Semiconductor automatic cleaning monitoring system and method based on image processing

By evaluating image abnormality and evaluating importance of the number of images and surface transistors of semiconductor wafers, the problem of inaccurate judgment of the degree of dirtyness in the prior art is solved, and accurate analysis of cleaning time and improvement of cleaning efficiency are achieved.

CN118571794BActive Publication Date: 2024-11-15JIANGSU XINGGUANG SEMICONDUCTOR CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202410743121.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-06-11
Publication Date
2024-11-15
Estimated Expiration
2044-06-11

AI Technical Summary

Technical Problem

The prior art cannot comprehensively judge the degree of dirtyness of the semiconductor wafer based on the finished image of the semiconductor wafer and the number of surface transistors, resulting in inaccurate analysis of the cleaning time of the cleaning equipment, affecting the cleaning efficiency and device performance.

Method used

By acquiring the image and surface transistors of the semiconductor wafer, dividing the cleaning sub-regions, performing image abnormality evaluation and importance evaluation, and using the evaluation results to determine the degree of dirtyness, thereby automatically determining the cleaning time.

Benefits of technology

It improves the cleaning efficiency, avoids damage to semiconductor wafers during the cleaning process, and achieves a comprehensive and accurate analysis of cleaning time.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118571794B_ABST
    Figure CN118571794B_ABST
Patent Text Reader

Abstract

The present invention discloses a semiconductor automatic cleaning monitoring system and method based on image processing, which belongs to the field of image processing. The present invention obtains the number of surface transistors in a cleaning sub-area and introduces it into a cleaning sub-area importance assessment strategy to perform cleaning sub-area importance assessment, obtains image anomaly assessment results and cleaning sub-area importance assessment results to assess the semiconductor contamination level, and makes an automatic cleaning time judgment based on the semiconductor contamination level. The contamination level of the semiconductor wafer is comprehensively judged based on the finished product image of the semiconductor wafer and the number of surface transistors, and then a comprehensive and accurate analysis of the cleaning time of the cleaning equipment is performed, thereby improving the cleaning efficiency while avoiding damage to the semiconductor wafer caused by cleaning.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the field of image processing, and in particular relates to a semiconductor automatic cleaning monitoring system and method based on image processing. Background Art

[0002] Semiconductor automatic cleaning is a process of cleaning the surface of semiconductor devices to ensure that the surface is clean and free of contaminants, thereby improving the performance and reliability of semiconductor devices. With the continuous advancement of semiconductor manufacturing processes, automatic cleaning technology is also constantly developing, from simple mechanical cleaning to complex chemical cleaning, and then to the current automatic cleaning system based on artificial intelligence. However, the existing technology cannot make a comprehensive judgment on the degree of contamination of semiconductor wafers based on the finished image of the semiconductor wafer and the number of surface transistors during cleaning, and thus cannot make a comprehensive and accurate analysis of the cleaning time of the cleaning equipment. The above problems exist in the existing technology;

[0003] For example, a Chinese patent with the authorization announcement number CN109427035B discloses a semiconductor device and an image recognition system. For example, a semiconductor device includes an image recognition device having a convolution arithmetic processing circuit. The convolution arithmetic processing circuit includes a coefficient register for setting coefficients of an integral coefficient table, a product calculation circuit for calculating the product of an input image and the coefficient, a channel register in which the number of channels of the integral coefficient table is set, a channel selection circuit for selecting an output destination of a cumulative addition arithmetic operation based on the number of channels, and a plurality of output registers for storing the results of the cumulative addition arithmetic operation. The integral coefficient table is a table in which a plurality of input coefficient tables are integrated, and the integral coefficient table has a size of N×N. The product calculation circuit can calculate N×N data at the same time;

[0004] At the same time, a semiconductor defect recognition method based on AI image recognition is disclosed in a Chinese patent with application publication number CN117197021A. According to the original coordinate position of the defect, the primary defect accuracy is calculated, and it is judged whether the primary defect accuracy reaches the primary threshold; if the primary threshold is not reached, the defect offset is found and adjusted; then the secondary defect accuracy is calculated according to the final coordinates of the defect, and it is judged whether the secondary defect accuracy reaches the secondary threshold; if the secondary threshold is reached, the type of defect is further identified, classified, and sorted. Its advantage is that it realizes automation without modifying or upgrading the equipment hardware itself, gets rid of the dependence on personnel, greatly reduces misjudgments and misoperations during operation, improves production efficiency, defect success rate and stability, and directly reduces labor costs;

[0005] The above patents all have the problems raised by this background technology: the prior art is unable to make a comprehensive judgment on the degree of contamination of the semiconductor wafer based on the finished product image of the semiconductor wafer and the number of surface transistors during cleaning, and thus is unable to make a comprehensive and accurate analysis of the cleaning time of the cleaning equipment. In order to solve these problems, this application designs a semiconductor automatic cleaning monitoring system and method based on image processing. Summary of the invention

[0006] In view of the deficiencies in the prior art, the present invention proposes a semiconductor automatic cleaning monitoring system and method based on image processing, which obtains an image of a semiconductor wafer that needs to be cleaned, and simultaneously obtains a finished product image of the semiconductor wafer and the number of surface transistors, divides the cleaning sub-areas, obtains the semiconductor wafer image of the cleaning sub-area and the finished product image of the semiconductor wafer, and imports them into an image anomaly assessment strategy to perform image anomaly assessment, obtains the number of surface transistors in the cleaning sub-area and imports them into a cleaning sub-area importance assessment strategy to perform cleaning sub-area importance assessment, obtains the image anomaly assessment results and the cleaning sub-area importance assessment results to evaluate the degree of semiconductor contamination, makes an automatic cleaning time judgment based on the degree of semiconductor contamination, makes a comprehensive judgment on the degree of contamination of the semiconductor wafer based on the finished product image of the semiconductor wafer and the number of surface transistors, and then makes a comprehensive and accurate analysis of the cleaning time of the cleaning equipment, thereby improving the cleaning efficiency while avoiding damage to the semiconductor wafer caused by cleaning.

[0007] To achieve the above-mentioned object, the present invention provides the following technical solution: a semiconductor automatic cleaning monitoring method based on image processing, which comprises the following specific steps: obtaining an image of a semiconductor wafer that needs to be cleaned, and simultaneously obtaining a finished product image of the semiconductor wafer and the number of surface transistors, and dividing the cleaning sub-areas;

[0008] Obtaining semiconductor wafer images of the cleaning sub-area and finished product images of the semiconductor wafer and importing them into the image anomaly assessment strategy for image anomaly assessment;

[0009] Obtain the number of transistors on the surface of the cleaning sub-area and import it into the cleaning sub-area importance evaluation strategy to perform cleaning sub-area importance evaluation;

[0010] Obtain image anomaly assessment results and cleaning sub-area importance assessment results to assess the degree of semiconductor contamination;

[0011] The automatic cleaning time is determined according to the degree of semiconductor contamination.

[0012] It should be specifically explained that the acquisition of the semiconductor wafer image that needs to be cleaned, the acquisition of the finished product image of the semiconductor wafer and the number of surface transistors, and the division of the cleaning sub-areas include the following specific steps:

[0013] S11, obtaining a semiconductor wafer to be cleaned, and dividing the semiconductor wafer into a plurality of cleaning sub-areas of equal set areas according to the area of ​​the semiconductor wafer. It should be noted that the cleaning sub-areas are preferably 1 square inch squares;

[0014] S12, obtaining surface image data of each cleaning sub-area and simultaneously obtaining finished product image data of the semiconductor wafer, importing the surface image data of the cleaning sub-area and the image data of the cleaning sub-area corresponding to the finished product of the semiconductor wafer into image processing software to obtain the pixel value of each pixel point of each image, and storing them in the first storage component;

[0015] S13, obtaining the number of surface transistors in the cleaning sub-area corresponding to the finished product of the semiconductor wafer, and storing it in the second storage component.

[0016] It should be specifically explained here that the acquisition of the semiconductor wafer image of the cleaning sub-area and the finished product image of the semiconductor wafer and the introduction of the image anomaly assessment strategy to perform image anomaly assessment includes the following specific contents:

[0017] S21, obtaining the pixel value of each pixel point of the semiconductor wafer image in the cleaning sub-region and the pixel value of each pixel point of the image of the cleaning sub-region corresponding to the finished semiconductor wafer;

[0018] S22, obtaining the pixel value of each pixel point of the semiconductor wafer image in the cleaning sub-region and the pixel value of each pixel point of the image of the semiconductor wafer product corresponding to the cleaning sub-region, and importing them into the corresponding pixel point pixel value difference coefficient calculation formula to calculate the corresponding pixel point pixel value difference coefficient, wherein the pth corresponding pixel point pixel value difference coefficient is: ,in, is the pixel value of the pth pixel of the semiconductor wafer image in the cleaning sub-area, The pixel value of the pth pixel point of the cleaning sub-region image corresponding to the finished product of the semiconductor wafer is obtained, the pixel point with the largest pixel value difference coefficient of the corresponding pixel point is set as the abnormal pixel point, and the distance between other pixel points in the cleaning sub-region and the abnormal pixel point is obtained;

[0019] S23, obtaining the distance between other pixels in the cleaning sub-region and the abnormal pixel, and obtaining the pixel value difference coefficient of the corresponding pixel in the cleaning sub-region, and substituting it into the image abnormality evaluation value calculation formula to calculate the image abnormality evaluation value of the cleaning sub-region, wherein the image abnormality evaluation value of the z-th cleaning sub-region is: , where d is the length of the cleaning sub-region, L is the width of the cleaning sub-region, N is the number of pixels in the cleaning sub-region, kzi is the pixel value difference coefficient of the ith corresponding pixel in the zth cleaning sub-region, and Lzi is the distance between the ith other pixel in the zth cleaning sub-region and the abnormal pixel.

[0020] It should be specifically explained here that the method of obtaining the number of surface transistors of the cleaning sub-region and importing it into the cleaning sub-region importance evaluation strategy to evaluate the importance of the cleaning sub-region includes the following specific contents:

[0021] The number of transistors on the surface of the cleaning sub-region is obtained, and the number of transistors on the surface of the cleaning sub-region is introduced into the cleaning sub-region importance calculation formula to calculate the cleaning sub-region importance, wherein the cleaning sub-region importance calculation formula of the z-th cleaning sub-region is: , where mz is the number of surface transistors in the cleaning sub-area, and M is the set threshold value of the number of surface transistors.

[0022] It should be specifically explained here that the acquisition of the image anomaly assessment result and the cleaning sub-region importance assessment result to assess the semiconductor contamination degree includes the following specific steps:

[0023] S41, obtaining the calculated image anomaly assessment value and cleaning sub-region importance of each cleaning sub-region;

[0024] S42, substituting the image abnormality evaluation value of each cleaning sub-region and the importance of the cleaning sub-region into the semiconductor contamination degree calculation formula to calculate the semiconductor contamination degree, wherein the semiconductor contamination degree calculation formula is: , where m is the number of cleaning sub-areas on the semiconductor wafer;

[0025] It should be noted that the determination of the automatic cleaning time according to the degree of semiconductor contamination includes the following specific contents:

[0026] The calculated semiconductor contamination degree is obtained, and the calculated semiconductor contamination degree is introduced into the cleaning time calculation formula to calculate the cleaning time, wherein the cleaning time calculation formula is: , where Tk is the set cleaning standard time, The degree of semiconductor contamination of the semiconductor corresponding to the standard cleaning time.

[0027] The semiconductor automatic cleaning monitoring system based on image processing is implemented based on the above-mentioned semiconductor automatic cleaning monitoring method based on image processing, and specifically includes a data acquisition module, an image anomaly assessment module, an importance assessment module, a contamination degree assessment module, a cleaning time judgment module and a control module, wherein the data acquisition module is used to obtain the image of the semiconductor wafer that needs to be cleaned, and at the same time obtain the finished image of the semiconductor wafer and the number of surface transistors to divide the cleaning sub-areas, and the image anomaly assessment module is used to obtain the semiconductor wafer image of the cleaning sub-area and the finished image of the semiconductor wafer and import it into the image anomaly assessment strategy to perform image anomaly assessment.

[0028] It should be specifically explained here that the importance assessment module is used to obtain the number of surface transistors in the cleaning sub-area and import it into the cleaning sub-area importance assessment strategy to perform cleaning sub-area importance assessment, the contamination level assessment module is used to obtain the image abnormality assessment results and the cleaning sub-area importance assessment results to evaluate the semiconductor contamination level, and the cleaning time judgment module is used to automatically judge the cleaning time according to the semiconductor contamination level.

[0029] It should be specifically explained here that the control module is used to control the operation of the data acquisition module, the image anomaly assessment module, the importance assessment module, the contamination degree assessment module and the cleaning time judgment module.

[0030] An electronic device comprises: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;

[0031] The processor executes the above-mentioned semiconductor automatic cleaning monitoring method based on image processing by calling the computer program stored in the memory.

[0032] A computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the semiconductor automatic cleaning monitoring method based on image processing as described above.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] The present invention obtains an image of a semiconductor wafer that needs to be cleaned, and simultaneously obtains a finished product image of the semiconductor wafer and the number of surface transistors, divides the cleaning sub-areas, obtains the semiconductor wafer image of the cleaning sub-area and the finished product image of the semiconductor wafer, and introduces them into an image abnormality assessment strategy to perform image abnormality assessment, obtains the number of surface transistors in the cleaning sub-area and introduces them into a cleaning sub-area importance assessment strategy to perform cleaning sub-area importance assessment, obtains the image abnormality assessment results and the cleaning sub-area importance assessment results to evaluate the degree of semiconductor contamination, automatically judges the cleaning time according to the degree of semiconductor contamination, comprehensively judges the degree of contamination of the semiconductor wafer according to the finished product image of the semiconductor wafer and the number of surface transistors, and then comprehensively and accurately analyzes the cleaning time of the cleaning equipment, thereby improving the cleaning efficiency and avoiding damage to the semiconductor wafer caused by cleaning. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a schematic diagram of the overall process of the semiconductor automatic cleaning monitoring method based on image processing of the present invention;

[0036] Figure 2 It is a schematic diagram of the overall framework of the semiconductor automatic cleaning monitoring system based on image processing of the present invention;

[0037] Figure 3 It is a schematic diagram of the cleaning sub-area of ​​the present invention. DETAILED DESCRIPTION

[0038] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0039] Example 1

[0040] The technical problem solved by this embodiment is that the prior art cannot comprehensively judge the degree of contamination of the semiconductor wafer according to the finished product image of the semiconductor wafer and the number of surface transistors during cleaning, and thus cannot comprehensively and accurately analyze the cleaning time of the cleaning equipment;

[0041] To solve the above problems, please refer to Figure 1 and Figure 3 The present invention provides a preferred embodiment: Figure 1 As shown, an image of a semiconductor wafer that needs to be cleaned is obtained, and at the same time, a finished product image of the semiconductor wafer and the number of surface transistors are obtained to divide the cleaning sub-areas;

[0042] In this embodiment, the above steps are specifically described, and the specific contents of the above steps include: S11, obtaining a semiconductor wafer that needs to be cleaned, such as Figure 3 As shown, the area of ​​the semiconductor wafer is divided into a number of cleaning sub-areas of equal set area. It should be noted that the cleaning sub-area is preferably a square of 1 square inch;

[0043] S12, obtaining surface image data of each cleaning sub-area and simultaneously obtaining finished product image data of the semiconductor wafer, importing the surface image data of the cleaning sub-area and the image data of the cleaning sub-area corresponding to the finished product of the semiconductor wafer into image processing software to obtain the pixel value of each pixel point of each image, and storing them in the first storage component;

[0044] The surface image data of the cleaning sub-area and the image data of the cleaning sub-area corresponding to the finished semiconductor wafer are imported into the image processing software to obtain the pixel value of each pixel point of each image. The specific operation steps are as follows:

[0045] 1. Import the surface image data of the cleaning sub-area and the image data of the cleaning sub-area corresponding to the finished semiconductor wafer into the image processing software. Usually, the data needs to be converted into an image format, such as JPEG, PNG, etc.;

[0046] 2. Open the imported image data in the image processing software, which usually displays an image window;

[0047] 3. Get the pixel value of each pixel, which can be achieved through the "sampling" or "measurement" function in the software. The specific operation method may vary depending on the software, but it usually displays the coordinates of each pixel and the corresponding color value;

[0048] It should be noted that when obtaining pixel values, it is necessary to ensure the accuracy and integrity of the image data to ensure the accuracy and reliability of the analysis results. In addition, factors such as image resolution and clarity need to be considered to ensure the accuracy and readability of the analysis results.

[0049] S13, obtaining the number of surface transistors in the cleaning sub-area corresponding to the finished product of the semiconductor wafer, and storing it in the second storage component;

[0050] Obtaining semiconductor wafer images of the cleaning sub-area and finished product images of the semiconductor wafer and importing them into the image anomaly assessment strategy for image anomaly assessment;

[0051] In this embodiment, the above steps are specifically described, and the specific contents of the above steps include: S21, obtaining the pixel value of each pixel point of the semiconductor wafer image in the cleaning sub-region and the pixel value of each pixel point of the image of the cleaning sub-region corresponding to the finished semiconductor wafer;

[0052] S22, obtaining the pixel value of each pixel point of the semiconductor wafer image in the cleaning sub-region and the pixel value of each pixel point of the image of the semiconductor wafer product corresponding to the cleaning sub-region, and importing them into the corresponding pixel point pixel value difference coefficient calculation formula to calculate the corresponding pixel point pixel value difference coefficient, wherein the pth corresponding pixel point pixel value difference coefficient is: ,in, is the pixel value of the pth pixel of the semiconductor wafer image in the cleaning sub-area, The pixel value of the pth pixel point of the cleaning sub-region image corresponding to the finished semiconductor wafer is obtained, and the pixel point with the largest pixel value difference coefficient of the corresponding pixel points is set as the abnormal pixel point, and the distance between the other pixel points in the cleaning sub-region and the abnormal pixel point is obtained; Here, an example is given to illustrate the calculation process of kp. The pixel value of the pth pixel point of the semiconductor wafer image in the cleaning sub-region is 122, and the pixel value of the pth pixel point of the cleaning sub-region image corresponding to the finished semiconductor wafer is 125, then kp=3 / 125;

[0053] S23, obtaining the distance between other pixels in the cleaning sub-region and the abnormal pixel, and obtaining the pixel value difference coefficient of the corresponding pixel in the cleaning sub-region, and substituting it into the image abnormality evaluation value calculation formula to calculate the image abnormality evaluation value of the cleaning sub-region, wherein the image abnormality evaluation value of the z-th cleaning sub-region is: , where d is the length of the cleaning sub-region, L is the width of the cleaning sub-region, N is the number of pixels in the cleaning sub-region, kzi is the pixel value difference coefficient of the ith corresponding pixel in the zth cleaning sub-region, and Lzi is the distance between the ith other pixel in the zth cleaning sub-region and the abnormal pixel; kzi here is calculated based on the pixel value difference coefficient of the corresponding pixel in step S22, and Lzi here is the distance between the ith other pixel and the abnormal pixel. Obtaining the distance between two pixels is a common conventional technical means, which can be implemented by most image processing software;

[0054] Here, the above steps are explained by a specific embodiment: In a specific embodiment, the cleaning sub-area is preferably a square of 1 square inch, for example, the length of the cleaning sub-area is 2 centimeters, the width is 3 centimeters, d+L=5, the image of the cleaning sub-area is imported into the image processing software to obtain the pixel value of the pixel point, which is obtained in detail in step S12, and the pixel value of the corresponding pixel point and the distance between the i-th other pixel point and the abnormal pixel point are obtained at the same time, and imported into the image abnormality evaluation value calculation formula to calculate the image abnormality evaluation value of the cleaning sub-area;

[0055] Obtain the number of transistors on the surface of the cleaning sub-area and import it into the cleaning sub-area importance evaluation strategy to perform cleaning sub-area importance evaluation;

[0056] In this embodiment, the above steps are specifically described. The specific contents of the above steps include: the step of obtaining the number of surface transistors in the cleaning sub-region and importing it into the cleaning sub-region importance evaluation strategy to evaluate the importance of the cleaning sub-region includes the following specific contents:

[0057] The number of transistors on the surface of the cleaning sub-region is obtained, and the number of transistors on the surface of the cleaning sub-region is introduced into the cleaning sub-region importance calculation formula to calculate the cleaning sub-region importance, wherein the cleaning sub-region importance calculation formula of the z-th cleaning sub-region is: , where mz is the number of surface transistors in the cleaning sub-region, and M is the set threshold value of the number of surface transistors; the number of surface transistors in the cleaning sub-region here has been defined during wafer design and can be directly obtained from the wafer design library;

[0058] Here, the above steps are explained by a specific embodiment: In a specific embodiment, the number of surface transistors in the cleaning sub-area is 1000, the threshold value of the number of surface transistors is set to 2000, and Pz=0.5;

[0059] Obtain image anomaly assessment results and cleaning sub-area importance assessment results to assess the degree of semiconductor contamination;

[0060] The above steps are specifically described in this embodiment, and the specific contents of the above steps include:

[0061] S41, obtaining the calculated image anomaly assessment value and cleaning sub-region importance of each cleaning sub-region;

[0062] S42, substituting the image abnormality evaluation value of each cleaning sub-region and the importance of the cleaning sub-region into the semiconductor contamination degree calculation formula to calculate the semiconductor contamination degree, wherein the semiconductor contamination degree calculation formula is: , where m is the number of cleaning sub-areas on the semiconductor wafer;

[0063] Here, the above steps are explained by a specific embodiment: In a specific embodiment, there are 10 cleaning sub-areas on the wafer, and the image abnormality evaluation value and the importance of the cleaning sub-area of ​​each cleaning sub-area are respectively substituted into the image abnormality evaluation value calculation formula and the importance of the cleaning sub-area for calculation, and then substituted into the semiconductor contamination degree calculation formula to calculate the semiconductor contamination degree;

[0064] Determine the automatic cleaning time according to the degree of semiconductor contamination;

[0065] In this embodiment, the above steps are specifically described. The specific contents of the above steps include: obtaining the calculated semiconductor contamination degree, and importing the calculated semiconductor contamination degree into the cleaning time calculation formula to calculate the cleaning time, wherein the cleaning time calculation formula is: , where Tk is the set cleaning standard time, It is the semiconductor contamination degree of the semiconductor corresponding to the standard cleaning time. It should be noted here that the semiconductor contamination degree of the semiconductor corresponding to the standard cleaning time and the set threshold value of the number of surface transistors are obtained in the following way: 5000 sets of semiconductor wafer images that need to be cleaned are obtained, and the finished product images of the semiconductor wafer and the number of surface transistors are obtained at the same time, and the cleaning time is imported into the cleaning time calculation formula to calculate the cleaning time. At the same time, the semiconductor wafer is placed in the semiconductor wafer cleaning module for cleaning, and the real-time cleaning time for cleaning that meets the semiconductor contamination degree of the semiconductor used is obtained. The calculated cleaning time and real-time cleaning time are imported into the fitting software, and the semiconductor contamination degree of the semiconductor corresponding to the cleaning standard time that meets the highest real-time cleaning time accuracy and the set threshold value of the number of surface transistors are output.

[0066] It should be noted in the present embodiment that the advantages of the present embodiment over the prior art are as follows: obtaining an image of a semiconductor wafer that needs to be cleaned, obtaining a finished product image of the semiconductor wafer and the number of surface transistors at the same time, dividing the cleaning sub-areas, obtaining the semiconductor wafer image of the cleaning sub-areas and the finished product image of the semiconductor wafer and importing them into the image abnormality assessment strategy to perform image abnormality assessment, obtaining the number of surface transistors in the cleaning sub-area and importing them into the cleaning sub-area importance assessment strategy to perform cleaning sub-area importance assessment, obtaining the image abnormality assessment results and the cleaning sub-area importance assessment results to evaluate the degree of semiconductor contamination, automatically judging the cleaning time according to the degree of semiconductor contamination, making a comprehensive judgment on the degree of contamination of the semiconductor wafer according to the finished product image of the semiconductor wafer and the number of surface transistors, and then conducting a comprehensive and accurate analysis of the cleaning time of the cleaning equipment, thereby improving the cleaning efficiency while avoiding damage to the semiconductor wafer caused by cleaning.

[0067] Example 2

[0068] like Figure 3 As shown, a semiconductor automatic cleaning monitoring system based on image processing is implemented based on the above-mentioned semiconductor automatic cleaning monitoring method based on image processing, and specifically includes a data acquisition module, an image anomaly assessment module, an importance assessment module, a contamination degree assessment module, a cleaning time judgment module and a control module, wherein the data acquisition module is used to acquire the semiconductor wafer image that needs to be cleaned, and at the same time acquire the finished product image of the semiconductor wafer and the number of surface transistors to divide the cleaning sub-areas, the image anomaly assessment module is used to acquire the semiconductor wafer image of the cleaning sub-area and the finished product image of the semiconductor wafer and import them into the image anomaly assessment strategy to perform image anomaly assessment, the importance assessment module is used to acquire the number of surface transistors in the cleaning sub-area and import them into the cleaning sub-area importance assessment strategy to perform cleaning sub-area importance assessment, the contamination degree assessment module is used to acquire the image anomaly assessment result and the cleaning sub-area importance assessment result to evaluate the semiconductor contamination degree, the cleaning time judgment module is used to judge the automatic cleaning time according to the semiconductor contamination degree, and the control module is used to control the operation of the data acquisition module, the image anomaly assessment module, the importance assessment module, the contamination degree assessment module and the cleaning time judgment module;

[0069] The image anomaly assessment module obtains the semiconductor wafer image of the cleaning sub-area and the finished product image of the semiconductor wafer and imports them into the image anomaly assessment strategy to perform image anomaly assessment;

[0070] In this embodiment, the above steps are specifically described, and the specific contents of the above steps include: S21, obtaining the pixel value of each pixel point of the semiconductor wafer image in the cleaning sub-region and the pixel value of each pixel point of the image of the cleaning sub-region corresponding to the finished semiconductor wafer;

[0071] S22, obtaining the pixel value of each pixel point of the semiconductor wafer image in the cleaning sub-region and the pixel value of each pixel point of the image of the semiconductor wafer product corresponding to the cleaning sub-region, and importing them into the corresponding pixel point pixel value difference coefficient calculation formula to calculate the corresponding pixel point pixel value difference coefficient, wherein the pth corresponding pixel point pixel value difference coefficient is: ,in, is the pixel value of the pth pixel of the semiconductor wafer image in the cleaning sub-area, The pixel value of the pth pixel point of the cleaning sub-region image corresponding to the finished semiconductor wafer is obtained, and the pixel point with the largest pixel value difference coefficient of the corresponding pixel points is set as the abnormal pixel point, and the distance between the other pixel points in the cleaning sub-region and the abnormal pixel point is obtained; Here, an example is given to illustrate the calculation process of kp. The pixel value of the pth pixel point of the semiconductor wafer image in the cleaning sub-region is 122, and the pixel value of the pth pixel point of the cleaning sub-region image corresponding to the finished semiconductor wafer is 125, then kp=3 / 125;

[0072] S23, obtaining the distance between other pixels in the cleaning sub-region and the abnormal pixel, and obtaining the pixel value difference coefficient of the corresponding pixel in the cleaning sub-region, and substituting it into the image abnormality evaluation value calculation formula to calculate the image abnormality evaluation value of the cleaning sub-region, wherein the image abnormality evaluation value of the z-th cleaning sub-region is: , where d is the length of the cleaning sub-region, L is the width of the cleaning sub-region, N is the number of pixels in the cleaning sub-region, kzi is the pixel value difference coefficient of the ith corresponding pixel in the zth cleaning sub-region, and Lzi is the distance between the ith other pixel in the zth cleaning sub-region and the abnormal pixel; kzi here is calculated based on the pixel value difference coefficient of the corresponding pixel in step S22, and Lzi here is the distance between the ith other pixel and the abnormal pixel. Obtaining the distance between two pixels is a common conventional technical means, which can be implemented by most image processing software;

[0073] The importance evaluation module obtains the number of transistors on the surface of the cleaning sub-area and imports it into the importance evaluation strategy of the cleaning sub-area to perform importance evaluation of the cleaning sub-area;

[0074] In this embodiment, the above steps are specifically described. The specific contents of the above steps include: the step of obtaining the number of surface transistors in the cleaning sub-region and importing it into the cleaning sub-region importance evaluation strategy to evaluate the importance of the cleaning sub-region includes the following specific contents:

[0075] The number of transistors on the surface of the cleaning sub-region is obtained, and the number of transistors on the surface of the cleaning sub-region is introduced into the cleaning sub-region importance calculation formula to calculate the cleaning sub-region importance, wherein the cleaning sub-region importance calculation formula of the z-th cleaning sub-region is: , where mz is the number of surface transistors in the cleaning sub-area, and M is the set threshold value of the number of surface transistors;

[0076] The control of the control module is achieved through control codes.

[0077] Example 3

[0078] This embodiment provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;

[0079] The processor executes the above-mentioned semiconductor automatic cleaning monitoring method based on image processing by calling the computer program stored in the memory.

[0080] The electronic device may have relatively large differences due to different configurations or performances, and may include one or more processors (Central Processing Units, CPU) and one or more memories, wherein the memory stores at least one computer program, which is loaded and executed by the processor to implement the semiconductor automatic cleaning monitoring method based on image processing provided by the above method embodiment. The electronic device may also include other components for realizing the functions of the device, for example, the electronic device may also have components such as a wired or wireless network interface and an input and output interface to input and output data. This embodiment will not be described in detail here.

[0081] Example 4

[0082] This embodiment provides a computer-readable storage medium having a rewritable computer program stored thereon;

[0083] When the computer program is executed on a computer device, the computer device is enabled to execute the above-mentioned semiconductor automatic cleaning monitoring method based on image processing.

[0084] For example, the computer readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a compact disc (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, etc.

[0085] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only one way of division, and there may be other ways of division in actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0086] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A semiconductor automatic cleaning monitoring method based on image processing, characterized in that: It includes the following specific steps: Acquire an image of a semiconductor wafer that needs to be cleaned, and simultaneously acquire a finished product image of the semiconductor wafer and the number of surface transistors, and divide the cleaning sub-areas; Obtaining the semiconductor wafer image of the cleaning sub-area and the finished product image of the semiconductor wafer and importing them into the image anomaly assessment strategy for image anomaly assessment; Obtain the number of transistors on the surface of the cleaning sub-area and import it into the cleaning sub-area importance evaluation strategy to perform cleaning sub-area importance evaluation; Obtain image anomaly assessment results and cleaning sub-area importance assessment results to assess the degree of semiconductor contamination; Determine the automatic cleaning time according to the degree of semiconductor contamination; The step of obtaining an image of a semiconductor wafer that needs to be cleaned, obtaining a finished product image of the semiconductor wafer and the number of surface transistors, and dividing the cleaning sub-areas comprises the following specific steps: S11, obtaining a semiconductor wafer that needs to be cleaned, and dividing the semiconductor wafer into a plurality of cleaning sub-areas of equal set areas according to the area of ​​the semiconductor wafer; S12, obtaining surface image data of each cleaning sub-area and simultaneously obtaining finished product image data of the semiconductor wafer, importing the surface image data of the cleaning sub-area and the image data of the cleaning sub-area corresponding to the finished product of the semiconductor wafer into image processing software to obtain the pixel value of each pixel point of each image, and storing them in the first storage component; S13, obtaining the number of surface transistors in the cleaning sub-area corresponding to the finished semiconductor wafer, and storing it in the second storage component; the obtaining of the semiconductor wafer image in the cleaning sub-area and the finished semiconductor wafer image and importing them into the image anomaly assessment strategy for image anomaly assessment includes the following specific contents: S21, obtaining the pixel value of each pixel point of the semiconductor wafer image in the cleaning sub-region and the pixel value of each pixel point of the image of the cleaning sub-region corresponding to the finished semiconductor wafer; S22, obtaining the pixel value of each pixel point of the semiconductor wafer image in the cleaning sub-region and the pixel value of each pixel point of the image of the semiconductor wafer product corresponding to the cleaning sub-region, and importing them into the corresponding pixel point pixel value difference coefficient calculation formula to calculate the corresponding pixel point pixel value difference coefficient, wherein the pth corresponding pixel point pixel value difference coefficient is: ,in, is the pixel value of the pth pixel of the semiconductor wafer image in the cleaning sub-area, The pixel value of the pth pixel point of the cleaning sub-region image corresponding to the finished product of the semiconductor wafer is obtained, the pixel point with the largest pixel value difference coefficient of the corresponding pixel point is set as the abnormal pixel point, and the distance between other pixel points in the cleaning sub-region and the abnormal pixel point is obtained; S23, obtaining the distance between other pixels in the cleaning sub-region and the abnormal pixel, and obtaining the pixel value difference coefficient of the corresponding pixel in the cleaning sub-region, and substituting it into the image abnormality evaluation value calculation formula to calculate the image abnormality evaluation value of the cleaning sub-region, wherein the image abnormality evaluation value of the z-th cleaning sub-region is: , where d is the length of the cleaning sub-region, L is the width of the cleaning sub-region, N is the number of pixels in the cleaning sub-region, kzi is the pixel value difference coefficient of the ith corresponding pixel in the zth cleaning sub-region, and Lzi is the distance between the ith other pixel in the zth cleaning sub-region and the abnormal pixel; the method of obtaining the number of surface transistors of the cleaning sub-region and importing it into the cleaning sub-region importance evaluation strategy to evaluate the importance of the cleaning sub-region includes the following specific contents: The number of transistors on the surface of the cleaning sub-region is obtained, and the number of transistors on the surface of the cleaning sub-region is introduced into the cleaning sub-region importance calculation formula to calculate the cleaning sub-region importance, wherein the cleaning sub-region importance calculation formula of the z-th cleaning sub-region is: , wherein mz is the number of surface transistors in the cleaning sub-region, and M is the set threshold value of the number of surface transistors; the acquisition of image abnormality assessment results and cleaning sub-region importance assessment results to assess the degree of semiconductor contamination includes the following specific steps: S41, obtaining the calculated image anomaly assessment value and cleaning sub-region importance of each cleaning sub-region; S42, substituting the image abnormality evaluation value of each cleaning sub-region and the importance of the cleaning sub-region into the semiconductor contamination degree calculation formula to calculate the semiconductor contamination degree, wherein the semiconductor contamination degree calculation formula is: , where m is the number of cleaning sub-areas on the semiconductor wafer.

2. The semiconductor automatic cleaning monitoring method based on image processing according to claim 1, characterized in that: The determination of the automatic cleaning time according to the degree of semiconductor contamination includes the following specific contents: The calculated semiconductor contamination degree is obtained, and the calculated semiconductor contamination degree is introduced into the cleaning time calculation formula to calculate the cleaning time, wherein the cleaning time calculation formula is: , where Tk is the set cleaning standard time, The degree of semiconductor contamination of the semiconductor corresponding to the standard cleaning time.

3. A semiconductor automatic cleaning monitoring system based on image processing, which is implemented based on the semiconductor automatic cleaning monitoring method based on image processing according to any one of claims 1-2, characterized in that: It specifically includes a data acquisition module, an image anomaly assessment module, an importance assessment module, a contamination degree assessment module, a cleaning time judgment module and a control module, wherein the data acquisition module is used to obtain the semiconductor wafer image that needs to be cleaned, and at the same time obtain the finished product image of the semiconductor wafer and the number of surface transistors to divide the cleaning sub-areas, and the image anomaly assessment module is used to obtain the semiconductor wafer image of the cleaning sub-area and the finished product image of the semiconductor wafer and import them into the image anomaly assessment strategy to perform image anomaly assessment.

4. The semiconductor automatic cleaning monitoring system based on image processing as claimed in claim 3, characterized in that: The importance assessment module is used to obtain the number of surface transistors in the cleaning sub-area and import it into the cleaning sub-area importance assessment strategy to perform cleaning sub-area importance assessment. The contamination level assessment module is used to obtain the image abnormality assessment result and the cleaning sub-area importance assessment result to evaluate the semiconductor contamination level. The cleaning time judgment module is used to automatically judge the cleaning time according to the semiconductor contamination level. The control module is used to control the operation of the data acquisition module, the image abnormality assessment module, the importance assessment module, the contamination level assessment module and the cleaning time judgment module.

5. An electronic device comprising: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; It is characterized in that the processor executes the semiconductor automatic cleaning monitoring method based on image processing as described in any one of claims 1-2 by calling the computer program stored in the memory.

6. A computer-readable storage medium, characterized in that: Instructions are stored, and when the instructions are executed on a computer, the computer executes the semiconductor automatic cleaning monitoring method based on image processing as described in any one of claims 1-2.

Citation Information

Cited By

  • System for automatic analysis of sem images based on full database and process defect detection

    CN120707514B