Sapphire slice quality evaluation method based on machine vision

Abstract

The invention relates to the technical field of image processing, in particular to a sapphire slice quality evaluation method based on machine vision, and the method comprises the steps: obtaining a sapphire slice image, and extracting slice defect features through an image processing algorithm; identifying a defect gathering area of the sapphire slice through a sliding window algorithm; evaluating the defect coupling degree of the defect aggregation area based on the slice defect features; and evaluating the slice quality based on the defect coupling degree and performing slice quality early warning. According to the method, the defect coupling degree of the defect aggregation area is quantified by analyzing the defect depth distribution condition of the sapphire slice, and the accuracy and reliability of slice quality evaluation are effectively improved.

Description

Technical Field

[0001] The present application relates to the technical field of image processing, and in particular to a sapphire slice quality assessment method based on machine vision. Background Art

[0002] Sapphire is a typical hard and brittle material with a unique crystal structure, excellent chemical stability and good optical properties. At the same time, sapphire has the characteristics of high hardness, corrosion resistance and high melting point, so it is widely used in optoelectronics, national defense, aerospace and other fields.

[0003] Slicing is the first step in sapphire processing. The quality of slicing largely determines the quality and efficiency of subsequent processing. Due to the hardness and brittleness of sapphire itself and its difficult processing characteristics and properties, various defects will inevitably occur in the sapphire slicing process. Among them, the appearance defect types of sapphire slices mainly include scratch defects and bubble defects.

[0004] The existing technology usually only focuses on the surface morphological characteristics of defects, such as length, width, area, etc., when evaluating the quality of sapphire slices. The distribution of scratch depth, bubble depth and bubble volume not only directly affects the mechanical properties of the slices, but is also closely related to the efficiency and accuracy of subsequent processing. For example, larger defect depth and defect volume will significantly increase the risk of slice breakage, affecting the product yield and processing efficiency. Therefore, the existing technology ignores the defect depth and defect volume characteristics and their correlation when evaluating the quality of sapphire slices, resulting in insufficient accuracy in slice quality assessment and omission of potential risks. Summary of the invention

[0005] In order to overcome the defects and shortcomings of the prior art, the present application provides a sapphire slice quality assessment method based on machine vision, which analyzes the scratch defect depth and bubble defect depth distribution of the sapphire slice and quantifies the defect coupling degree in the defect aggregation area, thereby effectively improving the accuracy and reliability of the slice quality assessment.

[0006] In order to achieve the above objectives, this application adopts the following technical solutions: In a first aspect, the present application provides a sapphire slice quality assessment method based on machine vision, comprising the following steps: Acquire sapphire slice images and extract slice defect features through image processing algorithms; Identify defect clustering areas of sapphire slices using a sliding window algorithm; Evaluate the degree of defect coupling in defect clustering areas based on slice defect characteristics; The slicing quality is evaluated based on the degree of defect coupling and a slicing quality warning is issued.

[0007] Preferably, the step of identifying the defect concentration area of ​​the sapphire slice comprises: Get the coordinates of the defect centroid point in the slice defect feature and pass the window size And the sliding step length is The sliding window traverses the sapphire slice image, where ; Count the coordinates of the defect centroid points within the sliding window and construct a set of defect centroid point coordinates ,in, and are the coordinate values ​​of the first defect centroid on the x-axis and y-axis respectively, and are the coordinate values ​​of the second defect centroid on the x-axis and y-axis respectively, and Respectively The coordinate values ​​of the defect centroid on the x-axis and y-axis, is the number of defect centroid points within the sliding window; The defect aggregation index of the sliding window is calculated by the defect centroid point coordinate set, and the sliding window with a defect aggregation index greater than a preset defect aggregation threshold is taken as the defect aggregation area.

[0008] Preferably, the calculating of the defect aggregation index of the sliding window includes: Get the defect centroid coordinate set of the sliding window; The nearest neighbor distance calculation formula , get the first The nearest neighbor distance between the defect centroid and other defect centroids ; The nearest neighbor distance mean is calculated using the formula , get the mean nearest neighbor distance ; The defect aggregation index is calculated by the formula , get the defect aggregation index of the sliding window ,in, is the expected nearest neighbor distance when defects are evenly distributed, , is the sliding window area, is the number of defect centroids within the sliding window.

[0009] Preferably, the step of evaluating the defect coupling degree of the defect clustering area includes: Obtaining slice defect characteristics within the defect clustering area; According to the defect characteristics of the slice, the defects in the defect clustering area are divided into scratch defects and bubble defects, and the scratch defect depth, bubble defect depth and bubble defect volume are obtained; Defect Coupling Index Evaluate the degree of defect coupling in defect clustering areas, defect coupling index The calculation formula is: ; in, The defect clustering area Scratch defect depth, is the number of scratch defects in the defect aggregation area, is the slice thickness, The defect clustering area The volume of bubble defects, is the number of bubble defects in the defect aggregation area, is the slice volume of the defect aggregation area, The defect clustering area The scratch defect and The defect depth correlation coefficient of a bubble defect is , The defect clustering area The depth of the bubble defect, is the scratch defect depth weight, is the bubble defect volume weight, is the defect depth association weight.

[0010] Preferably, the method of dividing the defects in the defect aggregation area into scratch defects and bubble defects according to the slice defect characteristics includes: Obtain the defect area and defect perimeter in the slice defect characteristics; Circularity calculation formula , get the circularity of the defect ,in, is the defect area, is the defect perimeter, is pi; Defects with a circularity greater than a preset circularity threshold are classified as bubble defects, and defects with a circularity less than or equal to the preset circularity threshold are classified as scratch defects.

[0011] Preferably, the evaluating the slice quality and issuing a slice quality warning includes: Obtain the defect coupling index of the defect clustering area; The average value of the defect coupling index of all defect clustering areas is used as the slice quality risk index, and the slice quality is evaluated by the slice quality risk index; When the slice quality risk index is greater than the preset slice quality threshold, a slice quality warning is issued; when the slice quality risk index is less than or equal to the preset slice quality threshold, no slice quality warning is issued.

[0012] It should be noted here that the scratch defect depth weight, bubble defect volume weight, defect depth association weight, preset defect aggregation threshold, preset circularity threshold and preset slice quality threshold are determined as follows: 5,000 sets of sapphire slice images are collected and slice defect features are extracted through image processing algorithms to distinguish whether the slice quality meets production requirements. The slice defect features are substituted into the slice quality risk index calculation formula for calculation, and the calculated slice quality risk index and the distinction result are simultaneously imported into the fitting software, and the optimal scratch defect depth weight, bubble defect volume weight, defect depth association weight, preset defect aggregation threshold, preset circularity threshold and preset slice quality threshold that meet the distinction accuracy of the distinction result are output.

[0013] In a second aspect, the present application provides a sapphire slice quality assessment system based on machine vision, comprising: A feature extraction module, used to obtain sapphire slice images and extract slice defect features through image processing algorithms; A defect clustering region identification module is used to identify the defect clustering region of the sapphire slice by a sliding window algorithm; A defect coupling degree assessment module is used to assess the defect coupling degree of the defect clustering area based on the slice defect characteristics; The slice quality warning module is used to evaluate the slice quality and issue slice quality warning based on the degree of defect coupling.

[0014] 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 slice quality assessment method based on machine vision by calling the computer program stored in the memory.

[0015] 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 slice quality assessment method based on machine vision.

[0016] Compared with the prior art, this application has the following advantages and beneficial effects: This application first extracts slice defect features in sapphire slice images through an image processing algorithm, then identifies the defect clustering area of ​​the sapphire slice through a sliding window algorithm, and finally quantifies the defect coupling degree of the defect clustering area by analyzing the distribution of scratch defect depth and bubble defect depth of the sapphire slice, thereby effectively improving the accuracy and reliability of slice quality assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] 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 slice quality assessment method based on machine vision provided in an embodiment of the present application; Figure 2 Schematic diagram of the structure of a sapphire slice quality assessment system based on machine vision provided in an embodiment of the present application; Figure 3 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] 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.

[0019] like Figure 1 As shown, Figure 1 : is a schematic diagram of the overall process of the sapphire slice quality assessment method based on machine vision provided in an embodiment of the present application, which specifically includes the following steps: S110: Acquire a sapphire slice image and extract slice defect features through an image processing algorithm, wherein the sapphire slice image is acquired by a CCD camera, incident light is irradiated onto the surface of the sapphire slice through an illumination device, and reflected light enters the CCD camera, and the CCD camera converts the target area signal of the sapphire slice into an electrical signal, and then uses an image acquisition card to convert the electrical signal into a digital signal to obtain the sapphire slice image; The steps of extracting slice defect features through image processing algorithms include: (1) Preprocessing the sapphire slice image, the preprocessing includes image denoising and image enhancement, the image denoising method is any one of median filtering, Gaussian filtering and bilateral filtering, and the image enhancement method is any one of histogram equalization, Gamma transform and Laplace transform; (2) Extracting slice defects from the preprocessed sapphire slice image through image segmentation algorithms, which include threshold segmentation algorithms (such as global threshold, adaptive threshold), edge detection segmentation algorithms (such as Canny, Sobel), region growing algorithms, and deep learning segmentation algorithms (such as U-Net, FCN, DeepLab); (3) The slice defect features are extracted through the feature extraction algorithm. The slice defect features include the defect perimeter, defect area and defect centroid coordinates. The feature extraction algorithm includes contour extraction algorithm, morphological algorithm and connected domain analysis algorithm.

[0020] S120: identifying defect clustering areas of the sapphire slice by a sliding window algorithm; In the processing of sapphire slices, the instability of the cutting process usually leads to various appearance defects on the slice surface, such as crack defects, bubble defects and scratch defects. Appearance defects not only affect the appearance quality of the slice, but also increase the workload of post-processing. By using the sliding window algorithm to locally scan the sapphire slice image and gradually traverse the entire slice area, the defect centroid points in each sliding window are counted and the defect aggregation index is calculated, and then the defect distribution in each sliding window is analyzed to identify the area with dense defect distribution, which provides data support for subsequent defect coupling analysis and slice quality assessment, and identifies the defect aggregation area of ​​sapphire slices, including: Get the coordinates of the defect centroid point in the slice defect feature and pass the window size And the sliding step length is The sliding window traverses the sapphire slice image, where ; A plane rectangular coordinate system is constructed with the center point of the sapphire slice as the coordinate origin, the horizontal direction of the slice (the length direction of the sapphire slice image) as the x-axis, and the vertical direction of the slice (the width direction of the sapphire slice image) as the y-axis. The coordinates of the defect centroid points within the sliding window are counted and a defect centroid point coordinate set is constructed. ,in, and are the coordinate values ​​of the first defect centroid on the x-axis and y-axis respectively, and are the coordinate values ​​of the second defect centroid on the x-axis and y-axis respectively, and Respectively The coordinate values ​​of the defect centroid on the x-axis and y-axis, is the number of defect centroid points within the sliding window; Calculate the defect aggregation index of the sliding window through the defect centroid point coordinate set and use the sliding window with a defect aggregation index greater than a preset defect aggregation threshold as the defect aggregation area; The defect aggregation index quantifies the degree of defect aggregation by analyzing the spatial distribution of defect centroids in the sliding window. When the defects are evenly distributed, the nearest neighbor distance between the defect centroids is large and the aggregation index is low. When the defects are concentrated in certain areas, the nearest neighbor distance between the defect centroids is small, and the aggregation index is high. The defect aggregation index of the sliding window is calculated, including: Get the defect centroid coordinate set of the sliding window; The nearest neighbor distance calculation formula , get the first The nearest neighbor distance between the defect centroid and other defect centroids ; The nearest neighbor distance mean is calculated using the formula , get the mean nearest neighbor distance ; The defect aggregation index is calculated by the formula , get the defect aggregation index of the sliding window ,in, is the expected nearest neighbor distance when the defects are evenly distributed, that is, the distance between the defect centroids in the case of random distribution (i.e., the distribution of the defect centroids is completely irregular), , is the sliding window area, is the number of defect centroids within the sliding window.

[0021] S130: evaluating the degree of defect coupling in the defect clustering area based on the defect characteristics of the slice; The scratch defect depth, bubble defect depth and bubble defect volume can directly reflect the spatial distribution of defects in the slice and its potential impact on the slice quality. Scratch defects usually affect the surface finish and optical properties, while bubble defects affect the mechanical strength and stability of the slice. By comprehensively analyzing the scratch defect depth, bubble defect depth and bubble defect volume, the coupling effect between defects can be more accurately evaluated, that is, whether there is the possibility of interaction and mutual reinforcement between defects, and the degree of defect coupling in the defect clustering area can be evaluated, including: Obtaining slice defect characteristics within the defect clustering area; According to the defect characteristics of the slice, the defects in the defect clustering area are divided into scratch defects and bubble defects, and the scratch defect depth, bubble defect depth and bubble defect volume are obtained, wherein the scratch defect depth, bubble defect depth and bubble defect volume are obtained by laser scanning confocal microscopy (LSCM), X-ray computed tomography (XCT) and white light interferometry; Defect Coupling Index Evaluate the degree of defect coupling in defect clustering areas, defect coupling index The calculation formula is: ; in, The defect clustering area Scratch defect depth, is the number of scratch defects in the defect aggregation area, is the slice thickness, It is the ratio of the square of the scratch depth to the square of the slice thickness, reflecting the influence of the scratch defect in the slice thickness direction. The defect clustering area The volume of bubble defects, is the number of bubble defects in the defect aggregation area, is the slice volume of the defect aggregation area, is the ratio of the total volume of bubbles in the defect aggregation area to the total volume of the slice, The defect clustering area The scratch defect and The defect depth correlation coefficient of a bubble defect is , The defect clustering area The depth of the bubble defect, is a natural constant. The defect depth correlation coefficient is used to quantify the depth proximity between the scratch defect and the bubble defect. When the difference between the scratch defect depth and the bubble defect depth is smaller, that is, the depth is closer, the defect depth correlation coefficient is larger. is the scratch defect depth weight, is the bubble defect volume weight, is the defect depth association weight; The purpose of dividing defects into scratch defects and bubble defects is to provide a basis for the subsequent measurement of defect depth according to the morphological characteristics of the defects. Scratch defects usually have a lower circularity and a narrow and long shape, while bubble defects usually have a higher circularity and a shape close to a circle. According to the characteristics of the slice defects, the defects in the defect aggregation area are divided into scratch defects and bubble defects, including: Obtain the defect area and defect perimeter in the slice defect characteristics; Circularity calculation formula , get the circularity of the defect ,in, is the defect area, is the defect perimeter, is the ratio of pi; Defects with a circularity greater than a preset circularity threshold are classified as bubble defects, and defects with a circularity less than or equal to the preset circularity threshold are classified as scratch defects.

[0022] S140: Evaluate slice quality based on defect coupling degree and issue slice quality warning; By calculating the defect coupling index, the mutual influence between different defects in the defect clustering area can be effectively identified, and then the slice quality can be quantitatively analyzed to ensure timely identification and early warning of potential slice quality problems, evaluate slice quality and issue slice quality early warning, including: Obtain the defect coupling index of the defect clustering area; The average value of the defect coupling index of all defect clustering areas is used as the slice quality risk index, and the slice quality is evaluated by the slice quality risk index; When the slice quality risk index is greater than the preset slice quality threshold, a slice quality warning is issued; when the slice quality risk index is less than or equal to the preset slice quality threshold, no slice quality warning is issued.

[0023] like Figure 2 As shown, Figure 2 : is a schematic diagram of the structure of a sapphire slice quality assessment system based on machine vision provided in an embodiment of the present application. This embodiment provides a sapphire slice quality assessment system based on machine vision, including: Feature extraction module 210, used to obtain sapphire slice images and extract slice defect features through image processing algorithms; A defect clustering region identification module 220 is used to identify the defect clustering region of the sapphire slice by a sliding window algorithm; A defect coupling degree evaluation module 230 is used to evaluate the defect coupling degree of the defect clustering area based on the slice defect characteristics; The slice quality warning module 240 is used to evaluate the slice quality based on the defect coupling degree and issue a slice quality warning.

[0024] In the embodiment of the present application, the defect clustering region identification module 220 is used to identify the defect clustering region of the sapphire slice by using a sliding window algorithm. The defect clustering region of the sapphire slice is identified by: Get the coordinates of the defect centroid point in the slice defect feature and pass the window size And the sliding step length is The sliding window traverses the sapphire slice image, where ; Count the coordinates of the defect centroid points within the sliding window and construct a set of defect centroid point coordinates ,in, and are the coordinate values ​​of the first defect centroid on the x-axis and y-axis respectively, and are the coordinate values ​​of the second defect centroid on the x-axis and y-axis respectively, and Respectively The coordinate values ​​of the defect centroid on the x-axis and y-axis, is the number of defect centroid points within the sliding window; Calculate the defect aggregation index of the sliding window through the defect centroid point coordinate set and use the sliding window with a defect aggregation index greater than a preset defect aggregation threshold as the defect aggregation area; Calculate the defect aggregation index of the sliding window, including: Get the defect centroid coordinate set of the sliding window; The nearest neighbor distance calculation formula , get the first The nearest neighbor distance between the defect centroid and other defect centroids ; The nearest neighbor distance mean is calculated using the formula , get the mean nearest neighbor distance ; The defect aggregation index is calculated by the formula , get the defect aggregation index of the sliding window ,in, is the expected nearest neighbor distance when defects are evenly distributed, , is the sliding window area, is the number of defect centroids within the sliding window.

[0025] In the embodiment of the present application, the defect coupling degree evaluation module 230 is used to evaluate the defect coupling degree of the defect clustering area based on the slice defect characteristics, and the defect coupling degree of the defect clustering area is evaluated, including: Obtaining slice defect characteristics within the defect clustering area; According to the defect characteristics of the slice, the defects in the defect clustering area are divided into scratch defects and bubble defects, and the scratch defect depth, bubble defect depth and bubble defect volume are obtained; Defect Coupling Index Evaluate the degree of defect coupling in defect clustering areas, defect coupling index The calculation formula is: ; in, The defect clustering area Scratch defect depth, is the number of scratch defects in the defect aggregation area, is the slice thickness, The defect clustering area The volume of bubble defects, is the number of bubble defects in the defect aggregation area, is the slice volume of the defect aggregation area, The defect clustering area The scratch defect and The defect depth correlation coefficient of a bubble defect is , The defect clustering area The depth of the bubble defect, is the scratch defect depth weight, is the bubble defect volume weight, is the defect depth association weight; According to the characteristics of slice defects, the defects in the defect clustering area are divided into scratch defects and bubble defects, including: Obtain the defect area and defect perimeter in the slice defect characteristics; Circularity calculation formula , get the circularity of the defect ,in, is the defect area, is the defect perimeter, is pi; Defects with a circularity greater than a preset circularity threshold are classified as bubble defects, and defects with a circularity less than or equal to the preset circularity threshold are classified as scratch defects.

[0026] In the embodiment of the present application, the slice quality warning module 240 is used to evaluate the slice quality and issue a slice quality warning based on the defect coupling degree. The evaluation of the slice quality and the slice quality warning include: Obtain the defect coupling index of the defect clustering area; The average value of the defect coupling index of all defect clustering areas is used as the slice quality risk index, and the slice quality is evaluated by the slice quality risk index; When the slice quality risk index is greater than the preset slice quality threshold, a slice quality warning is issued; when the slice quality risk index is less than or equal to the preset slice quality threshold, no slice quality warning is issued.

[0027] The above-mentioned parameters and steps for each unit module to implement corresponding functions in the sapphire slice quality assessment system based on machine vision of the present application can refer to the parameters and steps in the embodiment of the sapphire slice quality assessment method based on machine vision above, and will not be repeated here.

[0028] like Figure 3 As shown, an embodiment of the present invention 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 slice quality assessment method based on machine vision as provided in the above embodiment, which can be loaded and executed by the processor 320.

[0029] The memory 310 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 310 can include a program storage area and a data storage area, wherein the program storage area can store instructions for implementing an operating system, instructions for at least one function, and instructions for implementing the sapphire slice quality assessment method based on machine vision provided in the above embodiment, etc.; the data storage area can store data involved in the sapphire slice quality assessment method based on machine vision provided in the above embodiment, etc.

[0030] 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.

[0031] 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 3 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.

[0032] An embodiment of the present application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and execute a sapphire slice quality assessment method based on machine vision as provided in the above embodiment.

[0033] In this embodiment, the computer-readable storage medium may be a tangible device that holds and stores instructions used by the instruction execution device. The computer-readable storage medium may be, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination thereof. Specifically, the computer-readable storage medium may be a portable computer disk, a hard disk, a USB flash drive, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a podium random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, an optical disk, a magnetic disk, a mechanical encoding device, and any combination thereof.

[0034] The terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not expressly listed, or also includes elements inherent to such process, method, article, or apparatus.

[0035] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of application involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the aforementioned application concept. For example, the above features are replaced with (but not limited to) technical features with similar functions applied in the present application.

Claims

1. A sapphire slice quality assessment method based on machine vision, characterized in that: The steps include: Acquire sapphire slice images and extract slice defect features through image processing algorithms; Identify defect clustering areas of sapphire slices using a sliding window algorithm; Evaluate the degree of defect coupling in defect clustering areas based on slice defect characteristics; The slicing quality is evaluated based on the degree of defect coupling and a slicing quality warning is issued.

2. The sapphire slice quality assessment method based on machine vision according to claim 1, characterized in that: The method of identifying the defect concentration area of ​​the sapphire slice comprises: Get the coordinates of the defect centroid point in the slice defect feature and pass the window size And the sliding step length is The sliding window traverses the sapphire slice image, where ; Count the coordinates of the defect centroid points within the sliding window and construct a set of defect centroid point coordinates ,in, and are the coordinate values ​​of the first defect centroid on the x-axis and y-axis respectively, and are the coordinate values ​​of the second defect centroid on the x-axis and y-axis respectively, and Respectively The coordinate values ​​of the defect centroid on the x-axis and y-axis, is the number of defect centroid points within the sliding window; The defect aggregation index of the sliding window is calculated by the defect centroid point coordinate set, and the sliding window with a defect aggregation index greater than a preset defect aggregation threshold is taken as the defect aggregation area.

3. The sapphire slice quality assessment method based on machine vision according to claim 2, characterized in that: The step of calculating the defect aggregation index of the sliding window includes: Get the defect centroid coordinate set of the sliding window; The nearest neighbor distance calculation formula , get the first The nearest neighbor distance between the defect centroid and other defect centroids ; The nearest neighbor distance mean is calculated using the formula , get the mean nearest neighbor distance ; The defect aggregation index is calculated by the formula , get the defect aggregation index of the sliding window ,in, is the expected nearest neighbor distance when defects are evenly distributed, , is the sliding window area, is the number of defect centroids within the sliding window.

4. The sapphire slice quality assessment method based on machine vision according to claim 1, characterized in that: The step of evaluating the degree of defect coupling in the defect clustering area includes: Obtaining slice defect characteristics within the defect clustering area; According to the defect characteristics of the slice, the defects in the defect clustering area are divided into scratch defects and bubble defects, and the scratch defect depth, bubble defect depth and bubble defect volume are obtained; Defect Coupling Index Evaluate the degree of defect coupling in defect clustering areas, defect coupling index The calculation formula is: ； in, The defect clustering area Scratch defect depth, is the number of scratch defects in the defect aggregation area, is the slice thickness, The defect clustering area The volume of bubble defects, is the number of bubble defects in the defect aggregation area, is the slice volume of the defect aggregation area, The defect clustering area The scratch defect and The defect depth correlation coefficient of a bubble defect is , The defect clustering area The depth of the bubble defect, is the scratch defect depth weight, is the bubble defect volume weight, is the defect depth association weight.

5. The sapphire slice quality assessment method based on machine vision according to claim 4, characterized in that: The method of dividing the defects in the defect aggregation area into scratch defects and bubble defects according to the slice defect characteristics includes: Obtain the defect area and defect perimeter in the slice defect characteristics; Circularity calculation formula , get the circularity of the defect ,in, is the defect area, is the defect perimeter, is the ratio of pi; Defects with a circularity greater than a preset circularity threshold are classified as bubble defects, and defects with a circularity less than or equal to the preset circularity threshold are classified as scratch defects.

6. The sapphire slice quality assessment method based on machine vision according to claim 1, characterized in that: The step of evaluating slice quality and issuing slice quality warning includes: Obtain the defect coupling index of the defect clustering area; The average value of the defect coupling index of all defect clustering areas is used as the slice quality risk index, and the slice quality is evaluated by the slice quality risk index; When the slice quality risk index is greater than the preset slice quality threshold, a slice quality warning is issued; when the slice quality risk index is less than or equal to the preset slice quality threshold, no slice quality warning is issued.

7. A sapphire slice quality assessment system based on machine vision, applied to the sapphire slice quality assessment method based on machine vision according to any one of claims 1 to 6, characterized in that: The system comprises: A feature extraction module, used to obtain sapphire slice images and extract slice defect features through image processing algorithms; A defect clustering region identification module is used to identify the defect clustering region of the sapphire slice by a sliding window algorithm; A defect coupling degree assessment module is used to assess the defect coupling degree of the defect clustering area based on the slice defect characteristics; The slice quality warning module is used to evaluate the slice quality and issue slice quality warning based on the degree of defect coupling.

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 slice quality assessment method based on machine vision 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 slice quality assessment method based on machine vision as described in any one of claims 1 to 6.

Images