CN117076960A - Panel defect point aggregation detection method, system, equipment and storage medium - Google Patents
Panel defect point aggregation detection method, system, equipment and storage medium Download PDFInfo
- Publication number
- CN117076960A CN117076960A CN202311031487.5A CN202311031487A CN117076960A CN 117076960 A CN117076960 A CN 117076960A CN 202311031487 A CN202311031487 A CN 202311031487A CN 117076960 A CN117076960 A CN 117076960A
- Authority
- CN
- China
- Prior art keywords
- panel
- defect
- clustering
- points
- point aggregation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000007547 defect Effects 0.000 title claims abstract description 174
- 238000004220 aggregation Methods 0.000 title claims abstract description 93
- 230000002776 aggregation Effects 0.000 title claims abstract description 90
- 238000001514 detection method Methods 0.000 title claims abstract description 48
- 238000003860 storage Methods 0.000 title claims abstract description 15
- 238000000034 method Methods 0.000 claims abstract description 42
- 238000012512 characterization method Methods 0.000 claims abstract description 35
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 22
- 238000012545 processing Methods 0.000 claims abstract description 22
- 238000004590 computer program Methods 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 12
- 238000004140 cleaning Methods 0.000 claims description 6
- 238000012216 screening Methods 0.000 claims description 5
- 238000005259 measurement Methods 0.000 claims description 3
- 230000002950 deficient Effects 0.000 description 23
- 238000010586 diagram Methods 0.000 description 6
- 238000004519 manufacturing process Methods 0.000 description 6
- 238000009792 diffusion process Methods 0.000 description 5
- 238000009826 distribution Methods 0.000 description 5
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The application provides a method, a system, equipment and a storage medium for detecting panel defect point aggregation, which relate to the technical field of defect detection, and the method comprises the following steps: acquiring characterization data of a panel, wherein the characterization data comprises coordinate information of all defect points on the surface of the panel; clustering and dividing the characterization data of the panel by adopting a clustering algorithm based on density to obtain a plurality of clustering areas; carrying out region merging processing on the plurality of clustered regions to obtain at least one merged region; and comprehensively detecting the defect points of the panel based on the clustering area and the merging area to obtain a defect point aggregation detection result. The application applies the clustering algorithm based on density to the defect detection of the industrial panel, and can more intuitively consider the defect point aggregation condition by introducing the modes of combining and adjusting the defect points among clusters and judging the total number of the defect points in the clusters, thereby improving the accuracy and the reliability of the defect point aggregation detection.
Description
Technical Field
The present application relates to the field of defect detection technologies, and in particular, to a method, a system, an apparatus, and a storage medium for detecting panel defect point aggregation.
Background
The panel defect point aggregation refers to a certain number of defective defect points with uneven sizes on the surface of the panel during the production and processing process of the industrial panel, and the defective defect points mostly have certain regional tendencies and appear to be aggregated in a certain region. For the panel with the bad defect point aggregation, if the aggregation condition of the bad defect points of the panel cannot be accurately observed, the missing detection is very easy to be caused, and the mass production risk of defective products is increased. Therefore, how to ensure accurate detection of the panel defective defect point aggregation area and reduce the mass production risk of defective products is a core problem to be solved in the field.
The prior art mainly comprises a machine vision mode, a clustering mode and the like. By means of machine vision, a defect can be identified from the captured image and preliminary localization can be performed. By means of clustering, the defect point aggregation can be further identified. These techniques have achieved a certain detection effect, and can detect the aggregation defect relatively quickly and effectively, but still have some problems. For example, image quality is affected by factors such as illumination, reflection, and shading, which may affect the results of defect detection. The clustering method has certain difficulty on the scene of complex defect point distribution on the surface of the panel.
In addition, the prior art has the following problems in detecting defect point aggregation: (1) lack of robustness: the distribution condition of defect points of the existing panel is very complex, the panel has the characteristics of clusters with different shapes, sizes and densities, and the robustness of the existing mode for all defect conditions is not good, so that the existing mode is aimed at specific defect conditions. (2) lack of intuitiveness: current technology typically involves complex algorithms and data processing procedures, lacking the ability to directly manipulate the data.
Disclosure of Invention
The application provides a method, a system, equipment and a storage medium for detecting panel defect point aggregation, which solve the problem that the prior art lacks robustness and observability when detecting the defect point aggregation.
In a first aspect, an embodiment of the present application provides a method for detecting panel defect point aggregation, including the steps of:
acquiring characterization data of a panel, wherein the characterization data comprises coordinate information of all defect points on the surface of the panel;
clustering and dividing the characterization data of the panel by adopting a clustering algorithm based on density to obtain a plurality of clustering areas;
carrying out region merging processing on the plurality of clustered regions to obtain at least one merged region;
and comprehensively detecting the defect points of the panel based on the clustering area and the merging area to obtain a defect point aggregation detection result.
In the above embodiment, the clustering algorithm based on density is applied to the defect detection of the industrial panel, and the aggregation condition can be more intuitively considered by introducing the modes of combining and adjusting the inter-cluster defect points and judging the total number of the defect points in the cluster, so that the accuracy and the reliability of the defect point aggregation detection are improved; compared with other complicated machine learning or deep learning modes, the method is more direct and concise, and the practical feasibility and efficiency of panel defect detection in industrial production are improved.
As some optional embodiments of the present application, after the characterization data of the panel is obtained, a cleaning process is required to be performed on the characterization data of the panel.
As some optional embodiments of the application, the cleaning treatment is performed on the characterization data of the panel, namely, the repeated defect points and the defect points which have no influence on the quality of the panel are removed.
In the above embodiment, by performing cleaning treatment on the characterization data, repeated and unaffected defect points can be removed, and effective characterization data can be generated, which is beneficial to improving the accuracy and reliability of defect point aggregation detection.
As some optional embodiments of the application, the clustering division processing is carried out on the characterization data of the panel by adopting a clustering algorithm based on density, so as to obtain a plurality of clustering areas, wherein the flow is as follows:
performing neighborhood screening treatment on all defect points based on a preset neighborhood radius and a minimum neighborhood point number to obtain a plurality of core objects;
based on the principle that the density can be reached and the density is connected, all defect points in the neighborhood radius of the core object are clustered and divided in a recursive mode, so that a plurality of clustered areas are obtained.
In the above embodiment, the defect points are clustered into a plurality of clustered regions by a density-based clustering algorithm, so that the later defect point clustering detection is facilitated.
As some optional embodiments of the present application, the flow of performing region merging processing on a plurality of clustered regions to obtain at least one merged region is as follows:
acquiring inter-cluster distances between the clustered areas based on the coordinate information of the defect points;
if the distance between clusters is smaller than a preset distance threshold, merging the corresponding cluster areas to obtain at least one merged area.
In the above embodiment, the clustering areas obtained by the density-based clustering algorithm can be adjusted and optimized by combining different clustering areas, so that the flexibility and adaptability of judging various aggregation conditions are further improved.
As some optional embodiments of the application, the process of comprehensively detecting the defect points of the panel based on the clustering area and the merging area to obtain the defect point aggregation detection result is as follows:
acquiring the number of defect points in the clustering area based on the coordinate information of the defect points;
if the number of the defect points in the at least one clustering area is not smaller than the preset number threshold, judging that the defect point aggregation condition exists in the panel, otherwise, acquiring the number of the defect points in the merging area based on the coordinate information of the defect points, if the number of the defect points in the at least one merging area is not smaller than the preset number threshold, judging that the defect point aggregation condition exists in the panel, otherwise, judging that the defect point aggregation condition does not exist in the panel.
In the above embodiment, the number of defect points in the clustering area and the merging area is determined in a staged manner, so that the flexibility and adaptability of determining various aggregation conditions are further improved, that is, the defect point aggregation detection can be performed on panels with different shapes, sizes and densities, and the robustness is higher.
As some optional embodiments of the application, the calculation of the cluster distance adopts a distance measurement mode of Euclidean distance, manhattan distance, chebyshev distance or Minkowski distance.
In a second aspect, the present application provides a panel defect point aggregation detection system, the system comprising:
the data acquisition unit is used for acquiring characterization data of the panel, wherein the characterization data comprise coordinate information of all defect points on the panel;
the clustering processing unit performs clustering division processing on the characterization data of the panel by adopting a density-based clustering algorithm to obtain a plurality of clustering areas;
the region merging unit is used for carrying out region merging processing on the plurality of clustered regions so as to obtain at least one merged region;
and the comprehensive detection unit is used for comprehensively detecting the defect points of the panel based on the clustering area and the merging area so as to obtain a defect point aggregation detection result.
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, the processor implementing the panel defect point aggregation detection method when executing the computer program.
In a fourth aspect, the present application provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the panel defect point aggregation detection method.
The beneficial effects of the application are as follows:
1. the clustering method based on density is applied to the defect detection of the industrial panel, and the aggregation condition can be more intuitively considered by introducing the modes of combining and adjusting the inter-cluster defect points and judging the total number of the intra-cluster defect points, so that the accuracy and the reliability of the defect point aggregation detection are improved.
2. According to the clustering method, the clustering areas given by the density-based clustering algorithm are adjusted and optimized by means of staged aggregation detection and combining the clustering areas, so that the flexibility and adaptability of judging various aggregation conditions are further improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present application and should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for detecting panel defect point aggregation according to an embodiment of the present application;
FIG. 2 is a schematic illustration of a density-based clustering algorithm in accordance with an embodiment of the present application;
FIG. 3 is a clustering schematic diagram of a density-based clustering algorithm according to an embodiment of the application;
FIG. 4 is a schematic illustration of diffusion aggregation according to an embodiment of the present application;
FIG. 5 is a schematic diagram of a diffusion aggregation merge according to an embodiment of the present application;
fig. 6 is a block diagram of a panel defect point aggregation detection system according to an embodiment of the present application.
Detailed Description
In order to better understand the above technical solutions, the following detailed description of the technical solutions of the present application is made by using the accompanying drawings and specific embodiments, and it should be understood that the specific features of the embodiments and the embodiments of the present application are detailed descriptions of the technical solutions of the present application, and not limiting the technical solutions of the present application, and the technical features of the embodiments and the embodiments of the present application may be combined with each other without conflict.
It should also be appreciated that in the foregoing description of at least one embodiment of the application, various features are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure aiding in the understanding of at least one embodiment of the application. This method of disclosure, however, is not intended to imply that more features than are required by the subject application. Indeed, less than all of the features of a single embodiment disclosed above.
In an industrial process, there are two prominent features of defective points of the panel: (1) distribution non-uniformity: the defect points of the panel may be unevenly distributed in different areas, some areas may collect a large number of defect points, while other areas are relatively few, and the distribution unevenness may be affected by various factors such as manufacturing process, material characteristics, process parameters, and operation factors, and in addition, the shape of the defect point collecting area may be irregular without obvious geometric patterns. (2) spatial correlation: the defective spots of the panel may have a certain spatial correlation, i.e. the distribution of defective spots on the panel may show a tendency to aggregate or cluster, which may be related to the manufacturing process, material properties or other factors of the panel, and in some cases the defective spots may aggregate together in a linear or planar form, showing a distinct spatial pattern.
In order to solve the limitations of the traditional machine vision mode and clustering mode in panel defect detection. The application provides a method for detecting panel defect point aggregation, referring to fig. 1, fig. 1 is a flow chart of the method, the method comprises the following steps:
(1) The method comprises the steps of obtaining characterization data of the panel, wherein the characterization data comprise coordinate information of all defect points on the surface of the panel, but the characterization data are not limited to the coordinate information of the defect points, and can also comprise attribution information, size information and the like of panel dicing, and the embodiment of the application is not limited to the coordinate information.
In the embodiment of the application, after the characterization data of the panel are acquired, the characterization data of the panel are required to be cleaned so as to generate effective characterization data, thereby being beneficial to improving the accuracy and reliability of defect point aggregation detection.
Specifically, the flow of the cleaning process for the characterization data of the panel is as follows:
(1.1) since there is a problem of repeated recording during the recording of the defective point coordinate information, two defective points having a point-to-point distance smaller than a preset de-duplication threshold are defined as repeated points by calculating the point-to-point distance of the defective points, and the repeated points are de-duplicated to eliminate redundant defective points. The preset de-duplication threshold can be set according to the service requirement, and the embodiment of the application is not limited, and preferably, the embodiment of the application selects the defect point with the inter-point distance smaller than 1um as the duplication point.
(1.2) because the problem of excessive recording exists in the process of recording the coordinate information of the defect points, the defect points which have no influence on the quality of the panel are removed, so that the subsequent defect point aggregation detection is more accurate, wherein the defect points which have no influence on the quality of the panel refer to the defect points which have no influence on the quality of the whole panel, for example, in the process of generating and processing the industrial panel, the panel is divided and cut according to the determined position relationship to form a plurality of panel cut pieces, so that certain gaps exist between different panel cut pieces, and the defect points in the gap areas are the defect points which have no influence on the quality of the panel.
(2) Clustering the characterization data of the panel by adopting a density-based clustering algorithm (DBSCAN), and dividing the defect points into n clusters to obtain n clustering areas.
The density-based clustering algorithm can effectively divide defect points in the characterization data into different clusters, and meanwhile noise points can be identified.
Specifically, the flow of the density-based clustering algorithm is as follows:
(2.1) setting a neighborhood radius epsilon and minimum neighborhood points, wherein a smaller neighborhood radius epsilon can lead to tighter clustering, a larger neighborhood radius epsilon can allow a larger distance to connect defect points, and meanwhile, the setting of the minimum neighborhood points can influence the screening of core objects, wherein the neighborhood radius epsilon and the minimum neighborhood points can be set automatically according to service requirements, and the embodiment of the application is not limited; preferably, the embodiment of the application sets the neighborhood radius epsilon to 20um and the minimum neighborhood point number to 2.
(2.2) carrying out neighborhood screening treatment on all defect points based on the neighborhood radius epsilon and the minimum neighborhood point number so as to obtain a plurality of core objects; namely, firstly, calculating the distance between each defect point and other defect points according to the coordinate information of each defect point, and if the distance between the two defect points is smaller than the neighborhood radius epsilon, treating the two defect points as neighbors.
Referring to fig. 2, fig. 2 is a schematic diagram illustrating a density clustering algorithm, for example, a circle is drawn by using a defect point A, B, C as a center, and a radius of the circle is a neighborhood radius epsilon; secondly, by calculating the number of other defect points in the neighborhood radius of each defect point, whether the defect point meets the condition of the core object can be determined, when the number of the defect points in the neighborhood radius epsilon of the defect point is larger than the minimum neighborhood point number, the defect point is defined as the core object, namely if the number of the defect points in the circle of the defect point B is larger than the minimum neighborhood point number, the defect point B is judged as the core object.
(2.3) based on the principle that the density can be reached and the density is connected, carrying out cluster division on all defect points in the neighborhood radius epsilon of the core object in a recursion mode, and dividing the defect points into n clusters to obtain n cluster areas; that is, if a certain defect point is within the neighborhood radius epsilon of the core object, or there is a path connected to the core object through other core objects, then the defect point and the core object can reach the density, and they belong to the same cluster; meanwhile, if a certain defect point is simultaneously reachable with at least two defect points, the at least two defect points with reachable densities are connected and belong to the same cluster, i.e. the cluster set formed by the defect points is C= { C 1 ,C 2 ,…,C n N clusters total.
Performing primary aggregation detection on the defect points of the panel based on the clustering area; i.e. first obtaining a set c= { C of clusters based on clustering 1 ,C 2 ,…,C n Counting the number of defect points contained in each cluster, and setting a proper number threshold according to the characteristics of panel data and the requirement of service defect aggregation, wherein the proper number threshold is used for judging whether the number of each cluster meets the requirement of aggregation; if the number of defective points in a cluster meets or exceeds a number threshold, it is considered to be a "clustered cluster"; otherwise, it is considered a "non-aggregated cluster"; finally, the panel containing the cluster (only one cluster is contained) is directly named as the cluster panel, and the set formed by the panel is named as G= { G 1 ,G 2 ,…,G m -a }; for a material without 'aggregation cluster'Is denoted as "uncertain panel", and its constituent set is denoted as u= { U 1 ,U 2 ,…,U n-m The number threshold is set according to the service requirement, which is not limited in the embodiment of the present application, and preferably, the number threshold is set to be 50 in the embodiment of the present application.
(4) And merging the clustered areas based on the inter-cluster distance to obtain at least one merged area, and performing re-aggregation detection on the defect points of the panel based on the merged area.
For the "uncertain panel", the re-aggregation decision is made by calculating the inter-cluster distance to further optimize the merging of the clustering results.
Specifically, the process of performing re-aggregation detection on the defective points of the panel is as follows:
(4.1) calculating an inter-cluster distance: for the "uncertain" panel set u= { U 1 ,U 2 ,…,U n-m And each panel in the array is calculated, so that a matrix between clusters is obtained, the matrix records the distance value between each pair of clusters, and the distance can be calculated in an Euclidean distance, manhattan distance, chebyshev distance or Minkowski distance equidistant measurement mode.
(4.2) distance threshold setting: according to the result of the inter-cluster calculation of the panel, a proper distance threshold is required to be set for judging whether the inter-cluster distance meets the combination requirement; in the initial judging stage, screening out the conditions of small-range special aggregation; however, when faced with a diffuse aggregation situation, each cluster typically contains a small number of defect points, but the clusters are located substantially on the same tile or on adjacent tiles. Therefore, the embodiment of the present application may combine the small clusters into one large cluster to more fully consider all aggregation situations, and preferably, the distance threshold set by the embodiment of the present application is 150.
(4.3) inter-cluster distance determination: the distance between each pair of clusters is compared with a distance threshold based on the inter-cluster distance matrix, and if the inter-cluster distance is less than or equal to the distance threshold, the two clusters are determined as candidate merging clusters and marked as to-be-merged.
(4.4) Cluster combining: according to the mark of the state to be merged, merging the clusters meeting the merging condition, and marking a set formed by all clusters as C ' = { C ' after the cluster merging operation on the panel ' 1 ,C’ 2 ,C’ 3 ,…,C’ N }。
(4.5) reaggregation determination: clustering C ' = { C ' for the sheet panel ' 1 ,C’ 2 ,C’ 3 ,…,C’ N Comparing the number of defective points contained in each cluster with the threshold number of defective points in the cluster set in step (3), and if the number of defective points in the cluster meets or exceeds the threshold number, treating the cluster as an "aggregate cluster"; if the panel contains an "aggregation cluster," the label of the panel is changed from an "indeterminate panel" to an "aggregation panel".
In the embodiment of the application, the defect points are comprehensively detected and described by taking the aggregation area of the defect points of a certain panel as an example:
(1) an aggregation determination is made for the first time based on the number of defective points per cluster (cluster area).
Referring to fig. 3, fig. 3 shows 6 clusters, denoted as c= { C, obtained by clustering defective points of a certain panel by a density-based clustering algorithm 1 ,C 2 ,…,C 6 Total number of defect points counted per cluster at the same time, denoted as m= { M 1 ,M 2 ,…,M 6 }. If expressed as C= { C 1 ,C 2 ,…,C 6 Number M of defective points contained in each cluster 1 ,M 2 ,…,M 6 90,56,95,156,54,69 respectively; because the number of defective points per cluster is greater than the number threshold 50, C 1 ,C 2 ,…,C 6 Are all "clustered", and this panel is directly denoted as "clustered panel".
It should be noted that a panel is referred to as an "aggregate panel" whenever there is a cluster on the panel with a number of defect points exceeding the number threshold.
(2) The merged clusters (merged regions) are calculated based on the inter-cluster distances, and the aggregation determination is performed again based on the number of defective points of the merged clusters.
The condition of the 'aggregated panel' screened by the primary aggregation judgment is characterized by small number of panel cut pieces and extremely high density, and the condition of the diffuse aggregation cannot be judged; referring to fig. 4, this type of aggregation is characterized by not having a large density, and the number of cut-outs of the panel is large due to defect points, but the average defect number occupied by the portion is far greater than that of the entire panel from the viewpoint of the entire panel, so for processing the aggregation of such panels, merging clusters should be calculated based on the inter-cluster distance, and aggregation determination should be performed again.
Referring to fig. 4, fig. 4 is a schematic view of diffusion aggregation, if a panel clusters 101 clusters by using a density-based clustering algorithm, but each cluster contains defect points of not more than 20 at maximum and not less than 2 at minimum, the number of defect points contained in each cluster is less than a number threshold 50, but most of the places where aggregation occurs are on the right side of the panel, therefore, the embodiment of the application merges the small clusters of diffusion aggregation into one large cluster based on the inter-cluster distance, so as to better judge the aggregation situation, namely, circularly traversing the 101 clusters, calculating the distance between cluster center points of every two clusters, and merging the two clusters if the distance is less than the distance threshold.
After recursive merging, a large cluster is obtained, please refer to fig. 5, fig. 5 is a schematic diagram of diffusion aggregation merging, and 578 defect points are all obtained; and comparing the number of the defect points contained in the large cluster with a set number threshold, and if the number of the defect point data in the cluster meets or exceeds the number threshold, treating the defect point data as an 'aggregation cluster', wherein the panel is marked as an 'aggregation panel'.
The number threshold for performing aggregation determination again may be the same as the number threshold for performing aggregation determination for the first time, and may also be set according to service requirements, which is not limited by the embodiment of the present application.
Example 2
The application provides a panel defect point gathering detection system, please refer to fig. 6, fig. 6 is a block diagram of the system, the system corresponds to the method of embodiment 1 one by one, the system comprises:
the data acquisition unit is used for acquiring characterization data of the panel, wherein the characterization data comprise coordinate information of all defect points on the panel;
the clustering processing unit performs clustering division processing on the characterization data of the panel by adopting a density-based clustering algorithm to obtain a plurality of clustering areas;
the region merging unit is used for carrying out region merging processing on the plurality of clustered regions so as to obtain at least one merged region;
and the comprehensive detection unit is used for comprehensively detecting the defect points of the panel based on the clustering area and the merging area so as to obtain a defect point aggregation detection result.
Example 3
The application provides a computer device comprising a memory, a processor and a computer program stored in the memory and operable on the processor, wherein the processor implements the panel defect point aggregation detection method of embodiment 1 when executing the computer program.
The computer device provided in this embodiment may implement the method described in embodiment 1, and in order to avoid repetition, a description thereof will be omitted.
Example 4
The present application provides a computer-readable storage medium having a computer program stored thereon, which when executed by a processor, implements a panel defect point aggregation detection method as described in embodiment 1.
The computer readable storage medium provided in this embodiment may implement the method described in embodiment 1, and will not be described herein in detail to avoid repetition.
The processor may be a central processing unit (CPU, central Processing Unit), but may also be other general purpose processors, digital signal processors (digital signal processor), application specific integrated circuits (Application Specific Integrated Circuit), off-the-shelf programmable gate arrays (Field programmable gate array) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory may be used to store the computer program and/or the module, and the processor may implement various functions of the panel defect point aggregation detection system of the present application by running or executing the data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, application programs required for at least one function (such as a sound playing function, an image playing function, etc.), and the like. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, smart memory card, secure digital card, flash memory card, at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device.
The panel defect point aggregation detection system, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding that the present application implements all or part of the flow of the method of the above-described embodiments, the steps of each method embodiment described above may also be implemented by a computer program stored in a computer readable storage medium, where the computer program when executed by a processor. Wherein the computer program comprises computer program code, object code forms, executable files, or some intermediate forms, etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash disk, a removable hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory, a random access memory, a point carrier signal, a telecommunication signal, a software distribution medium, and the like. It should be noted that the content of the computer readable medium can be appropriately increased or decreased according to the requirements of the legislation and the patent practice in the jurisdiction.
Having described the basic concept, it will be apparent to those skilled in the art that the foregoing detailed disclosure is by way of example only and is not intended to be limiting. Although not explicitly described herein, various modifications, improvements, and adaptations to the present disclosure are possible for those skilled in the art. Such modifications, improvements, and modifications are intended to be suggested within this specification, and therefore, such modifications, improvements, and modifications are intended to be included within the spirit and scope of the exemplary embodiments of the present application.
Claims (10)
1. A method for detecting a panel defect point aggregation, the method comprising the steps of:
acquiring characterization data of a panel, wherein the characterization data comprises coordinate information of all defect points on the panel;
clustering and dividing the characterization data of the panel by adopting a clustering algorithm based on density to obtain a plurality of clustering areas;
carrying out region merging processing on the plurality of clustered regions to obtain at least one merged region;
and comprehensively detecting the defect points of the panel based on the clustering area and the merging area to obtain a defect point aggregation detection result.
2. The method for detecting panel defect point aggregation according to claim 1, wherein after the characteristic data of the panel is obtained, cleaning processing is required for the characteristic data of the panel.
3. The method for detecting the defect point aggregation of the panel according to claim 2, wherein the cleaning process is performed on the characterization data of the panel to remove repeated defect points and defect points which have no influence on the quality of the panel.
4. The method for detecting the panel defect point aggregation according to claim 1, wherein the clustering and dividing process is performed on the characterization data of the panel by adopting a clustering algorithm based on density, so as to obtain a plurality of clustering areas, and the flow is as follows:
performing neighborhood screening treatment on all defect points based on a preset neighborhood radius and a minimum neighborhood point number to obtain a plurality of core objects;
based on the principle that the density can be reached and the density is connected, all defect points in the neighborhood radius of the core object are clustered and divided in a recursive mode, so that a plurality of clustered areas are obtained.
5. The method for detecting panel defect point aggregation according to claim 1, wherein the process of performing region merging processing on the plurality of clustered regions to obtain at least one merged region is as follows:
acquiring inter-cluster distances between the clustered areas based on the coordinate information of the defect points;
if the distance between clusters is smaller than a preset distance threshold, merging the corresponding cluster areas to obtain at least one merged area.
6. The method for detecting panel defect point aggregation according to claim 5, wherein the process of comprehensively detecting the panel defect points based on the clustering area and the merging area to obtain the defect point aggregation detection result is as follows:
acquiring the number of defect points in the clustering area based on the coordinate information of the defect points;
if the number of the defect points in the at least one clustering area is not smaller than the preset number threshold, judging that the defect point aggregation condition exists in the panel, otherwise, acquiring the number of the defect points in the merging area based on the coordinate information of the defect points, if the number of the defect points in the at least one merging area is not smaller than the preset number threshold, judging that the defect point aggregation condition exists in the panel, otherwise, judging that the defect point aggregation condition does not exist in the panel.
7. The method for detecting panel defect point aggregation according to claim 5, wherein: the distance between clusters is calculated by adopting a distance measurement mode of Euclidean distance, manhattan distance, chebyshev distance or Minkowski distance.
8. A panel defect point aggregation detection system, the system comprising:
the data acquisition unit is used for acquiring characterization data of the panel, wherein the characterization data comprise coordinate information of all defect points on the panel;
the clustering processing unit performs clustering division processing on the characterization data of the panel by adopting a density-based clustering algorithm to obtain a plurality of clustering areas;
the region merging unit is used for carrying out region merging processing on the plurality of clustered regions so as to obtain at least one merged region;
and the comprehensive detection unit is used for comprehensively detecting the defect points of the panel based on the clustering area and the merging area so as to obtain a defect point aggregation detection result.
9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized by: the processor, when executing a computer program, implements a panel defect point aggregation detection method as claimed in any one of claims 1-7.
10. A computer readable storage medium, wherein a computer program is stored on the computer readable storage medium, the computer program when executed by a processor implementing a panel defect point aggregation detection method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311031487.5A CN117076960A (en) | 2023-08-16 | 2023-08-16 | Panel defect point aggregation detection method, system, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311031487.5A CN117076960A (en) | 2023-08-16 | 2023-08-16 | Panel defect point aggregation detection method, system, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117076960A true CN117076960A (en) | 2023-11-17 |
Family
ID=88710844
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311031487.5A Pending CN117076960A (en) | 2023-08-16 | 2023-08-16 | Panel defect point aggregation detection method, system, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117076960A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117288770A (en) * | 2023-11-24 | 2023-12-26 | 张家港市鸿海精密模具有限公司 | Multi-dimensional detection method and system for surface defects of blow molding die |
| CN117593295A (en) * | 2024-01-18 | 2024-02-23 | 东莞市立时电子有限公司 | Nondestructive testing method for production defects of mobile phone data line |
-
2023
- 2023-08-16 CN CN202311031487.5A patent/CN117076960A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117288770A (en) * | 2023-11-24 | 2023-12-26 | 张家港市鸿海精密模具有限公司 | Multi-dimensional detection method and system for surface defects of blow molding die |
| CN117288770B (en) * | 2023-11-24 | 2024-02-06 | 张家港市鸿海精密模具有限公司 | Multi-dimensional detection method and system for surface defects of blow molding die |
| CN117593295A (en) * | 2024-01-18 | 2024-02-23 | 东莞市立时电子有限公司 | Nondestructive testing method for production defects of mobile phone data line |
| CN117593295B (en) * | 2024-01-18 | 2024-05-28 | 东莞市立时电子有限公司 | Nondestructive testing method for production defects of mobile phone data line |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117076960A (en) | Panel defect point aggregation detection method, system, equipment and storage medium | |
| CN115222741B (en) | Cable surface defect detection method | |
| TWI603072B (en) | Detecting defects on a wafer | |
| JP6618478B2 (en) | Automatic in-line inspection and measurement using projected images | |
| US7084968B2 (en) | Method for analyzing defect data and inspection apparatus and review system | |
| CN102396058B (en) | Detecting defects on a wafer | |
| JP4008291B2 (en) | Pattern inspection apparatus, pattern inspection method, and program | |
| CN108596903B (en) | Method for detecting defects of black edges and fragments of solar cell | |
| TW201839383A (en) | Dynamic care areas for defect detection | |
| CN102194725B (en) | Identification method of data point distribution area on coordinate plane and recording medium | |
| TW201511157A (en) | Methods and systems for detecting repeating defects on semiconductor wafers using design data | |
| JP2013160629A (en) | Defect inspection method, defect inspection apparatus, program, and output unit | |
| US11788973B2 (en) | Defect inspection device and defect inspection method | |
| CN108460776B (en) | Brake pad size detection method and device based on machine vision | |
| JP2015511310A (en) | Segmentation for wafer inspection | |
| US20090220142A1 (en) | Linear pattern detection method and apparatus | |
| CN107516085B (en) | Method for automatically removing black edges based on document image | |
| CN115775250A (en) | Golden finger circuit board defect rapid detection system based on digital image analysis | |
| US20110117683A1 (en) | Chip quality determination method and marking mechanism using same | |
| CN108039326B (en) | Method for setting scanning threshold value according to circuit design pattern | |
| CN108038841B (en) | Silicon wafer LD defect detection method and device | |
| JP4334527B2 (en) | Needle mark detection device and needle mark detection method | |
| JP4504612B2 (en) | Foreign matter inspection method and foreign matter inspection device | |
| EP2667354B1 (en) | Method and apparatus for analyzing stereoscopic or multi-view images | |
| JP2010019730A (en) | Critical flaw detecting method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |