Sapphire crystal bar polishing abnormal data analysis method based on image processing

Abstract

The invention relates to the technical field of data analysis, in particular to a sapphire crystal bar polishing abnormal data analysis method based on image processing, which comprises the following steps: acquiring crystal bar cutting data and a wafer surface image of sapphire; according to the crystal bar cutting data, the incidence relation between the wafer and the crystal bar is judged, and the wafer source similarity is analyzed; according to the wafer surface image, polishing anomaly data are recognized, and the polishing anomaly similarity is analyzed; and judging the source of the polishing abnormity by combining the wafer source similarity and the polishing abnormity similarity. By analyzing the wafer source similarity and the polishing anomaly similarity, the polishing anomaly source is identified, and the accuracy of identifying the polishing anomaly source of the crystal bar is improved.

Description

Background Art

[0002] Sapphire is a multifunctional crystal material with excellent properties such as high hardness, high melting point, good light transmittance, excellent thermal conductivity and electrical insulation, and stable chemical properties. It is widely used in semiconductor lighting, national defense and military industry, aerospace, smart wearables, electronic products and other fields.

[0003] The processing technology of sapphire wafers includes slicing, grinding, polishing and other processes. Sapphire crystal rods are cut into sapphire wafers, which are then ground and polished. Polishing is the last step in sapphire finishing, which directly determines the performance and service life of sapphire wafers.

[0004] Sapphire is a typical hard and brittle material that is difficult to process. Some sapphire wafers still have surface defects after grinding and polishing. The main reasons are: (1) Crystal defects caused by factors such as temperature and pulling speed during the growth of sapphire crystal rods. For example, the gas released by the decomposition of sapphire melt under high temperature conditions or the impurities contained in the original material during the heating process, the gas generated by the interaction between substances leads to bubble defects inside the crystal rod; (2) Surface defects caused by sapphire wafers during the processing processes such as slicing, grinding and polishing. For example, large particles of polishing powder or other impurities in the polishing liquid lead to scratch defects on the surface of sapphire wafers.

[0005] For chip anomalies caused by crystal rod defects, the abnormal characteristics are usually reflected in multiple adjacent cut chips of the same crystal rod, and the abnormal positions of multiple adjacent cut chips are similar. However, the existing sapphire crystal rod polishing anomaly data analysis method fails to identify the source of polishing anomalies by analyzing the spatial relationship between chips and the similar patterns of chip anomalies, resulting in insufficient accuracy in identifying the source of polishing anomalies. Summary of the invention

[0006] In order to overcome the defects and shortcomings of the prior art, the present application provides a sapphire crystal rod polishing anomaly data analysis method based on image processing, which identifies the source of polishing anomaly by analyzing the similarity of wafer sources and polishing anomaly similarity, thereby improving the accuracy of identifying the source of crystal rod polishing anomaly.

[0007] In order to achieve the above objectives, this application adopts the following technical solutions: In a first aspect, the present application provides a sapphire crystal rod polishing abnormality data analysis method based on image processing, comprising the following steps: Collect sapphire ingot cutting data and wafer surface images; Determine the correlation between the wafer and the ingot according to the ingot cutting data and analyze the similarity of the wafer sources; identifying polishing anomaly data based on the wafer surface image and analyzing polishing anomaly similarities; The source of the polishing anomaly is determined by combining the wafer source similarity and the polishing anomaly similarity.

[0008] Optionally, the step of analyzing the similarity of wafer sources includes: Acquire the crystal ingot cutting data, wherein the crystal ingot cutting data includes cutting wafer information and cutting sequence number information corresponding to the wafer; For any two chips and Chip , judging whether the two belong to the same crystal rod by the information of the cut wafer, when the wafer and Chip If they do not belong to the same crystal rod, the chip and Chip Similarity of chip sources ; When the chip and Chip When the wafers belong to the same crystal rod, the wafer source similarity is calculated according to the cutting sequence number information, and the wafer source similarity is used to quantitatively analyze the wafer source similarity.

[0009] Optionally, the calculation formula of the wafer source similarity is as follows: ; in, Indicates chip Cutting sequence number information, Indicates chip Cutting sequence number information, Indicates chip and Chip The number of cut wafers from the respective ingot, Indicates chip and Chip The similarity of the chip sources.

[0010] Optionally, the step of identifying polishing abnormal data includes: Acquiring the wafer surface image and performing preprocessing, wherein the preprocessing includes image normalization, image denoising and image enhancement; Performing feature extraction on the preprocessed wafer surface image by a feature extraction algorithm, wherein the feature extraction includes color feature extraction, texture feature extraction and edge feature extraction; An abnormal data classification model is constructed based on a machine learning algorithm, and the abnormal data classification model is trained through feature extraction results to obtain the polished abnormal data.

[0011] Optionally, the step of analyzing the similarity of polishing anomalies comprises: Acquire the polishing abnormality data and perform binarization processing on the wafer surface image according to the polishing abnormality data to obtain a wafer surface binarization image, wherein the pixel value of a normal pixel in the wafer surface binarization image is 0, and the pixel value of an abnormal pixel in the wafer surface binarization image is 1; For any two chips and Chip , compare the pixel values ​​of pixels at the same position in the binary image of the wafer surface, and calculate the polishing anomaly similarity by counting the number of abnormal pixels with matching positions and the total number of abnormal pixels. The calculation formula of the polishing anomaly similarity is as follows: ; in, Indicates chip The pixel coordinates in the binary image of the wafer surface The pixel value at Indicates chip The pixel coordinates in the binary image of the wafer surface The pixel value at Indicates chip and Chip The pixel coordinate set of Indicates chip and Chip The polishing anomaly similarity is used to quantitatively analyze the similarity of wafer polishing anomalies.

[0012] Optionally, the step of determining the source of the polishing abnormality includes: Obtaining the wafer source similarity and the polishing anomaly similarity; When the sum of the wafer source similarity and the polishing anomaly similarity is greater than or equal to a preset similarity threshold, it is determined that the source of the polishing anomaly is a crystal rod defect; When the sum of the wafer source similarity and the polishing anomaly similarity is less than the preset similarity threshold, it is determined that the source of the polishing anomaly is an amorphous rod defect.

[0013] It should be noted here that the preset similarity threshold is determined as follows: 5000 sets of crystal rod cutting data and chip surface images are collected to distinguish whether the source of the polishing anomaly is a crystal rod defect, the crystal rod cutting data and the chip surface image are substituted into the chip source similarity and polishing anomaly similarity calculation formulas for calculation, the calculated chip source similarity and polishing anomaly similarity and the distinction result are simultaneously imported into the fitting software, and the optimal preset similarity threshold that meets the distinction accuracy of the distinction result is output.

[0014] In a second aspect, the present application provides a sapphire crystal rod polishing abnormality data analysis system based on image processing, comprising: A data acquisition module, used to collect sapphire crystal rod cutting data and wafer surface images; A wafer source analysis module, used to determine the correlation between wafers and crystal bars according to the crystal bar cutting data and analyze the similarity of wafer sources; A polishing anomaly analysis module, used to identify polishing anomaly data and analyze polishing anomaly similarities based on the wafer surface image; The abnormality source identification module is used to determine the source of the polishing abnormality by combining the wafer source similarity and the polishing abnormality similarity.

[0015] In a third aspect, the present application provides an electronic device, comprising: a processor and a memory, wherein the memory stores a computer program that can be called by the processor, and the processor executes a sapphire crystal rod polishing abnormality data analysis method based on image processing by calling the computer program stored in the memory.

[0016] In a fourth aspect, the present application provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute a sapphire crystal rod polishing abnormality data analysis method based on image processing.

[0017] Compared with the prior art, this application has the following advantages and beneficial effects: The present application determines the correlation between the chip and the crystal rod through the crystal rod cutting data and quantitatively analyzes the similarity of the chip sources, identifies the polishing anomaly data through the chip surface image and quantitatively analyzes the similarity of the polishing anomaly, and then identifies the source of the polishing anomaly by combining the similarity of the chip sources and the similarity of the polishing anomaly, thereby effectively improving the accuracy and reliability of the identification of the source of the polishing anomaly. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings: Figure 1 It is a schematic diagram of the overall process of the sapphire crystal rod polishing abnormality data analysis method based on image processing provided in an embodiment of the present application; Figure 2 It is a schematic diagram of the sapphire processing process provided by the embodiment of the present application; Figure 3 It is a structural schematic diagram of a sapphire crystal rod polishing abnormality data analysis system based on image processing provided in an embodiment of the present application; Figure 4 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] The technical solution of the present application is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific features in the embodiments are detailed descriptions of the technical solution of the present application, rather than limitations on the technical solution of the present application. In the absence of conflict, the embodiments of the present application and the technical features in the embodiments can be combined with each other.

[0020] See also Figure 1 , Figure 1 : This is a schematic diagram of the overall process of the sapphire crystal rod polishing abnormality data analysis method based on image processing provided in an embodiment of the present application, which specifically includes the following steps: S110: Collecting sapphire crystal rod cutting data and wafer surface images.

[0021] S120: judging the correlation between the wafer and the ingot according to the ingot cutting data and analyzing the similarity of the wafer sources; See also Figure 2 , Figure 2 : is a schematic diagram of the sapphire processing process provided by an embodiment of the present application. Sapphire crystals are obtained through processes such as crystal growth, rod cutting, and rolling to obtain sapphire crystal rods. Sapphire crystal rods are sliced, ground, and polished to obtain sapphire wafers. The source similarity between different wafers is evaluated by confirming whether the wafers are derived from the same crystal rod and the proximity of the slicing positions on the same crystal rod. When different wafers are highly similar in source and the locations where their polishing anomalies occur also show consistency, it indicates that the polishing anomaly is caused by quality problems of the crystal rod itself. When the wafer source similarity is low, even if the polishing anomaly occurs, it indicates that the anomaly is generated during the wafer processing rather than an internal defect of the crystal rod. Therefore, analyzing the wafer source similarity helps to distinguish whether the source of the polishing anomaly is an internal defect of the crystal rod, thereby improving the accuracy of identifying the source of the polishing anomaly. The steps of analyzing the wafer source similarity include: Acquire crystal ingot cutting data, where the crystal ingot cutting data includes cutting wafer information and cutting sequence number information corresponding to the wafer, wherein the cutting wafer information is information of all wafers cut from the crystal ingot; For any two chips and Chip , by cutting the wafer information to determine whether the two belong to the same crystal rod, when the wafer and Chip If they do not belong to the same crystal rod, the chip and Chip Similarity of chip sources ; When the chip and Chip If they belong to the same crystal rod, the wafer source similarity is calculated by the cutting sequence number information, and the wafer source similarity is used to quantitatively analyze the wafer source similarity; After it is determined that the wafers belong to the same crystal rod, the relative position relationship of the wafers in the crystal rod can be reflected by determining the wafer cutting sequence interval. Since there may be local defects inside the crystal rod, and the local defects will be reflected in the wafers at adjacent or close positions, the smaller the cutting sequence interval is, the closer the physical positions between the wafers are, and the higher the source similarity is, indicating that similar defects may exist between the wafers. The wafer source similarity assessment can identify and analyze the potential correlation between the wafers, and further determine whether the defects come from the same crystal rod. The calculation formula for wafer source similarity is as follows: ; in, Indicates chip Cutting sequence number information, Indicates chip Cutting sequence number information, Indicates chip and Chip The similarity of wafer sources decreases linearly with the increase of the cutting sequence interval. Indicates chip and Chip The number of cut wafers from the respective ingot, Indicates the maximum cutting interval, that is, the cutting interval between the first and last wafers cut from the same crystal rod. Indicates chip and Chip The similarity of the chip sources.

[0022] S130: identifying polishing abnormality data according to the wafer surface image and analyzing the polishing abnormality similarity; Image processing technology can fully capture defects such as scratches, bubbles, and cracks in the wafer surface image, which not only helps to improve the defect detection speed, but also avoids the subjective errors caused by manual detection. Compared with traditional manual detection methods, image recognition technology can accurately identify tiny abnormal defects on the wafer surface. At the same time, combined with machine learning algorithms, it can continuously improve the wafer defect recognition ability. The steps to identify polishing abnormal data include: Acquire the wafer surface image and perform preprocessing, which includes image normalization, image denoising and image enhancement. Image normalization is a commonly used preprocessing step in the field of machine learning and computer vision. Image data is standardized through image normalization so that the data has a uniform scale. Image normalization includes Min-Max normalization, z-score normalization and L2 norm normalization; Image denoising refers to removing unnecessary noise information from the chip surface image. Noise information interferes with the extraction of feature information. Filtering operations are used to filter out noise in the image. The main filtering methods are divided into frequency domain filtering and spatial domain filtering. Among them, frequency filtering removes high-frequency or low-frequency components by filtering the frequency points of the image to smooth and sharpen the image. High-pass filtering can sharpen the image and enhance image detail information. Low-pass filtering is the opposite of high-frequency filtering. Low-pass filtering achieves the purpose of denoising and smoothing the image by filtering high-frequency components. Spatial domain filtering selectively suppresses or separates specific spatial frequency ranges according to the spatial changes of the digital quantity of the image to modify the image to achieve the purpose of denoising. Spatial domain filtering includes mean filtering, matched filtering, median filtering and Wiener filtering. Image enhancement improves image quality to make image features easier to perceive and identify. Image enhancement includes histogram equalization, gamma correction and super-resolution reconstruction. Histogram equalization is a traditional image enhancement method, which is mainly used to improve the contrast of the image. By stretching or evenly distributing the grayscale histogram of the image, the brightness distribution of the image is more balanced. Gamma correction is a nonlinear transformation commonly used in image enhancement, which can adjust the brightness and contrast of the image. By performing a power transformation on the pixel values ​​of the image, it can effectively amplify the details of the dark or bright parts. Super-resolution reconstruction aims to convert low-resolution images into high-resolution images, and uses generative adversarial networks (GANs) or deep convolutional networks for efficient reconstruction. The pre-processed wafer surface image is subjected to feature extraction by a feature extraction algorithm, and the feature extraction includes color feature extraction, texture feature extraction and edge feature extraction, wherein the color feature is extracted by a color moment or a color histogram, the texture feature is extracted by a gray level co-occurrence matrix (GLCM), a wavelet transform or a Gabor filter, and the edge feature is extracted by a Canny edge detection operator or a Sobel edge detection operator; Build an abnormal data classification model based on machine learning algorithms, train the abnormal data classification model through feature extraction results and obtain polished abnormal data. The machine learning algorithms include support vector machine (SVM), K nearest neighbor algorithm (KNN) and neural network. When surface anomalies of multiple wafers appear at the same position and have a high degree of similarity, combined with the source similarity of the wafers, that is, whether the wafers belong to the same crystal rod and the wafer cutting interval in the same crystal rod, it is possible to effectively distinguish whether the source of the anomaly is an internal defect of the crystal rod. The steps for analyzing the similarity of polishing anomalies include: Obtaining polishing abnormality data and performing binarization processing on the wafer surface image according to the polishing abnormality data to obtain a wafer surface binarization image, wherein the pixel value of a normal pixel in the wafer surface binarization image is 0, and the pixel value of an abnormal pixel in the wafer surface binarization image is 1; For any two chips and Chip , compare the pixel values ​​of the pixels at the same position in the binary image of the wafer surface, and calculate the polishing anomaly similarity by counting the number of abnormal pixels with matching positions and the total number of abnormal pixels. The calculation formula for the polishing anomaly similarity is as follows: ; in, Indicates chip The pixel coordinates in the binary image of the wafer surface The pixel value at Indicates chip The pixel coordinates in the binary image of the wafer surface The pixel value at Indicates chip and Chip The pixel coordinates of the set, the molecular part For counting chips and Chip The number of abnormal pixel matches at the same position in , only when and hour, That is, for two pixels with the same pixel coordinates in two chips, only when both pixels are abnormal pixels at the same time will they be included in the statistical range. Indicates chip The number of abnormal pixels in the binary image of the wafer surface, Indicates chip The number of abnormal pixels in the binary image of the wafer surface, Indicates chip and Chip The polishing anomaly similarity is used to quantitatively analyze the similarity of wafer polishing anomalies.

[0023] S140: Determine the source of the polishing anomaly by combining the similarity of wafer sources and the similarity of polishing anomalies; By quantifying the similarity of wafer sources and polishing anomalies, it is possible to effectively distinguish between ingot defects and processing defects, thereby providing a clear direction for subsequent quality tracing and improvement. The steps to determine the source of polishing anomalies include: Obtain wafer source similarity and polishing anomaly similarity; When the sum of the wafer source similarity and the polishing anomaly similarity is greater than or equal to the preset similarity threshold, it is determined that the source of the polishing anomaly is a crystal rod defect; When the sum of the chip source similarity and the polishing anomaly similarity is less than the preset similarity threshold, it is determined that the source of the polishing anomaly is an amorphous rod defect, that is, it comes from the chip processing flow, and then the source of the polishing anomaly is determined by backtracing the chip processing flow.

[0024] See also Figure 3 , Figure 3 : is a schematic diagram of the structure of a sapphire crystal rod polishing abnormality data analysis system based on image processing provided in an embodiment of the present application. This embodiment provides a sapphire crystal rod polishing abnormality data analysis system based on image processing, including: The data acquisition module 210 is used to collect sapphire crystal rod cutting data and wafer surface images; A wafer source analysis module 220 is used to determine the correlation between wafers and ingots based on the ingot cutting data and analyze the similarity of wafer sources; A polishing anomaly analysis module 230, for identifying polishing anomaly data based on the wafer surface image and analyzing polishing anomaly similarities; The abnormality source identification module 240 is used to determine the source of the polishing abnormality by combining the wafer source similarity and the polishing abnormality similarity.

[0025] In the embodiment of the present application, the wafer source analysis module 220 is used to determine the correlation between the wafer and the crystal ingot according to the crystal ingot cutting data and analyze the similarity of the wafer sources. The step of analyzing the similarity of the wafer sources includes: Obtaining crystal ingot cutting data, where the crystal ingot cutting data includes cutting wafer information and cutting sequence number information corresponding to the wafer; For any two chips and Chip , by cutting the wafer information to determine whether the two belong to the same crystal rod, when the wafer and Chip If they do not belong to the same crystal rod, the chip and Chip Similarity of chip sources ; When the chip and Chip If they belong to the same crystal rod, the wafer source similarity is calculated by the cutting sequence number information, and the wafer source similarity is used to quantitatively analyze the wafer source similarity; The calculation formula of chip source similarity is as follows: ; in, Indicates chip Cutting sequence number information, Indicates chip Cutting sequence number information, Indicates chip and Chip The number of wafers cut from the respective ingot, Indicates chip and Chip The similarity of the chip sources.

[0026] In the embodiment of the present application, the polishing abnormality analysis module 230 is used to identify the polishing abnormality data according to the wafer surface image and analyze the polishing abnormality similarity. The step of identifying the polishing abnormality data includes: Acquire the wafer surface image and perform preprocessing, which includes image normalization, image denoising and image enhancement; The pre-processed wafer surface image is subjected to feature extraction by a feature extraction algorithm, wherein the feature extraction includes color feature extraction, texture feature extraction and edge feature extraction; Build an abnormal data classification model based on machine learning algorithm, train the abnormal data classification model through feature extraction results and obtain polished abnormal data; The steps to analyze the similarity of polishing anomalies include: Obtaining polishing abnormality data and performing binarization processing on the wafer surface image according to the polishing abnormality data to obtain a wafer surface binarization image, wherein the pixel value of a normal pixel in the wafer surface binarization image is 0, and the pixel value of an abnormal pixel in the wafer surface binarization image is 1; For any two chips and Chip , compare the pixel values ​​of the pixels at the same position in the binary image of the wafer surface, and calculate the polishing anomaly similarity by counting the number of abnormal pixels with matching positions and the total number of abnormal pixels. The calculation formula for the polishing anomaly similarity is as follows: ; in, Indicates chip The pixel coordinates in the binary image of the wafer surface The pixel value at Indicates chip The pixel coordinates in the binary image of the wafer surface The pixel value at Indicates chip and Chip The pixel coordinate set of Indicates chip and Chip The polishing anomaly similarity is used to quantitatively analyze the similarity of wafer polishing anomalies.

[0027] In the embodiment of the present application, the abnormality source identification module 240 is used to determine the source of the polishing abnormality by combining the wafer source similarity and the polishing abnormality similarity. The step of determining the source of the polishing abnormality includes: Obtain wafer source similarity and polishing anomaly similarity; When the sum of the wafer source similarity and the polishing anomaly similarity is greater than or equal to the preset similarity threshold, it is determined that the source of the polishing anomaly is a crystal rod defect; When the sum of the wafer source similarity and the polishing anomaly similarity is less than a preset similarity threshold, it is determined that the source of the polishing anomaly is an amorphous rod defect.

[0028] The above-mentioned parameters and steps for each unit module to implement corresponding functions in the sapphire crystal rod polishing abnormality data analysis system based on image processing of the present application can refer to the parameters and steps in the embodiment of the sapphire crystal rod polishing abnormality data analysis method based on image processing above, and will not be repeated here.

[0029] like Figure 4 As shown, an embodiment of the present application further provides an electronic device 300, including a memory 310, a processor 320 and a communication bus 330; the memory 310 and the processor 320 are connected via the communication bus 330. The memory 310 stores a sapphire crystal rod polishing abnormality data analysis method based on image processing as provided in the above embodiment, which can be loaded and executed by the processor 320.

[0030] The memory 310 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 310 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function, and instructions for implementing the sapphire crystal rod polishing abnormality data analysis method based on image processing provided in the above embodiment, etc.; the data storage area may store data involved in the sapphire crystal rod polishing abnormality data analysis method based on image processing provided in the above embodiment, etc.

[0031] The processor 320 may include one or more processing cores. The processor 320 executes various functions and processes data of the present application by running or executing instructions, programs, code sets or instruction sets stored in the memory 310, calling the data stored in the memory 310. The processor 320 may be at least one of an application-specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field programmable gate array (FPGA), a central processing unit (CPU), a controller, a microcontroller and a microprocessor. It can be understood that for different devices, the electronic device used to implement the above-mentioned processor 320 function can also be other, and the embodiment of the present application is not specifically limited.

[0032] The communication bus 330 may include a path to transmit information between the above components. The communication bus 330 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The communication bus 330 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one double arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0033] This embodiment provides a computer-readable storage medium on which an erasable computer program is stored. When the computer program is run on a computer device, the computer device executes the above-mentioned sapphire crystal rod polishing abnormality data analysis method based on image processing.

[0034] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0035] It should be understood that determining B based on A does not mean determining B only based on A. B can also be determined based on A and / or other information.

[0036] Those skilled in the art will appreciate that the present application may be implemented as a system, method or computer program product.

[0037] Therefore, the present application can be specifically implemented in the following forms, namely: it can be complete hardware, it can be complete software (including firmware, resident software, microcode, etc.), or it can be a combination of hardware and software, which is generally referred to as "circuit", "module" or "system" in the present application. In addition, in some embodiments, the present application can also be implemented in the form of a computer program product in one or more computer-readable media, and the computer-readable medium contains computer-readable program code.

[0038] Any combination of one or more computer-readable media may be used. A computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, device, or device.

[0039] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for analyzing sapphire crystal rod polishing abnormality data based on image processing, characterized in that: The following steps are involved: Collect sapphire ingot cutting data and wafer surface images; Determine the correlation between the wafer and the ingot according to the ingot cutting data and analyze the similarity of the wafer sources; identifying polishing anomaly data based on the wafer surface image and analyzing polishing anomaly similarities; The source of the polishing anomaly is determined by combining the wafer source similarity and the polishing anomaly similarity.

2. The method for analyzing sapphire crystal rod polishing abnormality data based on image processing according to claim 1, characterized in that: The step of analyzing the similarity of wafer sources comprises: Acquire the crystal ingot cutting data, wherein the crystal ingot cutting data includes cutting wafer information and cutting sequence number information corresponding to the wafer; For any two chips and Chip , judging whether the two belong to the same crystal rod by the information of the cut wafer, when the wafer and Chip If they do not belong to the same crystal rod, the chip and Chip Similarity of chip sources ; When the chip and Chip When the wafers belong to the same crystal rod, the wafer source similarity is calculated according to the cutting sequence number information, and the wafer source similarity is used to quantitatively analyze the wafer source similarity.

3. The sapphire crystal rod polishing abnormality data analysis method based on image processing according to claim 2 is characterized in that: The calculation formula of the wafer source similarity is as follows: ； in, Indicates chip Cutting sequence number information, Indicates chip Cutting sequence number information, Indicates chip and Chip The number of wafers cut from the respective ingot, Indicates chip and Chip The similarity of the chip sources.

4. The method for analyzing sapphire crystal rod polishing abnormality data based on image processing according to claim 1, characterized in that: The step of identifying polishing abnormal data comprises: Acquiring the wafer surface image and performing preprocessing, wherein the preprocessing includes image normalization, image denoising and image enhancement; Performing feature extraction on the preprocessed wafer surface image by a feature extraction algorithm, wherein the feature extraction includes color feature extraction, texture feature extraction and edge feature extraction; An abnormal data classification model is constructed based on a machine learning algorithm, and the abnormal data classification model is trained through feature extraction results to obtain the polished abnormal data.

5. The method for analyzing sapphire crystal rod polishing abnormality data based on image processing according to claim 1, characterized in that: The step of analyzing the similarity of polishing anomalies comprises: Acquire the polishing abnormality data and perform binarization processing on the wafer surface image according to the polishing abnormality data to obtain a wafer surface binarization image, wherein the pixel value of a normal pixel in the wafer surface binarization image is 0, and the pixel value of an abnormal pixel in the wafer surface binarization image is 1; For any two chips and Chip , compare the pixel values ​​of pixels at the same position in the binary image of the wafer surface, and calculate the polishing anomaly similarity by counting the number of abnormal pixels with matching positions and the total number of abnormal pixels. The calculation formula of the polishing anomaly similarity is as follows: ； in, Indicates chip The pixel coordinates in the binary image of the wafer surface The pixel value at Indicates chip The pixel coordinates in the binary image of the wafer surface The pixel value at Indicates chip and Chip The pixel coordinate set of Indicates chip and Chip The polishing anomaly similarity is used to quantitatively analyze the similarity of wafer polishing anomalies.

6. The method for analyzing sapphire crystal rod polishing abnormality data based on image processing according to claim 1, characterized in that: The step of determining the source of the polishing abnormality comprises: Obtaining the wafer source similarity and the polishing anomaly similarity; When the sum of the wafer source similarity and the polishing anomaly similarity is greater than or equal to a preset similarity threshold, it is determined that the source of the polishing anomaly is a crystal rod defect; When the sum of the wafer source similarity and the polishing anomaly similarity is less than the preset similarity threshold, it is determined that the source of the polishing anomaly is an amorphous rod defect.

7. A sapphire crystal rod polishing abnormality data analysis system based on image processing, applied to the sapphire crystal rod polishing abnormality data analysis method based on image processing according to any one of claims 1 to 6, characterized in that: The system comprises: A data acquisition module, used to collect sapphire crystal rod cutting data and wafer surface images; A wafer source analysis module, used to determine the correlation between wafers and crystal bars according to the crystal bar cutting data and analyze the similarity of wafer sources; A polishing anomaly analysis module, used to identify polishing anomaly data and analyze polishing anomaly similarities based on the wafer surface image; The abnormality source identification module is used to determine the source of the polishing abnormality by combining the wafer source similarity and the polishing abnormality similarity.

8. An electronic device, comprising: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; characterized in that the processor executes the sapphire crystal rod polishing abnormality data analysis method based on image processing as described in any one of claims 1 to 6 by calling the computer program stored in the memory.

9. A computer-readable storage medium, characterized in that: Instructions are stored, and when the instructions are executed on a computer, the computer executes the sapphire crystal rod polishing abnormality data analysis method based on image processing as described in any one of claims 1 to 6.

Images