CN119027371A - A focus evaluation method, system, device and medium for wafer alignment mark - Google Patents
A focus evaluation method, system, device and medium for wafer alignment mark Download PDFInfo
- Publication number
- CN119027371A CN119027371A CN202410990631.6A CN202410990631A CN119027371A CN 119027371 A CN119027371 A CN 119027371A CN 202410990631 A CN202410990631 A CN 202410990631A CN 119027371 A CN119027371 A CN 119027371A
- Authority
- CN
- China
- Prior art keywords
- focusing
- evaluation function
- sequence image
- wafer alignment
- alignment mark
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30148—Semiconductor; IC; Wafer
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Quality & Reliability (AREA)
- Testing Or Measuring Of Semiconductors Or The Like (AREA)
Abstract
本发明属于光学计量技术领域，并公开了一种面向晶圆对准标记的聚焦评价方法、系统、设备及介质，包括：获取晶圆对准标记在离焦‑聚焦‑离焦过程中的序列图像集；对所述序列图像集中各序列图像进行边缘提取；基于Brenner函数和Roberts函数在进行边缘提取后的各序列图像的不同方向上进行聚焦评价，得到各所述序列图像对应的初始聚焦评价函数值；对各所述序列图像对应的初始聚焦评价函数值进行方差计算，得到最终聚焦评价函数值；基于所述最终聚焦评价函数值判断晶圆对准标记的聚焦状态。本发明所述技术方案能够实现晶圆对准标记快速准确地自动聚焦。
The present invention belongs to the field of optical metrology technology, and discloses a focusing evaluation method, system, device and medium for wafer alignment marks, including: obtaining a sequence image set of wafer alignment marks in the process of defocus-focus-defocus; performing edge extraction on each sequence image in the sequence image set; performing focusing evaluation in different directions of each sequence image after edge extraction based on Brenner function and Roberts function to obtain initial focusing evaluation function values corresponding to each sequence image; performing variance calculation on the initial focusing evaluation function values corresponding to each sequence image to obtain final focusing evaluation function values; judging the focusing state of the wafer alignment mark based on the final focusing evaluation function value. The technical solution of the present invention can realize fast and accurate automatic focusing of the wafer alignment mark.
Description
Technical Field
The invention belongs to the technical field of optical measurement, and particularly relates to a focusing evaluation method, a focusing evaluation system, focusing evaluation equipment and a focusing evaluation medium for a wafer alignment mark.
Background
In the integrated circuit industry, the detection of the micro-nano structure quantity penetrates through the whole manufacturing process, wherein the key links comprise circuit design, wafer preparation, silicon wafer manufacturing, packaging, testing and the like. Although scanning electron microscopy, transmission electron microscopy, atomic force microscopy can all achieve dimension measurement on the nanoscale scale, their disadvantages are significant: the measuring speed is slow, the cost is high, and the equipment operation is complex. Therefore, the optical measurement technology with the advantages of high measurement speed, no contact, no damage, easy on-line integration and the like is widely applied to chip manufacturing process control and yield management. In the field of optical measurement technology, an auto-focusing technology is a key image acquisition and processing method, which is a core technology in the field. Aiming at the requirement of wafer alignment mark quantity detection, the automatic focusing technology can provide a high-precision, high-speed and reliable focus positioning method so as to obtain clear images and accurate measurement results. The technical core mainly comprises three parts of selection of a focusing window, selection of a focusing evaluation function and feedback control of a search algorithm on a stepping motor, wherein the focusing evaluation function is a weight of the three parts, and the performance of the focusing evaluation function can intuitively reflect the efficiency and the accuracy of automatic focusing.
At present, the focusing evaluation function is mainly divided into a space domain class, a frequency domain class and an information entropy class: the airspace function mainly comprises a Roberts function, a Brenner function, a Tenenrad function and the like, the functions describe focusing of an image through the change condition of the gray gradient of the image, the calculation is simple, the stability is high, and the airspace function is a focus evaluation function which is used in a large amount at present, but the noise immunity of the functions is weak; the frequency domain functions mainly comprise Fourier transform functions, discrete cosine transform functions and the like, are high in sensitivity, but sensitive to noise, are poor in instantaneity due to overlarge calculated amount, and are difficult to apply to actual conditions; the information entropy function describes the richness of information, the focusing of the image is in direct proportion to the change condition of the corresponding gray value interval, namely the information entropy value of the whole area, and the function has better unbiasedness, but the calculation process is complex and is easy to be influenced by illumination, and the sensitivity and the noise immunity are low.
In the prior art, SML and Roberts function are combined, and the proposed SMD-Roberts evaluation function has higher focusing sensitivity, so that accuracy and stability of focus evaluation value calculation can be improved, but time cost required by multiple times of focus evaluation operation is higher. The improved dual-threshold teningrad focusing evaluation function is suitable for cutter sequence images, and the improved method is difficult to apply to various occasions of measuring objects due to the fact that the final focusing evaluation result and the measuring accuracy depend on proper threshold selection although the high-frequency information of the images is fully utilized.
In the prior art, an improved Laplace focusing evaluation function is utilized to evaluate the focusing of imaging, and the method can improve the sensitivity of an evaluation algorithm to the x and y directions, but has weak noise resistance and low stability.
In the prior art, a part image high-precision focusing evaluation function for improving gradient weighting is also utilized for evaluation, an edge pixel point gradient value is obtained based on a 4-direction Sobel operator, the gradient accuracy is improved, then a pixel point gradient weighting coefficient is obtained according to the gray level distribution difference value of the edge pixel point and an 8-neighborhood pixel point, the sensitivity and noise immunity of a gradient weighting algorithm are enhanced, and the calculation is complicated and the instantaneity is poor.
Disclosure of Invention
The invention aims to provide a focusing evaluation method, a focusing evaluation system, focusing evaluation equipment and a focusing evaluation medium for a wafer alignment mark, so as to solve the problems in the prior art.
In order to achieve the above object, the present invention provides a focus evaluation method for a wafer alignment mark, including:
Acquiring a sequence image set of a wafer alignment mark in the defocus-focus-defocus process;
performing edge extraction on each sequence image in the sequence image set;
Performing focus evaluation on each sequence image in different directions after edge extraction based on a Brenner function and a Roberts function to obtain initial focus evaluation function values corresponding to each sequence image;
Performing variance calculation on the initial focus evaluation function value corresponding to each sequence image to obtain a final focus evaluation function value;
and judging the focusing state of the wafer alignment mark based on the final focusing evaluation function value.
Optionally, extracting edges of each sequence image in the sequence image set specifically includes:
Determining an initial sampling interval, and calculating the gray value standard deviation of each sequence image;
calculating gray value difference values of each pixel to be marked and adjacent pixel points in each sequence image, wherein the adjacent pixel points comprise pixel points positioned on the right and below the current pixel point to be marked;
And comparing the gray value standard deviation with the gray value difference of the adjacent pixel points, and determining whether the current pixel point to be marked is an edge pixel point or not based on a comparison result.
Optionally, comparing the gray value standard deviation with the gray value difference of the adjacent pixel points specifically includes:
If the gray value difference values of the adjacent pixel points are smaller than the gray value standard deviation, judging that the pixel point to be marked currently is not an edge pixel point;
If the gray value difference value larger than the gray value standard deviation exists in the gray value difference values of the adjacent pixel points, judging the current pixel point to be marked as an edge pixel point;
and after judging that the current pixel to be marked is an edge pixel, marking the current pixel to be marked and reducing the initial sampling interval.
Optionally, focus evaluation is performed on different directions of each sequence image after edge extraction based on a Brenner function and a Roberts function, which specifically includes:
Performing evaluation calculation in the directions of 0 DEG and 90 DEG of the sequence images by using a Brenner function, and adding pixel gray difference calculation in the direction of 90 DEG to obtain a first focusing evaluation function value; and (3) carrying out gray gradient calculation on the 45 DEG and 135 DEG directions of the sequence images by using a 3X 3Roberts operator to obtain a second focusing evaluation function value, and multiplying the first focusing evaluation function value and the second focusing evaluation function value to obtain an initial focusing evaluation function value.
Optionally, variance calculation is performed on the initial focus evaluation function value corresponding to each sequence image, and a specific calculation formula is as follows:
Where N is the number of sequential images, U Bre2d_Rob2d is the initial average evaluation function value of the sequential image set, and F Bre2d_Rob2d_Var is the final focus evaluation function value.
A focus evaluation system for wafer alignment marks, comprising:
The image acquisition module is used for acquiring a sequence image set of the wafer alignment mark in the defocus-focus-defocus process;
The focusing evaluation module is used for extracting edges of each sequence image in the sequence image set; performing focus evaluation on each sequence image in different directions after edge extraction based on a Brenner function and a Roberts function to obtain initial focus evaluation function values corresponding to each sequence image; performing variance calculation on the initial focus evaluation function value corresponding to each sequence image to obtain a final focus evaluation function value; and judging the focusing state of the wafer alignment mark based on the final focusing evaluation function value.
An electronic device includes a memory for storing a computer program and a processor that runs the computer program to cause the electronic device to perform the method of focus evaluation for wafer alignment marks.
A computer readable storage medium storing a computer program which when executed by a processor implements the method of focus evaluation for wafer alignment marks.
The invention has the technical effects that:
1. Aiming at the application background of the diversity change of the edge direction of the wafer alignment mark image characteristic, the invention combines the advantages of the Brenner function and the Roberts function, and can efficiently extract the edge gradient information of the mark image in multiple directions, thereby realizing high-precision focusing.
2. According to the invention, the traditional focusing evaluation function and the adaptive sampling are fused, so that the calculation complexity is improved, but the calculation amount is greatly reduced through the adaptive sampling, the efficient calculation is realized, the problems of poor noise resistance and stability of the traditional focusing evaluation function are solved, and the problems of complex calculation and poor instantaneity of the fused focusing function are solved.
3. According to the method, the variance calculation is carried out on the evaluation function, so that the problem of poor noise immunity of the airspace focusing function is improved, and better robustness is achieved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application. In the drawings:
FIG. 1 is a schematic view of a wafer alignment mark sequence image set according to an embodiment of the present invention;
FIG. 2 is a flowchart of a specific implementation of a focus evaluation method in an embodiment of the present invention;
fig. 3 is a graph comparing the focus evaluation function curves obtained by performing evaluation calculation on the sequence images of defocus-focus-defocus of 11 alignment marks by using 6 focus evaluation functions in the embodiment of the present invention.
Detailed Description
Various exemplary embodiments of the invention will now be described in detail, which should not be considered as limiting the invention, but rather as more detailed descriptions of certain aspects, features and embodiments of the invention.
It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. In addition, for numerical ranges in this disclosure, it is understood that each intermediate value between the upper and lower limits of the ranges is also specifically disclosed. Every smaller range between any stated value or stated range, and any other stated value or intermediate value within the stated range, is also encompassed within the invention. The upper and lower limits of these smaller ranges may independently be included or excluded in the range.
Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Although the invention has been described with reference to a preferred method, any method similar or equivalent to those described herein can be used in the practice or testing of the present invention. All documents mentioned in this specification are incorporated by reference for the purpose of disclosing and describing the methodologies associated with the documents. In case of conflict with any incorporated document, the present specification will control.
It will be apparent to those skilled in the art that various modifications and variations can be made in the specific embodiments of the application described herein without departing from the scope or spirit of the application. Other embodiments will be apparent to those skilled in the art from consideration of the specification of the present application. The specification and examples of the present application are exemplary only.
As used herein, the terms "comprising," "including," "having," "containing," and the like are open ended terms, meaning including, but not limited to.
It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other. The application will be described in detail below with reference to the drawings in connection with embodiments.
Example 1
As shown in fig. 1-fig. 1, in this embodiment, a focus evaluation method for a wafer alignment mark is provided, including: acquiring a sequence image set of a wafer alignment mark in the defocus-focus-defocus process; performing edge extraction on each sequence image in the sequence image set; performing focus evaluation on each sequence image in different directions after edge extraction based on a Brenner function and a Roberts function to obtain initial focus evaluation function values corresponding to each sequence image; performing variance calculation on the initial focus evaluation function value corresponding to each sequence image to obtain a final focus evaluation function value; and judging the focusing state of the wafer alignment mark based on the final focusing evaluation function value.
Performing adaptive sampling of pixel points and extraction of pixel points of marked characteristic edges on the obtained wafer alignment mark defocus-focus-defocus sequence image; then, evaluating and calculating the sampled pixel points by using an improved function combined with a Brenner function and a Roberts function to obtain an initial focus evaluation function value of each image; in order to reduce the sensitivity of the improvement function to noise, carrying out variance calculation on the initial focus evaluation function value of the image set to obtain a final focus evaluation function value; according to the focusing evaluation function curve of the image, the focal length can be regulated and controlled in real time, and an accurately focused image is obtained, so that high-precision detection of the wafer alignment mark is realized. The focusing evaluation function constructed by the embodiment has strong noise immunity, good stability and timeliness, and is suitable for automatic focusing of the wafer alignment marks with the varied edge directions.
1. Aiming at the application background of the diversity change of the edge direction of the wafer alignment mark image characteristic, the embodiment combines the advantages of the Brenner function and the Roberts function, and can efficiently extract the edge gradient information of the mark image in multiple directions, thereby realizing high-precision focusing.
2. According to the embodiment, the traditional focusing evaluation function and the adaptive sampling are fused, the calculation complexity is improved, but the calculation amount is greatly reduced through the adaptive sampling, so that efficient calculation is realized, the problems of poor noise resistance and stability of the traditional focusing evaluation function are solved, and the problems of complex calculation and poor instantaneity of the fused focusing function are solved.
3. According to the method, the variance calculation is finally carried out on the evaluation function, so that the problem of poor noise immunity of the airspace focusing function is improved, and better robustness is achieved.
The embodiment not only meets the unimodal property and unbiasedness of the focusing curve, but also has good stability and noise immunity, and also has an adaptive focusing evaluation function with high calculation efficiency, thereby realizing rapid and accurate automatic focusing of the wafer alignment mark.
The embodiment provides a self-adaptive focusing evaluation method for a wafer alignment mark aiming at the application background of the diversity change of the edge direction of the wafer alignment mark on the basis of the problems of high noise immunity, poor stability, large calculation amount of a fusion focusing evaluation function, poor instantaneity and the like of the traditional focusing evaluation function.
In order to achieve the purpose, the technical scheme adopted by the embodiment is as follows:
The adaptive focusing evaluation method for the wafer alignment mark comprises the steps of firstly, performing adaptive sampling of mark image pixel points and extraction of characteristic edge pixel points of the obtained wafer alignment mark defocusing-focusing-defocusing sequence images; then, an improvement function Brenner2d-Roberts2d is applied to evaluate and calculate the sampled pixel points, and an initial focus evaluation function value of each image is obtained; in order to reduce the sensitivity of the improved function to noise, enhance the stability and noise resistance of the improved function, variance calculation is carried out on the initial evaluation function value, and a final focus evaluation function Brenner2d-Roberts2d-Var is obtained.
The method specifically comprises the following steps:
(1) Marking image pixel adaptive sampling:
Determining different sampling intervals S 1 according to the sizes of the out-of-focus-out-of-focus sequence images of the wafer alignment marks obtained by the camera:
S1＝max(X，Y)/n
Wherein X, Y is the height and width of each image; and n is the sampling times, and the sampling times are determined according to the calculation efficiency of the evaluation function and the size of the wafer alignment mark.
(2) Extracting edge pixel points of marked image features:
In order to reduce the calculated amount, the simplest gray difference is adopted to judge the characteristic edge pixel point of the alignment mark image, and the gray value standard deviation sigma of each image is firstly calculated and is used as the judging basis of whether the gray value standard deviation sigma is the edge pixel point or not:
wherein f (x, y) is the gray value of the pixel point (x, y); XY is the image pixel size; u is the average pixel value of the whole image.
Then, calculating the gray level difference value between the pixel point (x, y) and the pixel point (x+1, y) on the right side and the pixel point (x, y+1) on the lower side, judging the gray level difference value with the gray level standard deviation sigma of the image, determining whether the pixel point is an edge pixel point, and timely adjusting the self-adaptive sampling interval S:
R＝|f(x,y)-f(x+1,y)
D＝|f(x,y)-f(x,y+1)
Wherein R is the absolute value of the gray difference between the pixel point (x, y) and the pixel point (x+1, y) on the right; d is the absolute value of the gray difference between the pixel (x, y) and the pixel (x, y+1) to the right.
Through the judgment of the gray level difference R, D and the standard deviation sigma, if R and D are smaller than the standard deviation sigma, the pixel point is not the marked edge, sampling is continued according to the large spacing S 1, if R or D is larger than the standard deviation sigma, the pixel point is the marked edge, the sampling spacing is reduced to S 2,S2, and the pixel point is determined according to the evaluation function calculation efficiency and the wafer alignment mark size.
(3) Improved focus evaluation function:
Considering that the wafer alignment mark has a plurality of characteristic edge directions, the present embodiment intends to perform evaluation function calculation for all four directions (0 °, 45 °, 90 °, 135 °) of the mark image. First, a Brenner function, which is one of the airspace-like focus evaluation functions having the lowest calculation complexity, is selected in the 0 ° and 90 ° directions, but it calculates only the squares of pixel gradation differences in the 0 ° direction which differ by two units, so that the pixel gradation difference calculation in the 90 ° direction is increased without increasing the calculation amount:
The Roberts function is then chosen in the 45 ° and 135 ° directions, which is to calculate the gray gradient values in the diagonal direction, but since the Brenner function operator is a 3×3 matrix, the Roberts operator is adjusted from 2×2 to 3×3 matrix in order to unify the gray gradient calculation dimensions:
finally, multiplying the two functions to obtain an initial focusing evaluation function which finally satisfies four directions:
(4) Variance calculation
The number of the evaluation function is increased to 4 from the calculation direction of the evaluation function, although the problem of azimuth limitation of the traditional airspace focusing evaluation function in gray gradient calculation is solved to a certain extent, the method is better suitable for the evaluation calculation of wafer alignment mark images, the calculation result is more accurate, the focusing sensitivity is improved to some extent, but the noise sensitivity of the evaluation function is increased, and the noise resistance is greatly weakened, so that variance calculation is added on the basis of the improvement function, the stability of the evaluation function can be enhanced by the variance function, and the evaluation function is not easy to be interfered by noise:
Where N is the number of out-of-focus-out-of-focus sequence image sets, u Bre2d_Rob2d is the initial average evaluation function value from the in-focus sequence image sets, and F Bre2d_Rob2d_Var is the final focus evaluation function.
And finally, fitting a focus evaluation function curve according to a focus evaluation function value of the defocusing-focusing-defocusing sequence image, so as to realize accurate focus positioning, acquire a clear focus image and realize high-precision detection of the wafer alignment mark.
The specific application object of this embodiment, the wafer bonding alignment mark autofocus system, is by which a sequential image set of alignment mark defocus-focus-defocus is first taken, as shown in fig. 1. After the image set is acquired, the adaptive focusing evaluation method for the wafer alignment mark is calculated according to the embodiment, and the flow is shown in fig. 2.
The detailed calculation flow comprises the following steps:
(1) Determination of pixel self-adaptive sampling interval S 1 and sampling times n:
the size of the defocusing-focusing-defocusing sequence image is 1280 multiplied by 1024 according to the wafer alignment mark obtained by the camera; n is the sampling times, and the calculation efficiency of the evaluation function and the size of the wafer alignment mark are determined to be 60; sampling interval S 1 is rounded to obtain:
S1＝max(X，Y)/n
＝max(1280，1024)/60
≈21
(2) Determining a wafer alignment mark characteristic edge pixel point extraction threshold sigma:
Firstly, calculating the gray value standard deviation sigma of each image, and taking the gray value standard deviation sigma as a judging basis of whether the gray value standard deviation sigma is an edge pixel point or not:
wherein f (x, y) is the gray value of the pixel point (x, y); XY is the image pixel size; u is the average pixel value of the whole image.
Then, calculating the gray level difference value between the pixel point (x, y) and the pixel point (x+1, y) on the right side and the pixel point (x, y+1) on the lower side, judging the gray level difference value with the gray level standard deviation sigma of the image, determining whether the pixel point is an edge pixel point, and timely adjusting the self-adaptive sampling interval:
R＝|f(x,y)-f(x+1,y)
D＝|f(x,y)-f(x,y+1)
Wherein R is the absolute value of the gray difference between the pixel point (x, y) and the pixel point (x+1, y) on the right; d is the absolute value of the gray difference between the pixel (x, y) and the pixel (x, y+1) to the right.
Through the judgment of the gray value difference R, D and the standard deviation sigma, if R and D are smaller than the standard deviation sigma, the pixel point is not the marked edge, sampling is continued according to the large spacing S 1, if R or D is larger than the standard deviation sigma, the pixel point is the marked edge, the sampling spacing is reduced to S 2,S2, and the pixel point is determined to be 1 according to the calculation efficiency of the evaluation function and the size of the alignment mark structure.
(3) And carrying out convolution calculation on the sampling pixel points by utilizing a multidirectional gradient focusing evaluation function:
First, the Brenner operator is used to calculate the pixel gray difference between two units in the 0 ° and 90 ° directions:
The pixel gray difference for two diagonal units is then calculated using the modified Roberts operator in the 45 ° and 135 ° directions:
Finally, multiplying the two functions to obtain an initial focusing evaluation function which finally satisfies four directions:
(4) Performing variance operation on the focusing evaluation function of the marker sequence image set:
Wherein N is the number of the out-of-focus-out-of-focus sequence image sets of the alignment mark to be 11, u Bre2d_Rob2d is the initial average evaluation function value of the out-of-focus sequence image sets, and F Bre2d_Rob2d_Var is the final focus evaluation function.
And finally, fitting a focus evaluation function curve according to a focus evaluation function value of the out-of-focus-out-of-focus sequence image of the alignment mark, thereby realizing accurate focus positioning, obtaining a clear focus image and realizing high-precision detection of the wafer alignment mark.
To verify the performance improvement of the method of this embodiment, this embodiment and the conventional focus evaluation function: performance contrast experiments and analyses were performed on Brenner, roberts, tenengrad functions, DFT two-dimensional discrete Fourier transformed image focus evaluation functions, and on Bre d_var evaluation functions proposed in 2023, dong Zhengqiong, etc.
And respectively calculating out the defocusing-focusing-defocusing sequence images of the same wafer alignment mark by using the focusing evaluation function, and normalizing the obtained focusing evaluation value to obtain a focusing evaluation function curve comparison graph, as shown in figure 3. In order to more objectively compare and analyze the performance differences of different focusing evaluation functions, performance indexes of the steep region width W, the definition ratio R, the steep Stp and the sensitivity f sen are selected to analyze and compare the focusing evaluation functions, and as shown in table 1, the relevant indexes are as follows.
Steep zone width W: the width W of the steep region represents the width of the region with a larger curve change rate on the evaluation function curve, and is a very important parameter depending on the specific shape and change rate of the evaluation function curve, and can be used as a selection reference index of the maximum step size of the focusing process.
Sharpness ratio R: the definition ratio R represents the resolution of the focus evaluation function on images with different defocus levels, and the larger the focus ratio is, the larger the difference between the focus value of the clear image and the focus value of the blurred image is, the easier the focus function is to be resolved:
Where f max is the maximum of the focusing function and f min is the minimum of the focusing function.
Abruptness Stp: the steepness Stp represents the steepness of the focus evaluation function curve at the peak, and the greater the steepness, the greater the resolution of the function to images of different defocus levels.
Sensitivity f sen: the sensitivity f sen represents the intensity of the change near the maximum value of the focus evaluation function, and the higher the sensitivity is, the more intense the change of the function is, the better the image focusing effect is, and conversely, the more gentle the change of the function is:
Where f max is the maximum of the focusing function and f (z max +θ) is the function value when the abscissa changes by θ.
TABLE 1 focusing evaluation function Performance evaluation index summary table
As can be seen from table 1, the steep region width W index of the improved focus evaluation function of the present embodiment is the smallest compared with other evaluation functions; the sharpness ratio R is several orders of magnitude higher than the other evaluation functions; the steepness Stp is 7 times of the minimum value of other evaluation functions; the sensitivity f sen is 14 times the minimum of the other evaluation functions.
In summary, the present embodiment is an autofocus evaluation algorithm for wafer alignment mark images with strong noise immunity, good stability and timeliness, and suitable for edge direction diversity.
A focus evaluation system for wafer alignment marks, comprising:
The image acquisition module is used for acquiring a sequence image set of the wafer alignment mark in the defocus-focus-defocus process;
The focusing evaluation module is used for extracting edges of each sequence image in the sequence image set; performing focus evaluation on each sequence image in different directions after edge extraction based on a Brenner function and a Roberts function to obtain initial focus evaluation function values corresponding to each sequence image; performing variance calculation on the initial focus evaluation function value corresponding to each sequence image to obtain a final focus evaluation function value; and judging the focusing state of the wafer alignment mark based on the final focusing evaluation function value.
An electronic device includes a memory for storing a computer program and a processor that runs the computer program to cause the electronic device to perform the method of focus evaluation for wafer alignment marks.
A computer readable storage medium storing a computer program which when executed by a processor implements the method of focus evaluation for wafer alignment marks.
The present application is not limited to the above-mentioned embodiments, and any changes or substitutions that can be easily understood by those skilled in the art within the technical scope of the present application are intended to be included in the scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.
Claims (8)
1. The focusing evaluation method for the wafer alignment mark is characterized by comprising the following steps of:
Acquiring a sequence image set of a wafer alignment mark in the defocus-focus-defocus process;
performing edge extraction on each sequence image in the sequence image set;
Performing focus evaluation on each sequence image in different directions after edge extraction based on a Brenner function and a Roberts function to obtain initial focus evaluation function values corresponding to each sequence image;
Performing variance calculation on the initial focus evaluation function value corresponding to each sequence image to obtain a final focus evaluation function value;
and judging the focusing state of the wafer alignment mark based on the final focusing evaluation function value.
2. The method for evaluating focusing on a wafer alignment mark according to claim 1, wherein the edge extraction is performed on each sequence image in the sequence image set, specifically comprising:
Determining an initial sampling interval, and calculating the gray value standard deviation of each sequence image;
calculating gray value difference values of each pixel to be marked and adjacent pixel points in each sequence image, wherein the adjacent pixel points comprise pixel points positioned on the right and below the current pixel point to be marked;
And comparing the gray value standard deviation with the gray value difference of the adjacent pixel points, and determining whether the current pixel point to be marked is an edge pixel point or not based on a comparison result.
3. The method for evaluating focusing on a wafer alignment mark according to claim 12, wherein comparing the gray value standard deviation with the gray value difference between adjacent pixels, specifically comprises:
If the gray value difference values of the adjacent pixel points are smaller than the gray value standard deviation, judging that the pixel point to be marked currently is not an edge pixel point;
If the gray value difference value larger than the gray value standard deviation exists in the gray value difference values of the adjacent pixel points, judging the current pixel point to be marked as an edge pixel point;
and after judging that the current pixel to be marked is an edge pixel, marking the current pixel to be marked and reducing the initial sampling interval.
4. The method for evaluating focusing on a wafer alignment mark according to claim 1, wherein the method for evaluating focusing on the basis of the Brenner function and the Roberts function in different directions of each sequence of images after edge extraction specifically comprises:
Performing evaluation calculation in the directions of 0 DEG and 90 DEG of the sequence images by using a Brenner function, and adding pixel gray difference calculation in the direction of 90 DEG to obtain a first focusing evaluation function value; and (3) carrying out gray gradient calculation on the 45 DEG and 135 DEG directions of the sequence images by using a 3X 3Roberts operator to obtain a second focusing evaluation function value, and multiplying the first focusing evaluation function value and the second focusing evaluation function value to obtain an initial focusing evaluation function value.
5. The focus evaluation method for wafer alignment mark according to claim 1, wherein the variance calculation is performed on the initial focus evaluation function value corresponding to each of the sequence images, and the specific calculation formula is:
Where N is the number of sequential images, U Bre2d_Rob2d is the initial average evaluation function value of the sequential image set, and F Bre2d_Rob2d_Var is the final focus evaluation function value.
6. A focus evaluation system for a wafer alignment mark, comprising:
The image acquisition module is used for acquiring a sequence image set of the wafer alignment mark in the defocus-focus-defocus process;
The focusing evaluation module is used for extracting edges of each sequence image in the sequence image set; performing focus evaluation on each sequence image in different directions after edge extraction based on a Brenner function and a Roberts function to obtain initial focus evaluation function values corresponding to each sequence image; performing variance calculation on the initial focus evaluation function value corresponding to each sequence image to obtain a final focus evaluation function value; and judging the focusing state of the wafer alignment mark based on the final focusing evaluation function value.
7. An electronic device comprising a memory for storing a computer program and a processor that runs the computer program to cause the electronic device to perform a wafer alignment mark oriented focus evaluation method according to any one of claims 1-5.
8. A computer-readable storage medium, characterized in that it stores a computer program which, when executed by a processor, implements a wafer alignment mark oriented focus evaluation method as claimed in any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410990631.6A CN119027371A (en) | 2024-07-23 | 2024-07-23 | A focus evaluation method, system, device and medium for wafer alignment mark |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410990631.6A CN119027371A (en) | 2024-07-23 | 2024-07-23 | A focus evaluation method, system, device and medium for wafer alignment mark |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN119027371A true CN119027371A (en) | 2024-11-26 |
Family
ID=93538181
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410990631.6A Pending CN119027371A (en) | 2024-07-23 | 2024-07-23 | A focus evaluation method, system, device and medium for wafer alignment mark |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119027371A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112433339A (en) * | 2020-12-10 | 2021-03-02 | 济南国科医工科技发展有限公司 | Microscope fine focusing method based on random forest |
| CN114972084A (en) * | 2022-05-13 | 2022-08-30 | 杭州汇萃智能科技有限公司 | Image focusing accuracy evaluation method and system |
| CN117252915A (en) * | 2023-10-08 | 2023-12-19 | 江苏理工学院 | Part image high-precision focusing method and device based on improved gradient weighting |
| CN117452784A (en) * | 2023-10-20 | 2024-01-26 | 江苏影速集成电路装备股份有限公司 | Control method of photoetching alignment system capable of automatically selecting exposure time and object distance |
| CN117788306A (en) * | 2023-12-18 | 2024-03-29 | 上海贝特威自动化科技有限公司 | Multithreading-based multi-focal-length tab image fusion method |
-
2024
- 2024-07-23 CN CN202410990631.6A patent/CN119027371A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112433339A (en) * | 2020-12-10 | 2021-03-02 | 济南国科医工科技发展有限公司 | Microscope fine focusing method based on random forest |
| CN114972084A (en) * | 2022-05-13 | 2022-08-30 | 杭州汇萃智能科技有限公司 | Image focusing accuracy evaluation method and system |
| CN117252915A (en) * | 2023-10-08 | 2023-12-19 | 江苏理工学院 | Part image high-precision focusing method and device based on improved gradient weighting |
| CN117452784A (en) * | 2023-10-20 | 2024-01-26 | 江苏影速集成电路装备股份有限公司 | Control method of photoetching alignment system capable of automatically selecting exposure time and object distance |
| CN117788306A (en) * | 2023-12-18 | 2024-03-29 | 上海贝特威自动化科技有限公司 | Multithreading-based multi-focal-length tab image fusion method |
Non-Patent Citations (2)
| Title |
|---|
| 熊锐 等: "一种适应多方向灰度梯度变化的自动对焦评价函数", 《激光与光电子学进展》, vol. 59, no. 4, 28 February 2022 (2022-02-28), pages 0418001 * |
| 董正琼 等: "基于梯度方差聚焦评价的三维显微测量方法", 《激光与光电子学进展》, 26 December 2023 (2023-12-26), pages 2 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109840889B (en) | High-precision vision measurement method, device and system based on bionic algorithm | |
| TWI324015B (en) | Autofocus searching method | |
| US8139868B2 (en) | Image processing method for determining matching position between template and search image | |
| Zhang et al. | A new focus measure method using moments | |
| CN102968792B (en) | Method for multi-focal-plane object imaging under microscopic vision | |
| CN111083365B (en) | Method and device for rapidly detecting optimal focal plane position | |
| CN102968800B (en) | A kind of evaluation method of image definition | |
| CN112019751B (en) | Calibration information based automatic focusing method | |
| CN111462225A (en) | Centroid identification and positioning method of infrared light spot image | |
| CN104637064A (en) | Defocus blurred image definition detection method based on edge intensity weight | |
| CN107664899A (en) | The method, apparatus and system of auto-focusing | |
| US8759763B2 (en) | Method and apparatus to measure step height of device using scanning electron microscope | |
| JP2009259036A (en) | Image processing device, image processing method, image processing program, recording medium, and image processing system | |
| CN113569679B (en) | Method, device and system for measuring elongation at break | |
| CN103714550B (en) | An automatic optimization method for image registration based on matching curve feature evaluation | |
| CN107092905A (en) | A kind of instrument localization method to be identified of electric inspection process robot | |
| CN106878617B (en) | A kind of focusing method and system | |
| CN106534661A (en) | Automatic focus algorithm accumulated based on strongest edge gradient Laplasse operator | |
| CN105093479A (en) | Automatic focusing method and device used for microscope | |
| CN119027371A (en) | A focus evaluation method, system, device and medium for wafer alignment mark | |
| Lee et al. | Real-time application of critical dimension measurement of TFT-LCD pattern using a newly proposed 2D image-processing algorithm | |
| CN106981065A (en) | A kind of image Absolute Central Moment innovatory algorithm based on exposure compensating | |
| CN108629227A (en) | The method and system on left and right vehicle wheel boundary are determined in the picture | |
| CN107292924B (en) | The characteristic automatic recognition method for the laser groove that a kind of pair of laser processing is formed | |
| CN117351494A (en) | Perspective distortion-based parking space character correction method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |