A panel defect point aggregation detection method, system, device and storage medium

Abstract

The application provides a panel defect point aggregation detection method, system, device and storage medium, and relates to the technical field of defect detection. The method comprises the following steps: acquiring characteristic data of a panel, wherein the characteristic data comprises coordinate information of all defect points on the surface of the panel; performing clustering and division processing on the characteristic data of the panel by using a density-based clustering algorithm to obtain a plurality of clustering regions; performing region merging processing on the plurality of clustering regions to obtain at least one merged region; and comprehensively detecting the defect points of the panel based on the clustering regions and the merged region to obtain a defect point aggregation detection result. The density-based clustering algorithm is applied to the industrial panel defect detection, and the defect point aggregation condition can be more intuitively considered by introducing the defect point merging adjustment between clusters and the total number threshold judgment of the defect points in the cluster, so that the accuracy and reliability of the defect point aggregation detection are improved.

Description

Technical Field

[0001] This invention relates to the field of defect detection technology, and more specifically, to a method, system, device, and storage medium for detecting clustered defect points on a panel. Background Technology

[0002] Panel defect clustering refers to the appearance of a certain number of unevenly sized defective points on the panel surface during the industrial panel manufacturing process. These defects often exhibit a regional tendency, clustering in specific areas. If the clustering of defective points cannot be accurately observed, it is very easy to miss detections, increasing the risk of mass production of defective products. Therefore, how to ensure accurate detection of defective point clusters and reduce the risk of mass production of defective products is a core problem that urgently needs to be solved in this field.

[0003] Existing technologies for detecting clustered defects on panel surfaces mainly include machine vision and clustering methods. Machine vision can identify and initially locate defects from captured images. Clustering methods can further identify clusters of defect points. These technologies have achieved certain detection results, enabling relatively fast and effective detection of clustered defects, but some problems remain. For example, image quality is affected by factors such as lighting, reflection, and shadows, which may influence the defect detection results. Clustering methods still face challenges in scenarios with complex defect point distributions on panel surfaces.

[0004] In addition, existing technologies have the following problems when detecting defect clusters: (1) Lack of robustness: The distribution of defect points on current panels is very complex. Panels have clusters with different shapes, sizes and densities. Existing methods are not very robust to all defect situations and are only applicable to specific defect situations. (2) Lack of intuitiveness: Current technologies usually involve complex algorithms and data processing, and lack the ability to directly manipulate data. Summary of the Invention

[0005] This invention provides a method, system, device, and storage medium for detecting clustered defects in panels, which solves the problems of lack of robustness and intuitiveness in the detection of clustered defects in existing technologies.

[0006] In a first aspect, embodiments of the present invention provide a method for detecting clustered defects in a panel, the method comprising the following steps:

[0007] Obtain characterization data of the panel, the characterization data including coordinate information of all defect points on the panel surface;

[0008] A density-based clustering algorithm was used to cluster the panel representation data to obtain several cluster regions.

[0009] performing region merging processing on the plurality of clustering regions to obtain at least one merged region;

[0010] performing comprehensive detection on the panel defect points based on the clustering regions and the merged region to obtain a defect point aggregation detection result.

[0011] In the above embodiment, by applying the density-based clustering algorithm to industrial panel defect detection, and by introducing inter-cluster defect point merging adjustment and intra-cluster defect point total number threshold judgment, the aggregation situation can be more intuitively considered, and the accuracy and reliability of defect point aggregation detection are improved. Compared with other complex machine learning or deep learning methods, it is more direct and simple, and improves the practical feasibility and efficiency of panel defect detection in industrial production.

[0012] As some optional embodiments of the present application, after obtaining the panel representation data, the panel representation data needs to be cleaned.

[0013] As some optional embodiments of the present application, cleaning the panel representation data is to eliminate repeated defect points and defect points that have no impact on panel quality.

[0014] In the above embodiment, by cleaning the representation data, repeated and non-impact defect points can be eliminated, effective representation data can be generated, and the accuracy and reliability of defect point aggregation detection can be improved.

[0015] As some optional embodiments of the present application, the process of using a density-based clustering algorithm to perform clustering and division processing on the panel representation data to obtain a plurality of clustering regions is as follows:

[0016] Performing neighborhood screening processing on all defect points based on a preset neighborhood radius and a minimum number of neighborhood points to obtain a plurality of core objects;

[0017] Based on the principle of density reachability and density connectivity, recursively performing clustering and division on all defect points within the neighborhood radius of the core object to obtain a plurality of clustering regions.

[0018] In the above embodiment, by using the density-based clustering algorithm to aggregate defect points into a plurality of clustering regions, it is convenient for later defect point aggregation detection.

[0019] As some optional embodiments of the present application, the process of performing region merging processing on the plurality of clustering regions to obtain at least one merged region is as follows:

[0020] Obtaining inter-cluster distances between the clustering regions based on coordinate information of the defect points;

[0021] If the inter-cluster distance is less than a preset distance threshold, the corresponding clustering areas are merged to obtain at least one merged area.

[0022] In the above embodiment, the clustering areas obtained by the density-based clustering algorithm are adjusted and optimized by merging different clustering areas, and the flexibility and adaptability of judging various aggregation conditions are further improved.

[0023] As some optional embodiments of the present application, the process of comprehensively detecting the panel defect points based on the clustering areas and the merged areas to obtain the defect point aggregation detection result is as follows:

[0024] The number of defect points in the clustering area is obtained based on the coordinate information of the defect points.

[0025] If the number of defect points in at least one clustering area is not less than a preset number threshold, it is determined that the panel has a defect point aggregation condition, otherwise, the number of defect points in the merged area is obtained based on the coordinate information of the defect points, and if the number of defect points in at least one merged area is not less than a preset number threshold, it is determined that the panel has a defect point aggregation condition, otherwise, it is determined that the panel does not have a defect point aggregation condition.

[0026] In the above embodiment, the numbers of defect points in the clustering areas and the merged areas are judged in a phased manner, and the flexibility and adaptability of judging various aggregation conditions are further improved, that is, the defect point aggregation detection can be performed on panels of different shapes, sizes and densities, and has high robustness.

[0027] As some optional embodiments of the present application, the calculation of the inter-cluster distance adopts a distance measurement manner of Euclidean distance, Manhattan distance, Chebyshev distance or Minkowski distance.

[0028] In a second aspect, the present application provides a panel defect point aggregation detection system, the system comprising:

[0029] A data acquisition unit, the data acquisition unit is used to acquire the characterization data of the panel, the characterization data comprising the coordinate information of all defect points on the panel;

[0030] A clustering processing unit, the clustering processing unit is used to perform clustering division processing on the characterization data of the panel by using a density-based clustering algorithm to obtain a plurality of clustering areas;

[0031] A region merging unit, the region merging unit is used to perform region merging processing on the plurality of clustering areas to obtain at least one merged area;

[0032] A comprehensive detection unit, the comprehensive detection unit is used to comprehensively detect the panel defect points based on the clustering areas and the merged areas to obtain a defect point aggregation detection result.

[0033] In a third aspect, the present application provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the panel defect point aggregation detection method when executing the computer program.

[0034] In a fourth aspect, the present application provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executable on a processor to implement the panel defect point aggregation detection method.

[0035] The present application has the following advantages:

[0036] 1. The present application applies the density-based clustering algorithm to industrial panel defect detection, and through the introduction of inter-cluster defect point merging adjustment and intra-cluster defect point total number threshold judgment, the aggregation condition can be more intuitively considered, and the accuracy and reliability of defect point aggregation detection are improved.

[0037] 2. The present application adjusts and optimizes the clustering region given by the density-based clustering algorithm through phased aggregation detection combined with clustering region merging, further improving the flexibility and adaptability of determining various aggregation conditions. BRIEF DESCRIPTION OF DRAWINGS

[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope, and for those skilled in the art, other related drawings can also be obtained without creative labor.

[0039] Figure 1 is a flow chart of the panel defect point aggregation detection method according to the embodiments of the present application;

[0040] Figure 2 is a demonstration schematic diagram of the density-based clustering algorithm according to the embodiments of the present application;

[0041] Figure 3 is a clustering schematic diagram of the density-based clustering algorithm according to the embodiments of the present application;

[0042] Figure 4 is a diffusion aggregation schematic diagram according to the embodiments of the present application;

[0043] Figure 5 is a diffusion aggregation schematic diagram according to the embodiments of the present application;

[0044] Figure 6is a structural block diagram of the panel defect point aggregation detection system described in the embodiments of the present application. DETAILED DESCRIPTION

[0045] In order to better understand the above technical solutions, the technical solutions of the present application will be described in detail below through the drawings and specific embodiments. It should be understood that the specific features in the embodiments and examples of the present application are detailed descriptions of the technical solutions of the present application, and are not limitations of the technical solutions of the present application. In the case of no conflict, the technical features in the embodiments and examples of the present application can be combined with each other.

[0046] It should also be understood that, in order to simplify the description of the present application disclosed and to help understand at least one embodiment of the present application, in the foregoing description of the embodiments of the present application, various features are sometimes combined into one embodiment, figure or description thereof. However, this method of disclosure does not mean that the features required by the present application are more than those mentioned in the claims. In fact, the features of the embodiments are less than all the features of the above-mentioned single embodiment.

[0047] In the industrial production process, there are two prominent features of panel defect points: (1) uneven distribution: panel defect points may be unevenly distributed in different areas, some areas may have a large number of defect points, while other areas are relatively few. This uneven distribution may be affected by many factors such as manufacturing process, material properties, process parameters and operating factors, in addition, the shape of the defect point aggregation area may be irregular, without obvious geometric pattern. (2) Spatial correlation: panel defect points may have certain spatial correlation, that is, the distribution of defect points on the panel may show a tendency of aggregation or clustering. This spatial correlation may be related to the manufacturing process of the panel, material properties or other factors. In some cases, defect points may be aggregated in a linear or planar form, showing obvious spatial patterns.

[0048] In order to solve the limitations of traditional machine vision and clustering methods in panel defect detection, the present application provides a panel defect point aggregation detection method, please refer to Figure 1 , Figure 1 is a flowchart of the method, the method comprising the following steps:

[0049] (1) Obtain the characterization data of the panel, which includes the coordinate information of all defect points on the panel surface, but is not limited to the coordinate information of the defect points, but also can contain panel cut attribution information, size information, etc. The embodiments of the present application do not limit this.

[0050] In the embodiments of the present application, after obtaining the characterization data of the panel, the characterization data of the panel needs to be cleaned to generate effective characterization data, which is beneficial to improve the accuracy and reliability of the defect point aggregation detection.

[0051] Specifically, the process of cleaning the characterization data of the panel is as follows:

[0052] (1.1) Since there may be a problem of repeated recording in the process of recording the defect point coordinate information, the inter-point distance of the defect points is calculated, two defect points with an inter-point distance less than a preset de-duplication threshold are defined as repeated points, and the repeated points are de-duplicated to eliminate redundant defect points. The preset de-duplication threshold can be set according to business requirements, and the present application does not limit it. Preferably, the present application selects the defect points with an inter-point distance less than 1 um as repeated points.

[0053] (1.2) Since there may be a problem of excessive recording in the process of recording the defect point coordinate information, the defect points that have no effect on the quality of the panel are removed to make the subsequent defect point aggregation detection more accurate. The defect points that have no effect on the quality of the panel refer to the defect points that have no effect on the quality of the entire panel. For example, in the industrial panel generation process, the panel is separated and cut according to the determined position relationship to form multiple panel cuttings, so that there is a certain gap between different panel cuttings. The defect points in the gap area are the defect points that have no effect on the quality of the panel.

[0054] (2) The density-based clustering algorithm (DBSCAN) is used to cluster the characterization data of the panel, and the defect points are divided into n clusters, i.e. n clustering regions are obtained.

[0055] The density-based clustering algorithm can effectively divide the defect points in the characterization data into different clusters, and can also identify noise points.

[0056] Specifically, the process of the density-based clustering algorithm is as follows:

[0057] (2.1) Set the neighborhood radius ε and the minimum neighborhood point number. A smaller neighborhood radius ε will result in tighter clustering, while a larger neighborhood radius ε will allow a larger distance to connect the defect points. At the same time, the setting of the minimum neighborhood point number will also affect the selection of core objects. The neighborhood radius ε and the minimum neighborhood point number can be set according to business requirements, and the present application does not limit it. Preferably, the neighborhood radius ε set by the present application is 20 um, and the minimum neighborhood point number is 2.

[0058] (2.2) Based on the neighborhood radius ε and the minimum number of neighborhood points, perform neighborhood filtering on all defect points to obtain several core objects; that is, first calculate the distance between each defect point and other defect points according to the coordinate information of each defect point. If the distance between two defect points is less than the neighborhood radius ε, then the two defect points are regarded as neighbors.

[0059] Please see Figure 2 , Figure 2 This is a demonstration diagram of a density-based clustering algorithm. For example, a circle is drawn with defect points A, B, and C as centers, and the radius of the circle is the neighborhood radius ε. Next, by calculating the number of other defect points within the neighborhood radius of each defect point, it can be determined whether the defect point meets the criteria for a core object. When the number of defect points within the neighborhood radius ε of a defect point is greater than the minimum number of neighborhood points, the defect point is defined as a core object. That is, if the number of defect points within the circle of defect point B is greater than the minimum number of neighborhood points, then defect point B is determined to be a core object.

[0060] (2.3) Based on the principles of density reachability and density connectivity, a recursive approach is used to cluster all defect points within the neighborhood radius ε of the core object, dividing the defect points into n clusters to obtain n clustering regions. That is, if a defect point is within the neighborhood radius ε of the core object, or if there is a path connecting it to the core object through other core objects, then the defect point is density reachable from the core object, and they belong to the same cluster. At the same time, if a defect point is density reachable from at least two other defect points, then the at least two density reachable defect points are density connected, and they belong to the same cluster. The set of clusters formed by defect points is denoted as C = {C1, C2, ..., C...} n There are a total of n clusters.

[0061] Initial clustering detection of defect points on the panel is performed based on clustered regions; that is, firstly, a set of clusters C = {C1, C2, ..., C} is obtained after clustering. n The number of defect points in each cluster is counted. Then, based on the characteristics of the panel data and the requirements for defect aggregation in the business, an appropriate threshold is set to determine whether the number of defect points in each cluster meets the aggregation requirements. If the number of defect points in a cluster reaches or exceeds the threshold, it is considered a "clustered cluster"; otherwise, it is considered a "non-clustered cluster". Finally, the panel containing a "clustered cluster" (as long as it contains only one "clustered cluster") is directly denoted as the "clustered panel", and the set formed by it is denoted as G = {G1, G2, ..., G...}. m}; A panel that does not contain any "clusters" is denoted as an "uncertain panel", and the set formed by it is denoted as U = {U1, U2, ..., U}. n-m The quantity threshold is set according to business needs, and the embodiments of the present invention do not impose restrictions. Preferably, the quantity threshold set in the embodiments of the present invention is 50.

[0062] (4) The clustering regions are merged based on the inter-cluster distance to obtain at least one merged region, and the defect points of the panel are detected again based on the merged region.

[0063] For the "uncertain panel", the re-aggregation determination is performed by calculating the inter-cluster distance to further optimize the merging of the clustering results.

[0064] Specifically, the process of detecting the defect points of the panel again is as follows:

[0065] (4.1) Calculate the inter-cluster distance: for each panel in the "uncertain panel" set U = {U1, U2, …, U n-m}, calculate the distance between all clusters on the panel, thereby obtaining an inter-cluster distance matrix, which records the distance value between each pair of clusters. The distance calculation can use Euclidean distance, Manhattan distance, Chebyshev distance or Minkowski distance, etc. The present application does not make any limitation.

[0066] (4.2) Distance threshold setting: according to the results of the panel inter-cluster distance calculation, a suitable distance threshold needs to be set to determine whether the inter-cluster distance meets the merging requirements; in the initial determination stage, the small range of special aggregation is selected; however, when facing the diffusion aggregation situation, each cluster will usually contain a small amount of defect points, but these clusters are basically located in the same tile or adjacent tile. Therefore, the present application embodiment can merge these small clusters into a large cluster to more comprehensively consider all aggregation situations. Preferably, the distance threshold set by the present application embodiment is 150.

[0067] (4.3) Inter-cluster distance determination: compare the distance between each pair of clusters based on the inter-cluster distance matrix with the distance threshold value. If the inter-cluster distance is less than or equal to the distance threshold value, the two clusters are determined as candidate merged clusters, and they are marked as to-be-merged state.

[0068] (4.4) Cluster merging: according to the to-be-merged state marking, the clusters that meet the merging conditions are merged. After the cluster merging operation on the panel, the set of all clusters is denoted as C' = {C'1, C'2, C'3, …, C' N}.

[0069] (4.5) Re-aggregation determination: for the cluster set C' = {C'1, C'2, C'3, …, C' NIf the number of defect points contained in each cluster meets or exceeds the number threshold of defect points in the cluster set in step (3), the cluster is regarded as an "aggregated cluster". If the panel contains an "aggregated cluster", the mark of the panel is changed from "uncertain panel" to "aggregated panel".

[0070] The embodiment of the present application takes the aggregated area of defect points of a certain panel as an example to illustrate the comprehensive detection of defect points.

[0071] ① The aggregation determination is initially performed based on the number of defect points of each cluster (clustering area).

[0072] Please refer to Figure 3 , Figure 3 Six clusters of defect points of a certain panel obtained after clustering processing based on the density-based clustering algorithm are denoted as C={C1, C2, …, C6}, and the total number of defect points counted in each cluster is denoted as M={M1, M2, …, M6}. If the number of defect points contained in each cluster in C={C1, C2, …, C6} is 90, 56, 95, 156, 54, and 69 respectively, and the number of defect points of each cluster is greater than the number threshold 50, C1, C2, …, C6 are all "aggregated clusters", and the panel is directly recorded as an "aggregated panel".

[0073] It should be noted that as long as there is a cluster of defect points on a certain panel whose number exceeds the number threshold, the panel will be recorded as an "aggregated panel".

[0074] ② The merged cluster (merged area) is calculated based on the distance between clusters, and the aggregation determination is performed again based on the number of defect points of the merged cluster.

[0075] The "aggregated panel" screened out after the initial aggregation determination often has the characteristics of small number of panel cutting blocks and extremely high density, and it is impossible to determine the diffusion aggregation; please refer to Figure 4 The characteristics of this type of aggregation are that the density is not large, and the number of panel cutting blocks occupied by the defect points is also relatively large, but from the whole panel, the average number of defect points occupied by this part is far greater than the average number of defect points of the whole panel, so for processing the aggregation of this panel, the merged cluster should be calculated based on the distance between clusters, and the aggregation determination should be performed again.

[0076] Please refer to Figure 4 , Figure 4For the diffusion aggregation diagram, if the panel is clustered by using the density-based clustering algorithm to obtain 101 clusters, but the maximum number of defect points contained in each cluster is not more than 20, and the minimum number of defect points contained in each cluster is not less than 2, and the number of defect points contained in each cluster is less than the number threshold 50, but most of the diffusion aggregation occurs on the right side of the panel, therefore, the embodiment of the application merges these small clusters of diffusion aggregation into one large cluster based on the distance between the clusters, in order to better judge such aggregation, that is, the 101 clusters are traversed in a loop, the distance between the cluster centers of each two clusters is calculated, and if the distance is less than the distance threshold, the two clusters are merged.

[0077] After merging by recursion, a large cluster is obtained, please refer to Figure 5 , Figure 5 For the diffusion aggregation merging diagram, a total of 578 defect points are obtained; and the number of defect points contained in the large cluster is compared with the set number threshold, if the number of defect point data in the cluster meets or exceeds the number threshold, it is regarded as an "aggregated cluster", and the panel is recorded as an "aggregated panel".

[0078] Among them, the number threshold for the second aggregation determination can be the same as the number threshold for the first aggregation determination, and at the same time, it can also be set according to business requirements, and the embodiment of the application does not limit it.

[0079] Embodiment 2

[0080] The application provides a panel defect point aggregation detection system, please refer to Figure 6 , Figure 6 For the structure diagram of the system, the system corresponds to the method of embodiment 1 one by one, and the system comprises:

[0081] A data acquisition unit, the data acquisition unit is used for acquiring the representation data of the panel, and the representation data comprises the coordinate information of all defect points on the panel;

[0082] A clustering processing unit, the clustering processing unit is used for clustering and dividing the representation data of the panel by using a density-based clustering algorithm, to obtain a plurality of clustering regions;

[0083] A region merging unit, the region merging unit is used for performing region merging processing on the plurality of clustering regions, to obtain at least one merged region;

[0084] A comprehensive detection unit, the comprehensive detection unit is used for performing comprehensive detection on the panel defect points based on the clustering regions and the merged region, to obtain a defect point aggregation detection result.

[0085] Embodiment 3

[0086] The application provides a computer device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the panel defect point aggregation detection method in embodiment 1 when executing the computer program.

[0087] The computer device provided in the embodiment can implement the method in embodiment 1, and details are not described herein again to avoid repetition.

[0088] Embodiment 4

[0089] The application provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the panel defect point aggregation detection method in embodiment 1.

[0090] The computer readable storage medium provided in the embodiment can implement the method in embodiment 1, and details are not described herein again to avoid repetition.

[0091] The processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0092] The memory can be used to store the computer program and / or modules, and the processor realizes various functions of the panel defect point aggregation detection system in the application by running or executing data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application program required by a function (such as a sound playing function, an image playing function, etc.) and the like. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory such as a hard disk, a memory, a plug-in hard disk, an intelligent memory card, a secure digital card, a flash memory card, at least one disk storage device, a flash memory device or other volatile solid-state memory device.

[0093] If the panel defect point aggregation detection system is implemented in the form of a software function unit and sold or used as a stand-alone product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment methods, and can also be stored in a computer-readable storage medium through a computer program, which, when executed by a processor, can implement the steps of the above-mentioned various method embodiments. Among them, the computer program includes computer program code, object code form, executable file or some intermediate form, etc. The computer-readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, point carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction.

[0094] The present application has described the basic concept, and it is obvious that the above detailed disclosure is only as an example and does not constitute a limitation on the specification for those skilled in the art. Although it is not explicitly stated here, those skilled in the art can make various modifications, improvements and corrections to the specification. Such modifications, improvements and corrections are suggested in the specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of the specification.

Claims

1. A method for detecting clustered defects in a panel, characterized in that, The method comprises the following steps: Obtaining characterization data of the panel, the characterization data comprising coordinate information of all defect points on the panel; Performing cluster division processing on the characterization data of the panel by using a density-based clustering algorithm to obtain a plurality of cluster regions; The process of performing cluster division processing on the characterization data of the panel by using a density-based clustering algorithm to obtain a plurality of cluster regions is as follows: Performing neighborhood screening processing on all defect points based on a preset neighborhood radius and a minimum number of neighborhood points to obtain a plurality of core objects; Based on the principles of density reachability and density connectivity, performing cluster division on all defect points within the neighborhood radius of the core objects in a recursive manner to obtain a plurality of cluster regions; Performing region merging processing on the plurality of cluster regions to obtain at least one merged region; Performing comprehensive detection on the defect points of the panel based on the cluster regions and the merged region to obtain defect point aggregation detection results; The process of performing comprehensive detection on the defect points of the panel based on the cluster regions and the merged region to obtain defect point aggregation detection results is as follows: Performing preliminary aggregation detection on the defect points of the panel based on the cluster regions, that is, obtaining the number of defect points of the cluster regions based on the coordinate information of the defect points; If the number of defect points of at least one cluster region is not less than a preset number threshold, it is determined that the panel has defect point aggregation, otherwise, obtaining the number of defect points of the merged regions based on the coordinate information of the defect points, and performing re-aggregation detection on the defect points of the panel based on the merged regions, that is, if the number of defect points of at least one merged region is not less than a preset number threshold, it is determined that the panel has defect point aggregation, otherwise, it is determined that the panel does not have defect point aggregation.

2. The method of claim 1, wherein the method further comprises: After obtaining the characterization data of the panel, the characterization data of the panel needs to be cleaned.

3. The method of claim 2, wherein the step of detecting the defect point is performed by using a defect point detection algorithm. Cleaning the characterization data of the panel is to eliminate repeated defect points and defect points that have no effect on the quality of the panel.

4. The method of claim 1, wherein the method further comprises: The process of performing region merging processing on the plurality of cluster regions to obtain at least one merged region is as follows: Obtaining the inter-cluster distance between the cluster regions based on the coordinate information of the defect points; If the inter-cluster distance is less than a preset distance threshold, the corresponding cluster regions are merged to obtain at least one merged region.

5. The method of claim 4, wherein: The calculation of the inter-cluster distance adopts distance measurement methods of Euclidean distance, Manhattan distance, Chebyshev distance or Minkowski distance.

6. A panel defect spot cluster detection system characterized by, The system comprises: A data acquisition unit, configured to obtain characterization data of the panel, the characterization data comprising coordinate information of all defect points on the panel; A cluster processing unit, configured to perform cluster division processing on the characterization data of the panel by using a density-based clustering algorithm to obtain a plurality of cluster regions; The process of performing cluster division processing on the characterization data of the panel by using a density-based clustering algorithm to obtain a plurality of cluster regions is as follows: Performing neighborhood screening processing on all defect points based on a preset neighborhood radius and a minimum number of neighborhood points to obtain a plurality of core objects; Based on the principles of density reachability and density connectivity, performing cluster division on all defect points within the neighborhood radius of the core objects in a recursive manner to obtain a plurality of cluster regions; a region merging unit configured to perform a region merging process on the plurality of clustering regions to obtain at least one merged region; a comprehensive detection unit configured to perform a comprehensive detection on the panel defect points based on the clustering regions and the merged region to obtain a defect point clustering detection result; The process of performing a comprehensive detection on the panel defect points based on the clustering regions and the merged region to obtain a defect point clustering detection result is as follows: perform a preliminary clustering detection on the panel defect points based on the clustering regions, i.e., obtain the number of defect points of the clustering regions based on the coordinate information of the defect points; if the number of defect points of at least one clustering region is not less than a preset number threshold, it is determined that the panel has a defect point clustering situation, otherwise, obtain the number of defect points of the merged regions based on the coordinate information of the defect points, and perform a re-clustering detection on the panel defect points based on the merged regions, i.e., if the number of defect points of at least one merged region is not less than the preset number threshold, it is determined that the panel has a defect point clustering situation, otherwise, it is determined that the panel does not have a defect point clustering situation.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: The processor implements the panel defect point clustering detection method of any one of claims 1-5 when executing the computer program.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium stores the computer program, and the computer program implements the panel defect point clustering detection method of any one of claims 1-5 when executed by the processor.

Images