Figure 14 is the grayscale distribution trend chart of all the preset test points. Through Figure 14The color change can clearly show that there are obvious differences in the gray-scale values of different preset points. All the preset test points cover the entire surface of the glass stage. Figure 15 It is the distribution map of the actual test points in the second measurement image. Figure 16a It is the relational expression of the two gray-scale values and the spacing dimension fitted by the first hole type; Figure 16b It is the relational expression of the two gray-scale values and the spacing dimension fitted by the second hole type; Figure 16c It is the relational expression of the two gray-scale values and the spacing dimension fitted by the third hole type; Figure 16d It is the relational expression of the two gray-scale values and the spacing dimension fitted by the fourth hole type; Figure 16e It is the relational expression of the two gray-scale values and the spacing dimension fitted by the fifth hole type.

[0086] In other embodiments, on the basis of adopting the key dimension measurement method of the metal mask in any of the above embodiments of the present application, other types of light sources (such as laser light sources) are used to replace the LED array light source, and the brightness uniformity is improved by optimizing the light source design. Using other types of light sources (such as laser light sources) to replace the LED array light source can fundamentally reduce the non-uniformity of the light source brightness and minimize the influence of the light source non-uniformity as much as possible. In other embodiments, on the basis of adopting the key dimension measurement method of the metal mask in any of the above embodiments of the present application, a high-precision glass stage is used, and the flatness difference is reduced by improving the manufacturing process. The high-precision glass stage can reduce the brightness difference caused by the unevenness of the stage and minimize the influence of the light source non-uniformity as much as possible.

[0087] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for measuring critical dimensions of a metal mask, which is applied to a measuring device for critical dimensions of a metal mask. The metal mask has a first surface and a second surface, and a plurality of first holes and a plurality of second holes are respectively arranged at intervals on the above surfaces. The critical dimension is the minimum opening dimension at the intersection of the first hole and the second hole. It is characterized in that, Including the following steps: S01. In a state where the backlight is turned on and there is no test sample on the glass stage, obtain a first measurement image, and calculate the inherent compensation value of the key dimension of the measurement device at different preset test points by combining the relationship between the gray scale value and the pitch size. , where A is the total number of all preset test points, a ≤ A, and both A and a are positive integers; among them, the pitch size of different preset test points obtained by substituting the gray scale values of the first measurement image at different preset test points into the relationship between the gray scale value and the pitch size is the inherent compensation value of the key dimension of the measurement device at different preset test points. ; S02. With the backlight turned on and the sample to be tested placed on the glass stage, obtain a second measurement image, as well as the position coordinates of the reference point and the b-th test point, where B is the total number of all test points, b ≤ B, and both B and b are positive integers, and B < A; S03. Determine whether the shape and size of the holes in the metal mask plate reflected by the relationship formula between the gray scale value and the pitch size used in S01 are consistent with the shape and size of the holes in the sample to be tested in S02; S04. When both the shape and size of the holes are consistent, or when the shapes of the holes are inconsistent, use the first correction method for measurement; when the shapes of the holes are consistent but the sizes are inconsistent, use a second correction method different from the first correction method for measurement.

2. The method for measuring critical dimensions of a metal mask according to claim 1, wherein It further includes: The first correction method is: Find the inherent compensation value of the critical dimension of the preset test point corresponding to the position coordinates of the b-th test point in the first measurement image , and based on generate the correction value of the critical dimension of the metal mask at the b-th test point , where is the critical dimension value of the standard position obtained through standard image testing; The second correction method is: Find the inherent compensation value of the key dimension of the preset test point corresponding to the position coordinates of the reference point in the first measurement image ; Find the inherent compensation value of the key dimension of the preset test point corresponding to the position coordinates of the b-th test point in the first measurement image ; Calculate the difference between the intrinsic compensation values of the critical dimensions of the b-th test point and the reference point in the second measurement image , , and based on generate the correction value of the critical dimension of the metal mask at the b-th test point .

3. The method for measuring the critical dimension of a metal mask plate according to claim 2, wherein S01, obtain a first measurement image, and calculate the inherent compensation value of the key dimensions of the measurement device at different preset test points by combining the relationship formula of the grayscale value and the spacing dimension , and the steps include: S011. Each time the backlight moves, take one first measurement image, and a total of A first measurement images are obtained. Each first measurement image obtains the image information of a preset test point; S012, obtain the average gray level value of the a-th first measurement image , and use the average gray level value of the a-th first measurement image as the gray level value, substitute it into the relational expression between the gray level value and the spacing dimension, and determine that the spacing dimension of the preset test point where the a-th first measurement image is located is the inherent compensation value of the preset test point where the a-th first measurement image is located , and the value range of a is 1, 2, 3... A.

4. The method for measuring the critical dimension of a metal mask plate according to claim 2, wherein S01. Obtain a first measurement image, and calculate the inherent compensation value of the key dimension of the measurement device at different preset test points by combining the relationship between the gray-scale value and the spacing dimension. , and the steps include: S0111. Each time the backlight moves, take one first measurement image, and a total of A first measurement images are obtained. Each first measurement image obtains the image information of a preset test point; S0112, respectively obtain the average gray scale value of A pieces of the first measurement images , , ...... , and select the mode value of the average gray scale values therefrom as the reference value; S0113, taking the difference between the average gray scale value of the a-th first measurement image and the mode value as the gray scale value and substituting it into the relational expression of the gray scale value and the spacing dimension to determine that the spacing dimension of the preset test point where the a-th first measurement image is located is the inherent compensation value of the preset test point where the a-th first measurement image is located , and the value range of a is 1, 2, 3... A.

5. The method for measuring critical dimensions of a metal mask according to claim 2 or 3, characterized in that, The average gray scale value of the a-th first measurement image The calculation method includes: S01a. Obtain multiple pixel regions according to the critical dimension of the test sample to be compensated and the designed hole pitch. The designed hole pitch is the minimum distance between two second holes. Among them, the first pixel region is a first circular region centered at the origin of the first measurement image with a radius equal to the sum of 1.5 times the critical dimension and 2 times the designed hole pitch; the second pixel region is an annular region obtained by subtracting the first circular region from a second circular region centered at the origin of the first measurement image with a radius equal to the sum of 2.5 times the critical dimension and 3 times the designed hole pitch; the third pixel region is the remaining pixel region after subtracting the first pixel region and the second pixel region from all regions of the a-th first measurement image; S01b. The gray scale weight of the first pixel region is 100%, the gray scale weight of the second pixel region is 95%, and the gray scale weight of the third pixel region is 80%; S01c, calculating the average grayscale calculated value of the a-th said first measurement image according to the following formula : , where is the average gray level value of the first pixel region, is the average gray level value of the second pixel region, is the average gray level value of the third pixel region.

6. The method for measuring the critical dimension of a metal mask according to claim 5, wherein The steps for establishing the relationship formula between the gray scale value and the pitch size include: S10. With the backlight turned on and fixedly set at the center of the glass stage and a standard sample placed on the glass stage, determine the reference point (0, 0) and obtain the standard image of the reference point; S20. Select multiple processing ranges from the standard image that are consistent with the standard hole size, and perform image processing on the image information within each processing range to obtain M processed images, forming a total of M data groups (Xi, Yi); Among them, the process of the image processing includes setting different grayscale thresholds between 0 and 255, observing whether the boundary contour of the standard hole can be formed within the processing range. If the boundary contour of the standard hole is formed at a certain grayscale threshold, record the grayscale value Xi of the processing range and the measured value Yi of the key dimension of the standard hole at this grayscale threshold. Among them, the measured value Yi of the key dimension of the standard hole is calculated through the boundary contour; S30. Perform data fitting on the M data groups (Xi, Yi) to obtain the relationship between the grayscale value X and the key dimension Y of the standard hole.

7. The critical dimension measurement method of the metal mask according to claim 6, wherein In the step of S30, the least squares method is used to perform data fitting on the multiple data groups (Xi, Yi), including the following steps: S301. Provide M of the data groups (Xi, Yi), and assume that the values in the M data groups (Xi, Yi) satisfy the linear model: , where k is the slope and b is the intercept, is the random error; S302. Through the following formulas (1), (2) and (3), calculate the slope k and the intercept b on the premise of minimizing the residual sum of squares RSS: Formula (1); Formula (2); Formula (3); S303, calculate the coefficient R through the following formula (4) and formula (5) 2 , if , then determine that the relational expression between the grayscale value and the pitch size satisfies the linear model , Formula (4); Formula (5).

8. The method for measuring the critical dimension of a metal mask according to claim 7, wherein, In step S303, if , it is assumed that the values in the M data groups (Xi, Yi) satisfy the quadratic model: , where d1 is the quadratic term coefficient, d2 is the linear term coefficient, e is a constant, is the random error; Solve d1, d2, e through operations and perform model verification.

9. The critical dimension measurement method of the metal mask according to claim 8, wherein The width of the backlight source is , and the backlight source moves at a frequency of mm / time, and the center of the measurement image obtained by each shot coincides with the center of the backlight source.

Images