Method and apparatus for tracing and grading defects before and after frosting of screen glass

Abstract

The application is suitable for the field of digital manufacturing technology, and particularly relates to a screen glass frosting defect tracing and grading method and device before and after frosting, which comprises the following steps: scattering characteristics of a frosting microstructure under a current frosting process state are characterized by a statistical reference scattering angle distribution; abnormal areas are separated by differential analysis of differences between each local area and the statistical reference; global scattering changes caused by frosting process fluctuations are eliminated as common mode components; only scattering abnormalities of local areas are retained, so that the strong scattering background of the frosting surface is overcome to eliminate the flooding of defect signals. Deviation patterns of the abnormal areas relative to the statistical reference scattering angle distribution are extracted as topographic signatures, and a defect production process is determined based on the topographic signatures; the abnormal areas are graded according to the defect production process and the topographic signatures; the frosting process is used to determine the defect topography, reduce false rejection and missed release, and improve the fine level of screen glass quality control.

Description

Technical Field

[0001] This application belongs to the field of digital manufacturing technology, and in particular relates to a method and equipment for tracing and classifying defects before and after screen glass frosting. Background Technology

[0002] Screen glass is a crucial component of electronic devices such as smartphones, tablets, and automotive displays. To reduce screen glare, improve display quality, and minimize fingerprints, screen glass is typically treated with a frosting process, creating a matte finish with random microstructures on its surface. Frosting processes generally include two types: chemical etching and sandblasting. Chemical etching uses hydrofluoric acid-based etching solutions to controllably etch the glass surface, forming random, uneven microstructures on the submicron to micron scale.

[0003] While the frosting process imparts excellent optical properties to screen glass, it also presents challenges in defect detection and quality control. The frosting process involves multiple steps, and defects can arise at different stages: the original sheet may carry defects such as cracks and scratches before frosting; during the frosting process, factors such as fluctuations in etching solution concentration, uneven temperature distribution, and residual etching solution can lead to defects such as uneven etching, white spots, and pitting; post-frosting defects such as scratches, indentations, and stains may be introduced during handling, cleaning, and packaging. The responsibility, reworkability, and handling methods for defects arising at different stages vary. Therefore, tracing the source of defects at each stage and classifying their handling is a crucial foundation for improving screen glass yield and optimizing frosting process parameters.

[0004] However, after frosting, the screen glass surface changes from a smooth mirror surface to a random microstructured frosted surface. The frosted microstructures generate strong diffuse scattering of incident light, forming a scattering background covering the entire surface. Defect signals are submerged in the strong scattering background generated by the frosted microstructures, resulting in extremely low defect contrast obtained by conventional optical detection methods on the frosted surface, making it difficult to reliably identify defects. Existing polarization enhancement schemes can only improve the grayscale contrast between specific targets and the background, and cannot distinguish between frosting process fluctuations and real defects; deep learning-based end-to-end recognition schemes rely on a large number of labeled samples, and the output results lack physical interpretability, failing to indicate the process stage to which the defect belongs; Fourier transform-based detection schemes mainly target the phase difference caused by uneven glass thickness, which is irrelevant to the defect detection problem under the scattering background of the frosted surface.

[0005] Furthermore, the art believes that frosted surfaces, due to their strong background scattering, have long been considered a difficult technical challenge to achieve, making it difficult to accurately detect defects online, let alone trace the source of defects at the process stage while detecting them. Summary of the Invention

[0006] This application provides a method and device for tracing and classifying defects before and after screen glass frosting. It can solve the problem that due to the strong background scattering on the frosted surface, it is difficult to achieve accurate online defect detection, and it is even more impossible to complete the process stage tracing of defects while detecting them.

[0007] In a first aspect, embodiments of this application provide a method for tracing and classifying defects before and after screen glass frosting, including: Obtain the scattering angle distribution of various local areas on the surface of the screen glass after frosting; Based on the scattering angle distribution of multiple local regions on the screen glass surface, the statistical reference scattering angle distribution of the frosted microstructure is determined; wherein, the statistical reference scattering angle distribution is used to characterize the scattering characteristics of the frosted microstructure under the current frosting process. The scattering angle distribution of each local region is compared with the statistical baseline scattering angle distribution to identify anomalous regions. A morphological signature of the scattering angle distribution of the anomalous regions is then extracted. The anomalous regions are local regions whose scattering angle distribution deviates from the statistical baseline scattering angle distribution. The morphological signature is used to characterize the deviation pattern of the anomalous regions relative to the statistical baseline scattering angle distribution. Based on the morphological signature, the defect generation process corresponding to the abnormal region is determined, and the abnormal region is classified according to the defect generation process and the morphological signature.

[0008] The technical solutions described in this application embodiment have at least the following technical effects: The screen glass defect tracing and grading method provided in this application embodiment characterizes the scattering characteristics of the frosted microstructure under the current frosting process state through a statistical benchmark scattering angle distribution. It separates abnormal regions by analyzing the differences between each local area and the statistical benchmark, eliminating global scattering changes caused by frosting process fluctuations as common-mode components, retaining only local scattering anomalies. This overcomes the overwhelming effect of strong scattering background on defect signals from the frosted surface, significantly improving the reliability of frosted surface defect identification. The deviation morphology of abnormal regions from the statistical benchmark scattering angle distribution is extracted as a morphological signature, and the defect generation process is determined based on the morphological signature. Abnormal regions are graded according to the defect generation process and morphological signature. Utilizing the deterministic modification effect of the frosting process on the defect morphology, it enables the inversion of the process history from the current defect morphology. It can complete process tracing of pre-frosting defects, frosting process defects, and post-frosting defects in a single inspection, eliminating the need for multiple cross-process inspections and defect matching before and after frosting, reducing false rejections and missed releases, and improving the precision of screen glass quality control.

[0009] Secondly, embodiments of this application provide a system for tracing and classifying defects before and after screen glass frosting, including: The acquisition unit is used to acquire the scattering angle distribution of various local areas on the surface of the screen glass after frosting. The distribution unit is used to determine the statistical reference scattering angle distribution of the frosted microstructure based on the scattering angle distribution of multiple local areas on the screen glass surface; wherein, the statistical reference scattering angle distribution is used to characterize the scattering characteristics of the frosted microstructure under the current frosting process state; An extraction unit is used to perform differential analysis on the scattering angle distribution of each local region and the statistical reference scattering angle distribution, separate abnormal regions from each local region, and extract the morphological signature of the scattering angle distribution of the abnormal regions; wherein, the abnormal region is a local region whose scattering angle distribution deviates from the statistical reference scattering angle distribution; the morphological signature is used to characterize the deviation shape of the abnormal region relative to the statistical reference scattering angle distribution; The result unit is used to determine the defect generation process corresponding to the abnormal region based on the morphological signature, and to classify the abnormal region according to the defect generation process and the morphological signature.

[0010] Thirdly, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in the first aspect above.

[0011] Fourthly, embodiments of this application provide a computer program product that, when run on an electronic device, causes the electronic device to perform the method described in the first aspect.

[0012] It is understood that the beneficial effects of the second to fourth aspects mentioned above can be found in the relevant descriptions in the above aspects, and will not be repeated here. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a schematic flowchart of a method for tracing and classifying defects before and after frosting of screen glass according to an embodiment of this application; Figure 2 This is a schematic diagram of the microstructure of the frosted glass surface provided in one embodiment of this application; Figure 3 This is a geometric schematic diagram of scattering angle distribution measurement provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of a screen glass defect tracing and grading system before and after frosting provided in one embodiment of this application; Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0015] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0016] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0017] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0018] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determination" or "if the described condition or event is detected" may be interpreted, depending on the context, as "once determination," "in response to determination," "once the described condition or event is detected," or "in response to the detection of the described condition or event."

[0019] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0020] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0021] Screen glass is a crucial component of electronic devices such as smartphones, tablets, and automotive displays. To reduce screen glare, improve display quality, and minimize fingerprints, screen glass is typically treated with a frosting process, creating a matte finish with random microstructures on its surface. Frosting processes generally include two types: chemical etching and sandblasting. Chemical etching uses hydrofluoric acid-based etching solutions to controllably etch the glass surface, forming random, uneven microstructures on the submicron to micron scale.

[0022] While the frosting process imparts excellent optical properties to screen glass, it also presents challenges in defect detection and quality control. The frosting process involves multiple steps, and defects can arise at different stages: the original sheet may carry defects such as cracks and scratches before frosting; during the frosting process, factors such as fluctuations in etching solution concentration, uneven temperature distribution, and residual etching solution can lead to defects such as uneven etching, white spots, and pitting; post-frosting defects such as scratches, indentations, and stains may be introduced during handling, cleaning, and packaging. The responsibility, reworkability, and handling methods for defects arising at different stages vary. Therefore, tracing the source of defects at each stage and classifying their handling is a crucial foundation for improving screen glass yield and optimizing frosting process parameters.

[0023] However, after frosting, the screen glass surface changes from a smooth mirror surface to a random microstructured frosted surface. The frosted microstructures generate strong diffuse scattering of incident light, forming a scattering background covering the entire surface. Defect signals are submerged in the strong scattering background generated by the frosted microstructures, resulting in extremely low defect contrast obtained by conventional optical detection methods on the frosted surface, making it difficult to reliably identify defects. Existing polarization enhancement schemes can only improve the grayscale contrast between specific targets and the background, and cannot distinguish between frosting process fluctuations and real defects; deep learning-based end-to-end recognition schemes rely on a large number of labeled samples, and the output results lack physical interpretability, failing to indicate the process stage to which the defect belongs; Fourier transform-based detection schemes mainly target the phase difference caused by uneven glass thickness, which is irrelevant to the defect detection problem under the scattering background of the frosted surface.

[0024] Furthermore, the art believes that frosted surfaces, due to their strong background scattering, have long been considered a difficult technical challenge to achieve, making it difficult to accurately detect defects online, let alone trace the source of defects at the process stage while detecting them.

[0025] To address the aforementioned issues, this application provides a method for tracing and classifying defects before and after frosting on screen glass. This method characterizes the scattering features of the frosted microstructure under the current frosting process state using a statistical benchmark scattering angle distribution. Abnormal regions are separated by differential analysis between each local region and the statistical benchmark. This eliminates global scattering changes caused by frosting process fluctuations as common-mode components, retaining only local scattering anomalies. This overcomes the overwhelming effect of strong scattering background on defect signals from the frosted surface, significantly improving the reliability of defect identification. The deviation morphology of abnormal regions from the statistical benchmark scattering angle distribution is extracted as a morphological signature. Based on the morphological signature, the defect-generating process is determined. Abnormal regions are classified according to the defect-generating process and the morphological signature. Utilizing the deterministic modification effect of the frosting process on the defect morphology, this method enables the inversion of the process history from the current defect morphology. It can complete process tracing of pre-frosting defects, frosting process defects, and post-frosting defects in a single inspection, eliminating the need for multiple cross-process inspections and defect matching before and after frosting. This reduces false rejections and missed releases, improving the precision of screen glass quality control.

[0026] The method for tracing and classifying defects before and after frosting of screen glass provided in this application embodiment can be applied to electronic devices. In this case, the electronic device is the executing subject of the method for tracing and classifying defects before and after frosting of screen glass provided in this application embodiment. This application embodiment does not impose any restrictions on the specific type of electronic device.

[0027] For example, electronic devices can be ultra-mobile personal computers (UMPCs), netbooks, desktop computers, computers, laptops, communication equipment, computing devices, satellite wireless equipment, etc.

[0028] To better understand the method for tracing and classifying defects before and after frosting of screen glass provided in this application, the specific implementation process of the method for tracing and classifying defects before and after frosting of screen glass provided in this application will be described in the following exemplary manner.

[0029] Figure 1 This paper illustrates a schematic flowchart of a method for tracing and classifying defects in screen glass before and after frosting, as provided in an embodiment of this application. The method includes: S100: Obtain the scattering angle distribution of various local areas on the screen glass surface after frosting.

[0030] It can be understood that scattering angle distribution refers to the distribution of scattered light intensity in a specific local area on the surface of a screen glass in various receiving angle directions, and it is the basic data for subsequent defect identification and tracing. The scattering characteristics of the frosted microstructure are related to the frosting process. Due to changes in the microstructure topology or scattering mechanism, the scattering angle distribution of defective areas differs from that of normal frosted areas.

[0031] In this embodiment, the frosted screen glass surface is divided into multiple local regions, and the scattering angle distribution of each local region is obtained. The scattering angle distribution uses the receiving angle as the independent variable and the intensity of scattered light in the corresponding receiving angle direction as the dependent variable. The number and interval of receiving angles can be determined according to the detection accuracy and detection cycle.

[0032] The acquisition of the scattering angle distribution relies on the measurement of scattered light intensity from multiple receiver angles. The measurement methods are well-known optical measurement techniques in the field, such as line scan imaging and synchronous acquisition by multi-channel detectors. By acquiring the scattering angle distribution, a complete spatial description of the scattering characteristics of the screen glass surface is established, providing a data foundation for subsequent determination of statistical benchmarks, separation of abnormal regions, and extraction of morphological signatures.

[0033] Optionally, a fixed incident angle light source and a multi-receiving angle detector array can be arranged on the screen glass conveying path to simultaneously collect the intensity of scattered light from multiple receiving angles during the continuous conveying of the screen glass.

[0034] In one possible implementation, step S100 involves obtaining the scattering angle distribution of various local areas on the frosted screen glass surface, including: S110: Acquire multiple sets of scattered light intensity data when a fixed incident angle is used to illuminate the screen glass surface at multiple different receiving angles.

[0035] This is understandable; please refer to [link / reference]. Figure 2 and 3 The fixed incident angle is the angle between the incident light and the normal to the screen glass surface, and it remains constant throughout the detection process to ensure consistent measurement conditions across different local areas and receiving angles. The receiving angle is the angle between the direction of scattered light reception and the normal to the screen glass surface, used to characterize the direction in which the detector collects scattered light. The receiving angle is based on the screen glass surface normal; a receiving angle of 0 degrees corresponds to scattered light collection along the normal direction, and as the receiving angle increases, the receiving direction gradually deviates from the normal. Multiple different receiving angles correspond one-to-one with different scattered light emission directions, and the intensity of scattered light collected at each receiving angle collectively constitutes the sampling points for the scattering angle distribution.

[0036] In this embodiment, a fixed incident angle can eliminate the influence of incident angle variation on the scattering angle distribution, making the scattering angle distributions between different local areas comparable; multi-receiver-angle coverage can capture the complete shape of the scattering angle distribution, providing sufficient sampling points for subsequent differentiation of morphological features such as wide-angle enhancement, narrow-angle enhancement and uniform offset.

[0037] Optionally, the incident angle is fixed at 45 degrees, and the receiving angles include four directions: 0 degrees, 15 degrees, 30 degrees, and 45 degrees. The 45-degree incident angle takes into account both specular reflection and diffuse scattering components in the detection of frosted surfaces; the receiving angle range of 0 to 45 degrees covers the main energy distribution range of frosted surface scattering, and the scattered signal at each receiving angle within this range has sufficient intensity, avoiding the amplification of measurement noise caused by weak signals at excessively large receiving angles.

[0038] S120, determines the scattering angle distribution of each local area on the screen glass surface based on all scattered light intensity data.

[0039] It can be understood that scattered light intensity data is the raw measurement data, while the scattering angle distribution is a data form obtained by organizing and arranging the raw measurement data. Based on multiple sets of scattered light intensity data, the scattered light intensity measured in each local area in each receiving angle direction is organized according to the receiving angle order, thus obtaining the scattering angle distribution of that local area.

[0040] In this embodiment, the scattering angle distribution is stored in the form of a sequence or vector, where each element in the sequence corresponds to the intensity of scattered light in a receiving angle direction. The scattering angle distribution serves as the standard input for subsequent differential analysis, unifying the data structure for each local region.

[0041] Optionally, the scattering angle distribution of each local area can also be normalized to eliminate the influence of light source intensity fluctuations on the amplitude of the scattering angle distribution.

[0042] This setup allows for the acquisition of scattered light intensity data and the determination of the scattering angle distribution by combining a fixed incident angle with multiple receiving angles, providing structurally consistent and conditionally uniform scattering angle distribution data for subsequent differential analysis.

[0043] S200, based on the scattering angle distribution of multiple local areas on the screen glass surface, determines the statistical reference scattering angle distribution of the frosted microstructure; wherein, the statistical reference scattering angle distribution is used to characterize the scattering characteristics of the frosted microstructure under the current frosting process.

[0044] It is understandable that the statistical baseline scattering angle distribution is a baseline distribution obtained by statistically fusing the scattering angle distributions of multiple local areas on the screen glass surface, reflecting the typical scattering behavior of the frosted microstructure under the current process parameters. Fluctuations in the frosting process (such as etching solution concentration drift and temperature changes) will cause changes in the statistical parameters of the frosted microstructure, but this change is a global change affecting the entire screen glass. The scattering mechanism of the frosted microstructure does not change, and the shape of the scattering angle distribution remains basically consistent; only the scattering intensity shifts overall.

[0045] In this embodiment, the statistical reference scattering angle distribution is determined by statistically analyzing the scattering angle distribution of multiple local regions on the screen glass surface. The scattering angle distributions of multiple local regions covering different positions on the screen glass surface are selected and statistically fused to obtain a reference distribution that can represent the typical scattering behavior of the frosted microstructure under the current frosting process.

[0046] The statistical baseline scattering angle distribution is essentially an internal reference for the screen glass itself, rather than an external standard sample. By using the statistical results of multiple local areas of the screen glass itself as a baseline, it can adaptively reflect the scattering characteristics of the frosted microstructure under the current process conditions, eliminating the baseline mismatch problem caused by differences in frosted microstructure under different batches and process parameters.

[0047] Optionally, statistical fusion can be achieved by taking the arithmetic mean or the median from each angle. The specific method can be determined based on the distribution characteristics of the scattering intensity data.

[0048] In one possible implementation, in step S200, determining the statistical reference scattering angle distribution of the frosted microstructure based on the scattering angle distribution of multiple local regions on the screen glass surface includes: S210, based on the scattering angle distribution of multiple local areas on the screen glass surface, a preliminary detection is performed on the screen glass surface, and normal areas among the local areas where the proportion of abnormal area detected in the preliminary detection is lower than a preset first proportion threshold are selected as candidate areas.

[0049] Understandably, preliminary detection is used to identify local areas on the screen glass surface where the scattering characteristics significantly deviate from the normal range, in order to exclude these areas from contaminating the statistical benchmark. The proportion of the area of ​​the preliminarily detected abnormal area to the total area of ​​the screen glass surface is calculated to obtain the preliminarily detected abnormal area proportion; when the preliminarily detected abnormal area proportion is lower than a preset first proportion threshold, it indicates that most areas of the screen glass surface are normal frosted areas, and normal areas among the local areas of these normal areas are selected as candidate areas.

[0050] Preliminary detection is not used to accurately locate the specific type of defect in a local area, but only to determine whether the scattering characteristics of the local area deviate from the normal range of frosted sand scattering characteristics. It has the advantages of low computational load and fast screening speed.

[0051] For example, preliminary detection can be performed by extracting at least one scattering feature parameter from the scattering angle distribution of each local region. The scattering feature parameter includes at least one of peak scattering angle, peak scattering intensity, full width at half maximum (FWHM), and scattering intensity integral. The scattering feature parameter is compared with a preset range of normal scattering feature parameters. When the scattering feature parameter exceeds the preset range of normal scattering feature parameters, the local region is identified as an abnormal region in the preliminary detection.

[0052] In this embodiment, the candidate region serves as the data source for the statistical benchmark, and the purity of the candidate region directly affects the accuracy of the statistical benchmark. Preliminary detection eliminates abnormal regions, preventing large-area defects or process anomalies from contaminating the statistical benchmark and ensuring the accuracy of the statistical benchmark in characterizing the scattering features of normal frosted microstructures.

[0053] Optionally, the preset first percentage threshold can be set to 20%. When the percentage of abnormal area detected in the initial test is less than 20%, the screen glass surface is mainly composed of normal frosted areas, and the normal frosted areas are used as candidate areas.

[0054] S220: The statistical reference scattering angle distribution is obtained by taking the statistical center value of the scattering angle distribution of multiple local regions in the candidate region for each angle.

[0055] It can be understood that taking the statistical center value for each angle is a statistical fusion method. It involves taking the statistical center value of the scattering intensity values ​​of multiple local regions along each receiving angle direction, and organizing these statistical center values ​​according to the receiving angle order to obtain the statistical baseline scattering angle distribution. The statistical center value can average out local random fluctuations across multiple local regions, extracting the common characteristics of scattering from the frosted microstructure.

[0056] In this embodiment of the application, when the scattering intensity data is approximately normally distributed, the arithmetic mean can be used as the statistical center value, and when outliers exist, the median can be used as the statistical center value.

[0057] Optionally, the selection of the statistical center value can be dynamically adjusted based on the number of candidate regions. When the number of candidate regions is small, the median can be used to enhance robustness.

[0058] With this setup, by initially detecting and screening candidate regions and taking the statistical center value at each angle, a statistical baseline scattering angle distribution characterizing the typical scattering behavior of normal frosted microstructures is obtained, providing a robust internal reference benchmark for differential analysis.

[0059] In one possible implementation, step S200, determining the statistical reference scattering angle distribution of the frosted microstructure based on the scattering angle distribution of multiple local regions on the screen glass surface, further includes: S230: When the percentage of abnormal area detected in the initial test is greater than the preset second percentage threshold, the corresponding screen glass is determined to be a glass with a process defect and the subsequent testing steps for the screen glass are stopped.

[0060] It is understandable that when the proportion of abnormal area detected in the initial test is greater than the preset second proportion threshold, it indicates that there is a large-scale abnormality on the surface of the screen glass. At this time, the statistical benchmark can no longer represent the scattering characteristics of the normal frosted microstructure. Continuing to perform subsequent abnormal area separation, morphology signature extraction and process determination will lose a reliable benchmark. Therefore, the screen glass as a whole is identified as a process abnormal glass, and further testing of the screen glass is directly stopped.

[0061] In this embodiment, large-scale anomalies are usually caused by overall abnormalities in the frosting process, such as etching solution failure or severe drift in process parameters. In this case, the overall quality of the screen glass is substandard, and there is no need to trace and classify each defect individually. The preset second percentage threshold should be greater than the preset first percentage threshold, for example, it can be set to 50%.

[0062] Optionally, when the screen glass is determined to be defective, a frosting process alarm can be triggered, prompting operators to check the frosting process parameters and locate the cause of the defect from the source of the process.

[0063] This setup, through the overall judgment of large-scale anomalies, avoids continuing to execute invalid detection processes under the condition of baseline failure, while providing an overall alarm entry point for sand frosting process anomalies, thus improving the robustness of the method under extreme operating conditions.

[0064] S300 performs a differential analysis between the scattering angle distribution of each local region and the statistical reference scattering angle distribution, separates the abnormal regions from each local region, and extracts the morphological signature of the scattering angle distribution of the abnormal regions; wherein, the abnormal region is a local region whose scattering angle distribution deviates from the statistical reference scattering angle distribution; the morphological signature is used to characterize the deviation shape of the abnormal region relative to the statistical reference scattering angle distribution.

[0065] It is understandable that differential analysis is performed at the angle-by-angle level of the scattering angle distribution. By comparing the differences in scattering intensity between each local region and the statistical baseline in each receiving angle direction, local regions that deviate from the statistical baseline are identified and separated into anomalous regions. Anomalous regions are candidate regions for potential defects, and morphological signatures are quantitative representations of the deviations in morphology of anomalous regions, carrying pattern information about changes in the scattering mechanism of anomalous regions.

[0066] In this embodiment, the scattering changes caused by fluctuations in the frosting process are consistent across the entire screen glass and belong to the common-mode component; the scattering changes caused by defects are limited to the local area where the defect is located and belong to the differential-mode component. Differential analysis uses a statistical benchmark as a reference to eliminate the common-mode component from each local area, making the differential-mode component, i.e., the abnormal area, stand out. This is the fundamental reason why this method can overcome the strong scattering background of the frosted surface.

[0067] Morphological signatures characterize the deviation of anomaly regions from the statistical baseline scattering angle distribution, i.e., the distribution pattern of the deviation of the anomaly region's scattering angle distribution from the statistical baseline scattering angle distribution in each receiving angle direction. Defects generated at different process stages have different morphological characteristics: defects in the original sheet before frosting (such as cracks), after preferential penetration and etching by the etching solution, form deep and narrow tunnel-like etch pits, whose scattering angle distribution shows a wide-angle morphology with enhanced large-angle scattering; defects introduced after frosting (such as scratches) damage the frosting microstructure, exposing the glass body, whose scattering angle distribution shows a narrow-angle morphology with enhanced small-angle specular scattering; defects in the frosting process (such as uneven etching) cause the statistical parameters of the frosting microstructure to drift, but the scattering mechanism remains, and its scattering angle distribution is consistent with the statistical baseline shape, with only the intensity shifting. Morphological signatures are precisely the quantitative characterization of the above deviation morphologies.

[0068] Optionally, differential analysis and morphological signature extraction can be performed sequentially in the same process flow, or they can be performed in stages.

[0069] In one possible implementation, in step S300, the scattering angle distribution of each local region is compared with the statistical baseline scattering angle distribution to separate abnormal regions from each local region, and the morphological signature of the scattering angle distribution of the abnormal regions is extracted, including: S310 performs angle-by-angle difference analysis between the scattering angle distribution of each local region and the statistical baseline scattering angle distribution, and separates the abnormal regions from each local region.

[0070] As can be understood, angle-by-angle differential analysis compares the scattering intensity of a local region with the scattering intensity of a statistical baseline in each receiving angle direction to obtain angle-by-angle difference information. Based on this angle-by-angle difference information, it is determined whether the local region deviates from the statistical baseline. Compared to overall intensity comparison, angle-by-angle comparison can capture the deviation characteristics of a local region in a specific angular direction, improving the sensitivity to changes in the defect scattering mechanism.

[0071] In this embodiment, the angle-by-angle differential analysis preserves the angle-by-angle information of the scattering angle distribution, and separates the abnormal regions where the scattering angle distribution deviates from the statistical benchmark from each local region, thereby realizing the location of the defect candidate region.

[0072] Optionally, in step S310, an angle-by-angle difference analysis is performed on the scattering angle distribution of each local region and the statistical baseline scattering angle distribution to separate out anomalous regions from each local region, including: S311, the scattering angle distribution of each local area is compared with the statistical reference scattering angle distribution at the corresponding receiving angle, and the scattering difference value of each receiving angle is obtained.

[0073] It can be understood that, for each receiving angle direction, the difference between the scattering intensity of the local area and the scattering intensity of the statistical reference scattering angle distribution in that receiving angle direction is calculated to obtain the scattering difference value for that receiving angle. A positive scattering difference value indicates that the scattering intensity of the local area in that receiving angle direction is higher than the statistical reference, while a negative value indicates that it is lower than the statistical reference.

[0074] In this embodiment of the application, angle-by-angle difference is a specific implementation method of differential analysis. By calculating the difference, the deviation of each local region from the statistical benchmark is quantified to each receiving angle direction, resulting in a set of scattering difference values ​​containing angle-by-angle deviation information.

[0075] S312, when the absolute value of the scattering difference value of a local area at any receiving angle is greater than the preset difference threshold, the local area is separated into an abnormal area.

[0076] It is understood that for each local region, the absolute value of the scattering difference at each receiving angle is calculated. When the absolute value of the scattering difference at any receiving angle is greater than a preset difference threshold, the local region is determined to deviate from the statistical benchmark and is separated into an abnormal region. An absolute value of the scattering difference greater than the preset difference threshold indicates that the scattering behavior of the local region in that receiving angle direction significantly deviates from the typical scattering behavior of normal frosted microstructures, which may be caused by defects.

[0077] In this embodiment, the preset difference threshold can be set according to the detection sensitivity requirements. The higher the sensitivity requirements, the smaller the preset difference threshold should be.

[0078] Optionally, the preset difference threshold can also be adaptively determined based on the measurement noise level of the statistical reference scattering angle distribution, so as to balance detection sensitivity and false detection rate.

[0079] This setup, by calculating the difference from angle to angle and comparing it with a threshold, quantitatively separates abnormal regions, ensuring the repeatability and objectivity of defect candidate region identification.

[0080] S320, extract the scattering angle distribution of the abnormal region, and determine the morphological signature of the abnormal region based on the deviation of the scattering angle distribution of the abnormal region from the statistical baseline scattering angle distribution.

[0081] It can be understood that the morphological signature characterizes the deviation pattern of the abnormal region from the statistical reference scattering angle distribution, that is, the distribution pattern of the deviation of the abnormal region's scattering angle distribution from the statistical reference scattering angle distribution in each receiving angle direction. It carries information about how the scattering mechanism of the abnormal region changes and is the core basis for process inversion.

[0082] In this embodiment of the application, by extracting the morphological signature, the deviation morphology of the abnormal region is quantified into feature data that can be used for process inversion, thus establishing a bridge between the current morphology of the defect and the process that generates the defect.

[0083] Optionally, S320, extract the scattering angle distribution of the anomalous region, and determine the morphological signature of the anomalous region based on the deviation of the scattering angle distribution of the anomalous region from the statistical reference scattering angle distribution, including: S321, extract the scattering angle distribution of the abnormal region, determine the scattering difference value of each receiving angle in the abnormal region relative to the statistical reference scattering angle distribution, and construct the scattering difference sequence of the abnormal region.

[0084] It can be understood that the scattering angle distribution corresponding to the abnormal area is extracted, and the difference between the scattering intensity of the abnormal area and the scattering intensity of the statistical reference scattering angle distribution in the receiving angle direction is calculated in each receiving angle direction. The scattering difference value under each receiving angle is obtained, and the scattering difference value under each receiving angle is arranged in the order of receiving angle to form a scattering difference sequence.

[0085] In this embodiment, the scattering difference sequence is a data form of morphological signature. Each element in the sequence corresponds to a scattering difference value in the direction of the receiving angle. The arrangement of the elements in the sequence with the change of the receiving angle carries all the information of the deviation morphology.

[0086] S322, the scattering difference sequence is identified as the morphological signature of the anomalous region.

[0087] It is understandable that treating the entire scattering difference sequence as a morphological signature of the anomalous region provides a clear data definition for the morphological signature, facilitating direct reference in subsequent steps and avoiding conceptual ambiguity. Subsequent inversion processes are then performed based on this morphological signature.

[0088] In this embodiment of the application, the scattering difference sequence fully preserves the angle-by-angle information of the deviation morphology, enabling the morphology signature to distinguish different deviation morphologies such as wide-angle enhancement, narrow-angle enhancement, and uniform offset.

[0089] Optionally, the morphological signature can also be obtained by normalizing the scattering difference sequence to eliminate the influence of defect size differences on the signature amplitude.

[0090] This setup, by defining the morphology signature using the scattering difference sequence, provides a clear and executable data definition for the morphology signature, and fully preserves the angle-by-angle information of the deviation from the morphology, providing a basis for process inversion.

[0091] S400: Determine the defect-generating process corresponding to the abnormal area based on the shape signature, and classify the abnormal area according to the defect-generating process and the shape signature.

[0092] This step is understood to facilitate process tracing and categorized handling of defects. Defect-causing processes include defects in the original sheet before frosting, defects in the frosting process itself, and defects introduced after frosting. The determination of the defect-causing process is based on the deviation morphology represented by the morphological signature, and is corrected by incorporating directional characteristics. Categorized handling is determined by combining the reworkability of the defect-causing process with information such as morphological signature, area, and location.

[0093] In this embodiment, the determination of the defect generation process utilizes the modification effect of the frosting process on the defect morphology. The frosting process is regarded as a recorder of the defect prehistory: before frosting, the defect is etched and amplified; the abnormal performance of the frosting process is statistical drift; after frosting, the defect destroys the microstructure. The three modification laws leave identifiable traces on the morphology. By reading these traces, the process inversion can be achieved.

[0094] Optionally, the grading results can be overlaid with the defect distribution map of the screen glass for easy viewing by operators to visually identify the process attribution and treatment level of the defects.

[0095] In one possible implementation, S400, the defect-generating process corresponding to the abnormal region is determined based on the morphological signature, and the abnormal region is classified according to the defect-generating process and the morphological signature, including: S410, the first scattering angle distribution of the screen glass surface illuminated by the first incident orientation under multiple receiving angles, and the second scattering angle distribution of the screen glass surface illuminated by the second incident orientation under multiple receiving angles are obtained respectively; wherein, the first incident orientation and the second incident orientation are orthogonal.

[0096] It can be understood that the first and second incident orientations are two mutually perpendicular orientations within the screen glass surface. The screen glass surface is illuminated from the first and second incident orientations, and scattered light intensity data is acquired at multiple receiving angles to determine the corresponding first and second scattering angle distributions. The frosted microstructure is statistically isotropic, and its scattering behavior of incident light is independent of the incident orientation. However, the scattering behavior of linear directional defects (such as scratches and cracks) is related to the incident orientation; the scattering angle distribution differs significantly when incident along the defect direction versus perpendicular to the defect direction.

[0097] In this embodiment, by comparing the scattering angle distribution of orthogonal incident directions, the directional characteristics of defects can be identified, and linear directional defects can be distinguished from planar defects. This is an important feature that distinguishes this method from the prior art.

[0098] Optionally, the first incident orientation and the second incident orientation can be set to the 0-degree orientation and the 90-degree orientation of the screen glass surface, respectively.

[0099] S420, determine the directional characteristics of the anomalous region based on the difference between the first scattering angle distribution and the second scattering angle distribution.

[0100] It is understandable that directional features characterize the degree of difference in scattering behavior of abnormal regions under orthogonal incident orientation, and are related to the geometric morphology of defects: linear directional defects (scratches, cracks) extend along a specific direction, and their scattering behavior is sensitive to the incident orientation; planar defects (irregular etching, white spots) are distributed in a planar manner on the surface, and their scattering behavior is not sensitive to the incident orientation.

[0101] In this embodiment, directional characteristics are determined by the difference in scattering angle distribution of orthogonal incident directions, thereby realizing the discrimination between linear and planar geometric morphologies of defects and providing auxiliary criteria for the accurate determination of defect generation processes.

[0102] Optionally, S420, the directional characteristics of the anomalous region are determined based on the difference between the first scattering angle distribution and the second scattering angle distribution, including: S421, the first scattering angle distribution and the second scattering angle distribution are subtracted angle by angle at the corresponding receiving angles to obtain the azimuth scattering difference value of each receiving angle.

[0103] It can be understood that, in each receiving angle direction, the difference in scattering intensity between the first scattering angle distribution and the second scattering angle distribution in the corresponding receiving angle direction is calculated to obtain the azimuth scattering difference value for each receiving angle. The azimuth scattering difference value characterizes the degree of difference in scattering behavior of the anomalous region under two orthogonal incident azimuths.

[0104] In this embodiment, the difference between the scattering angles of two orthogonal directions is quantified to each receiving angle direction by calculating the difference angle by angle, providing quantitative data for determining the directional characteristics.

[0105] S422, determine the directional characteristics of the anomalous region based on the scattering difference values ​​of all directions.

[0106] It is understandable that the directional characteristics of the abnormal region are comprehensively determined based on the azimuth scattering differences at each receiving angle. When the overall azimuth scattering difference is significant, the abnormal region has a linear directional characteristic, indicating that the defect extends in a linear manner; when the overall azimuth scattering difference is not significant, the abnormal region has a planar characteristic, indicating that the defect is distributed in a planar manner.

[0107] In this embodiment of the application, whether the azimuth scattering difference value is significant can be determined by comparing the azimuth scattering difference value with a preset azimuth difference threshold, or by comprehensively considering the statistical characteristics of the azimuth scattering difference values ​​at each receiving angle.

[0108] Optionally, directional features can also be represented by two levels of discrete values: linear directional features and area features, which are convenient for direct reference in subsequent processes.

[0109] This setup determines directional characteristics by the difference in scattering angle distribution from orthogonal incident directions, enabling the differentiation between linear and planar geometric morphologies of defects and providing auxiliary criteria for process judgment.

[0110] S430, based on morphological signatures and directional features, determines the defect-generating process corresponding to the abnormal area.

[0111] It is understandable that the defect generation process corresponding to the abnormal region is comprehensively determined based on the deviation morphology represented by the morphological signature and the defect geometry represented by the directional features. The deviation morphology provides information on the change pattern of the scattering mechanism, and the directional features provide information on the defect geometry. The joint determination of the two eliminates the classification ambiguity of a single criterion.

[0112] In this embodiment, the rule for determining the defect generation process can be as follows: When the morphological signature of the abnormal region shows a wide-angle enhancement pattern with a large receiving angle scattering difference value greater than a small receiving angle scattering difference value, and has linear directional characteristics, the defect generation process is determined to be a defect in the original sheet before frosting. This is because before frosting, the crack is preferentially penetrated and etched by the etching solution, forming a deep and narrow tunnel-like etch pit. The deep and narrow structure produces multiple reflections of the incident light, showing large-angle scattering enhancement, and the crack extends linearly with directionality; when ... When the narrow-angle enhancement morphology of the scattering difference value at the angle of convergence, and the linear directional characteristics are present, the defect is determined to be introduced by the frosting process. This is because scratches after frosting destroy the microstructure of the frosting, exposing the glass body to form a mirror surface. The mirror surface produces near-mirror reflection of the incident light, which manifests as enhanced scattering at a small angle. Moreover, the scratches extend in a linear manner and have directionality. When the abnormal area has planar characteristics, the defect is determined to be caused by a frosting process defect. This is because uneven etching and frosting process defects cause the statistical parameters of the frosting microstructure to drift, the scattering mechanism is maintained, and the defect is distributed in a planar manner and does not have directionality.

[0113] The aforementioned determination rules utilize the modification effect of the frosting process on defect morphology, inverting the current defect morphology into the defect-generating process, which is the core mechanism for achieving process traceability in this method. This mechanism does not rely on pre-frosting inspection data and can complete process traceability in a single inspection after frosting, significantly reducing inspection costs and process complexity.

[0114] Optionally, in other possible implementations, the determination of the defect generation process can also employ a machine learning classifier, using morphological signatures and directional features as input features, and outputting the classification result of the defect generation process. The machine learning classifier is pre-trained using labeled samples.

[0115] For ease of understanding, the table below provides examples of the morphological signatures and directional characteristics of three typical defect types (the values ​​are scattering differences relative to the statistical baseline, in au, for illustrative purposes only): This setup, through the joint determination of morphological signature and directional features, enables the accurate identification of the defect-generating process, transforming the modification law of defect morphology by the frosting process into an executable process inversion rule.

[0116] In one possible implementation, step S400, determining the defect generation process corresponding to the abnormal region based on the morphological signature, and classifying the abnormal region according to the defect generation process and the morphological signature, further includes: S440 classifies abnormal areas based on the defect-causing process and morphological signature.

[0117] It is understandable that the reworkability of abnormal areas is determined based on the process in which the defects occurred. Defects from the frosting process have a high reworkability and can be reworked by re-frosting or local etching. Defects on the original wafer before frosting have a low reworkability and usually require scrapping or downgrading. The reworkability of defects introduced after frosting depends on the depth and location of the defect; shallow defects can be reworked by polishing, while deep defects are usually not reworkable. Combining the reworkability with information such as the morphology, area, and location of the abnormal area, the treatment level of the abnormal area is determined. The treatment level includes release, downgrading, rework, and scrapping.

[0118] In this embodiment of the application, the grading rules can be as follows: For reworkable abnormal areas, the treatment level is determined between rework and downgrading. The specific level is determined according to the area, morphological signature strength and location of the abnormal area. For example, defects located in the edge area of ​​the screen glass can be downgraded, while defects located in the visible area need to be reworked. For non-reworkable abnormal areas, the treatment level is determined between scrap, downgrading and release. For example, defects with large area, high morphological signature strength and located in the visible area are scrapped, while defects with small area and located in the non-visible area can be downgraded or released.

[0119] Introducing the defect-generating process into a tiered handling system ensures that the handling method matches the reworkability of the defect. This avoids the mistaken scrapping of reworkable defects and the wrong release of non-reworkable defects due to a general scoring approach, thereby improving the rationality of handling decisions.

[0120] Optionally, the grading results can be overlaid with the defect distribution map of the screen glass, making it easier for operators to intuitively view the process attribution and treatment level of the defects; the treatment level can also be associated with the unique identifier of the screen glass for subsequent quality traceability.

[0121] This setup, by introducing the defect-generating process into a tiered handling system, achieves a match between defect handling methods and defect reworkability, reducing false rejections and missed releases, and improving the precision of screen glass quality control.

[0122] Corresponding to the screen glass defect tracing and classification method before and after frosting in the above embodiments, this application also provides a screen glass defect tracing and classification system before and after frosting. Each unit of the system can implement each step of the screen glass defect tracing and classification method before and after frosting. Figure 4 The diagram shows a structural block diagram of the screen glass defect tracing and grading system before and after frosting provided in the embodiments of this application. For ease of explanation, only the parts related to the embodiments of this application are shown.

[0123] Reference Figure 4 The system for tracing and classifying defects before and after frosting of the screen glass includes: The acquisition unit is used to acquire the scattering angle distribution of various local areas on the surface of the screen glass after frosting. The distribution unit is used to determine the statistical reference scattering angle distribution of the frosted microstructure based on the scattering angle distribution of multiple local areas on the screen glass surface; wherein, the statistical reference scattering angle distribution is used to characterize the scattering characteristics of the frosted microstructure under the current frosting process state; An extraction unit is used to perform differential analysis on the scattering angle distribution of each local region and the statistical reference scattering angle distribution, separate abnormal regions from each local region, and extract the morphological signature of the scattering angle distribution of the abnormal regions; wherein, the abnormal region is a local region whose scattering angle distribution deviates from the statistical reference scattering angle distribution; the morphological signature is used to characterize the deviation shape of the abnormal region relative to the statistical reference scattering angle distribution; The result unit is used to determine the defect generation process corresponding to the abnormal region based on the morphological signature, and to classify the abnormal region according to the defect generation process and the morphological signature.

[0124] It should be noted that the information interaction and execution process between the above systems / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0125] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit module can exist physically separately, or two or more unit modules can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0126] This application also provides an electronic device. Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 5 As shown, the electronic device 6 of this embodiment includes: at least one processor 60 ( Figure 5 Only one is shown in the image), at least one memory 61 ( Figure 5 (Only one is shown in the image) and a computer program 62 stored in the at least one memory 61 and executable on the at least one processor 60. When the processor 60 executes the computer program 62, it causes the electronic device 6 to perform the steps in any of the above embodiments of the method for tracing and classifying defects before and after screen glass frosting, or causes the electronic device 6 to perform the functions of each unit in the above system embodiments.

[0127] For example, the computer program 62 may be divided into one or more units, which are stored in the memory 61 and executed by the processor 60 to complete this application. The one or more units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program 62 in the electronic device 6.

[0128] Electronic device 6 can be a computing device or terminal device such as a desktop computer, laptop, handheld computer, or cloud server. This electronic device may include, but is not limited to, a processor 60 and a memory 61. Those skilled in the art will understand that... Figure 5 This is merely an example of electronic device 6 and does not constitute a limitation on electronic device 6. It may include more or fewer components than shown, or combine certain components, or different components, such as input / output devices, network access devices, buses, etc.

[0129] The processor 60 can be a Central Processing Unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0130] In some embodiments, the memory 61 may be an internal storage unit of the electronic device 6, such as a hard disk or memory of the electronic device 6. In other embodiments, the memory 61 may be an external storage device of the electronic device 6, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the electronic device 6. Furthermore, the memory 61 may include both internal and external storage units of the electronic device 6. The memory 61 is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory 61 can also be used to temporarily store data that has been output or will be output.

[0131] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.

[0132] This application provides a computer program product that, when run on an electronic device, causes the electronic device to perform the steps in any of the above method embodiments.

[0133] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to an electronic device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0134] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0135] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0136] In the embodiments provided in this application, it should be understood that the disclosed screen glass defect tracing and grading system / electronic device and method before and after frosting can be implemented in other ways. For example, the embodiments of the screen glass defect tracing and grading system / electronic device before and after frosting described above are merely illustrative. For instance, the division of units is merely a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0137] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0138] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for tracing and classifying defects before and after screen glass frosting, characterized in that, include: Obtain the scattering angle distribution of various local areas on the surface of the screen glass after frosting; Based on the scattering angle distribution of multiple local regions on the screen glass surface, the statistical reference scattering angle distribution of the frosted microstructure is determined; wherein, the statistical reference scattering angle distribution is used to characterize the scattering characteristics of the frosted microstructure under the current frosting process. The scattering angle distribution of each local region is compared with the statistical baseline scattering angle distribution to identify anomalous regions. A morphological signature of the scattering angle distribution of the anomalous regions is then extracted. The anomalous regions are local regions whose scattering angle distribution deviates from the statistical baseline scattering angle distribution. The morphological signature is used to characterize the deviation pattern of the anomalous regions relative to the statistical baseline scattering angle distribution. Based on the morphological signature, the defect generation process corresponding to the abnormal region is determined, and the abnormal region is classified according to the defect generation process and the morphological signature.

2. The method as described in claim 1, characterized in that, The acquisition of the scattering angle distribution of various local areas on the frosted screen glass surface includes: Acquire multiple sets of scattered light intensity data when a fixed incident angle is used to illuminate the screen glass surface at multiple different receiving angles; The scattering angle distribution of each local region on the screen glass surface is determined based on all the scattered light intensity data.

3. The method as described in claim 1, characterized in that, The step of determining the statistical reference scattering angle distribution of the frosted microstructure based on the scattering angle distribution of multiple local regions on the screen glass surface includes: Based on the scattering angle distribution of multiple local areas on the screen glass surface, a preliminary detection is performed on the screen glass surface, and normal areas among the local areas where the proportion of abnormal area detected in the preliminary detection is lower than a preset first proportion threshold are selected as candidate areas. The statistical baseline scattering angle distribution is obtained by taking the statistical center value of the scattering angle distribution of multiple local regions in the candidate region angle by angle.

4. The method as described in claim 3, characterized in that, The step of determining the statistical reference scattering angle distribution of the frosted microstructure based on the scattering angle distribution of multiple local regions on the screen glass surface further includes: When the percentage of the initially detected abnormal area is greater than the preset second percentage threshold, the corresponding screen glass is determined to be a glass with a process defect and subsequent detection steps on the screen glass are stopped.

5. The method as described in claim 1, characterized in that, The step of performing a differential analysis between the scattering angle distribution of each local region and the statistical baseline scattering angle distribution, separating abnormal regions from each local region, and extracting the morphological signature of the scattering angle distribution of the abnormal regions includes: An angle-by-angle difference analysis was performed on the scattering angle distribution of each local region and the statistical benchmark scattering angle distribution to separate out abnormal regions from each local region; Extract the scattering angle distribution of the abnormal region, and determine the morphological signature of the abnormal region based on the deviation of the scattering angle distribution of the abnormal region from the statistical baseline scattering angle distribution.

6. The method as described in claim 5, characterized in that, The step of performing angle-by-angle difference analysis between the scattering angle distribution of each local region and the statistical baseline scattering angle distribution, and separating out abnormal regions from each local region, includes: The scattering angle distribution of each local area is compared with the statistical reference scattering angle distribution at the corresponding receiving angle to obtain the scattering difference value of each receiving angle. When the absolute value of the scattering difference in the local region at any receiving angle is greater than a preset difference threshold, the local region is separated into an abnormal region.

7. The method as described in claim 5, characterized in that, The step of extracting the scattering angle distribution of the anomalous region and determining the morphological signature of the anomalous region based on the deviation of the scattering angle distribution of the anomalous region from the statistical benchmark scattering angle distribution includes: Extract the scattering angle distribution of the abnormal region, determine the scattering difference value of the abnormal region relative to the statistical reference scattering angle distribution at each receiving angle, and construct the scattering difference sequence of the abnormal region; The scattering difference sequence is determined as the morphological signature of the anomalous region.

8. The method as described in claim 1, characterized in that, The step of determining the defect generation process corresponding to the abnormal region based on the morphological signature, and classifying the abnormal region according to the defect generation process and the morphological signature, includes: The first scattering angle distribution of the screen glass surface illuminated by the first incident orientation under multiple receiving angles, and the second scattering angle distribution of the screen glass surface illuminated by the second incident orientation under multiple receiving angles are obtained respectively; wherein the first incident orientation and the second incident orientation are orthogonal to each other; The directional characteristics of the abnormal region are determined based on the difference between the first scattering angle distribution and the second scattering angle distribution; Based on the morphological signature and the directional features, the defect generation process corresponding to the abnormal region is determined.

9. The method as described in claim 8, characterized in that, Determining the directional characteristics of the abnormal region based on the difference between the first scattering angle distribution and the second scattering angle distribution includes: The first scattering angle distribution and the second scattering angle distribution are subtracted angle by angle at the corresponding receiving angle to obtain the azimuth scattering difference value for each receiving angle; The directional characteristics of the anomalous region are determined based on all the aforementioned azimuth scattering difference values.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1 to 9.

Images