OLED (Organic Light Emitting Diode) dispensing coating specification detection method, storage medium and equipment

Abstract

The invention relates to the technical field of OLED dispensing detection, in particular to an OLED dispensing coating specification detection method, a storage medium and equipment, and the method comprises the following steps: S10, collecting point cloud data; step S20, conveyor belt point cloud elimination based on straight-through filtering; step S30, product point cloud region segmentation based on point cloud registration; step S40, dividing a dispensing coating area based on European clustering; and S50, based on the dispensing coating specification detection of the point cloud minimum bounding box, whether the detection is passed or not is judged. According to the invention, related algorithms are designed, so that efficient and accurate OLED dispensing coating specification detection can be realized, and the production efficiency and the quality control level are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of OLED glue dispensing detection, and in particular to an OLED glue dispensing coating specification detection method, storage medium and equipment. Background Art

[0002] Glue dispensing refers to the process of preventing foreign matter and impurities from entering the circuit by dispensing glue at the OLED display panel and COF connector. This technology not only affects the performance and reliability of the product, but is also directly related to production efficiency and yield. Therefore, developing efficient and accurate detection technology has become one of the core needs for improving product quality. At present, glue dispensing detection technology mainly uses manual detection methods, relying on manual experience to judge the quality of glue dispensing. This method is not only inefficient, but also prone to human errors, and cannot meet the requirements of modern production for high precision and high efficiency.

[0003] Defects in the prior art (shortcomings and deficiencies in the prior art or products): With the advancement of machine vision and intelligent technology, detection systems based on traditional 2D vision and image processing are gradually applied to actual production. However, since most of the materials for dispensing and coating are transparent or highly transmittance colloid materials, the detection methods based on traditional 2D vision and image processing can only detect obvious defects, and have a high rate of missed detection for defects such as size specifications and slight glue breaks. This shows that the current automatic detection methods still have a lot of room for improvement in improving efficiency and accuracy. Therefore, the present invention aims to propose a 3D point cloud-based OLED dispensing and coating defect detection method to solve the problems of the existing methods. Summary of the invention

[0004] In view of the problems in the prior art, the present invention provides a method, storage medium and device for detecting the specifications of OLED glue dispensing coating, and designs relevant algorithms to realize efficient and accurate detection of the specifications of OLED glue dispensing coating, thereby improving production efficiency and quality control level.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] The present invention provides a method for detecting specifications of OLED dispensing coating, which is characterized by comprising the following steps:

[0007] Step S10: point cloud data collection;

[0008] Step S20, removing conveyor belt point cloud based on through-filtering;

[0009] Step S30, segmenting product point cloud regions based on point cloud registration;

[0010] Step S40, segmenting the glue coating area based on Euclidean clustering;

[0011] Step S50, the glue dispensing specification detection based on the minimum bounding box of the point cloud has been used to determine whether it has passed the detection;

[0012] The specific method of product point cloud region segmentation based on point cloud registration in step S30 is as follows: step S31, point cloud template preparation; step S32, point cloud rough registration based on SAC-IA; step S33, point cloud fine registration based on ICP; step S34, obtaining a new point cloud; based on the registration result, the boundary range is set with the point cloud template, and the point cloud contained outside the boundary is removed, and the obtained point cloud is the product point cloud region;

[0013] The specific method of the glue dispensing coating area segmentation based on Euclidean clustering in step S40 is as follows: step S41, setting the search radius; step S42, selecting the size interval of the cluster; step S43, selecting KD-tree as the search method.

[0014] Among them, the method of point cloud data collection in step S10 is: point cloud data collection is to obtain three-dimensional point cloud information of the product through a 3D linear array scanning camera, record millions of three-dimensional points in a short time, and form a point cloud data set containing shape, contour and surface details.

[0015] The method for removing conveyor belt point cloud based on through-filtering in step S20 includes the following steps:

[0016] Step S21, setting filtering parameters: first, defining the spatial range of the through-filter;

[0017] Step S22, apply filter: use the straight-through filtering algorithm to filter the point cloud data; remove all points that are not within the specified range and only keep the points that meet the conditions.

[0018] Among them, in step S31, the point cloud actually collected by the same sensor in the production line is used to make the point cloud template and the registered point cloud have the same size and point cloud density; then, the point cloud processing software is used to crop the glue coating area of ​​the point cloud data to obtain the point cloud template.

[0019] The method for rough registration of point clouds based on SAC-IA in step S32 is:

[0020] Step S321, solving the rotation matrix:

[0021] The rotation matrix R can be obtained by selecting the point pair (p s ,p t ) is calculated; assuming p s and p t are points in the source point cloud and the target point cloud respectively. The rotation matrix can be calculated by the following steps:

[0022] Calculate the centroid:

[0023]

[0024] Calculate the covariance matrix H

[0025]

[0026] Calculate the rotation matrix R by singular value decomposition (SVD)

[0027] [U,S,V T ]=SVD(H),R=VU T

[0028] Step S322, solving the translation matrix t:

[0029] The translation matrix calculation formula is as follows:

[0030] t=c t -R·c s .

[0031] The method for accurate point cloud registration based on ICP in step S33 is:

[0032] Step S331, setting the maximum number of iterations and the convergence threshold;

[0033] Step S332, inputting the pose matrix obtained by the rough registration as the initial registration matrix;

[0034] Step S333: nearest point matching, finding the nearest point from the target point cloud, and matching the two;

[0035] Step S334, calculate the covariance matrix H:

[0036]

[0037] Step S335: Calculate R * =VU T ,t * =u q -Ru q ,R=R * R,t=t+t * ;

[0038] Step S336, perform rigid transformation on the source point cloud until the maximum number of iterations set in the first step or the convergence threshold is met, and the final rotation matrix and translation matrix are output;

[0039] Step S337: Get a new point cloud:

[0040] P s'=RP s +t;

[0041] Based on the registration results, the boundary range is set with the point cloud template, and the point cloud outside the boundary is removed. The obtained point cloud is the product point cloud area.

[0042] The method for determining whether the glue dispensing specification detection based on the minimum bounding box of the point cloud in step S50 has been passed is as follows:

[0043] Step S511: Based on PCA principal component analysis;

[0044] The basic idea of ​​PCA is to find the direction with the greatest variation, which can be achieved by following the steps below:

[0045] Data centering: The coordinates of the data points are first averaged to facilitate centering. The specific method is to subtract the mean of all points in the point cloud. The specific principle and calculation method are as follows:

[0046] Assume that there are n points in the point cloud X = {x1, x2, ..., x n}, each point contains d dimensions, then the mean is:

[0047]

[0048] The centered point cloud is:

[0049] X centered =X-μ

[0050] Step S512: Calculate the covariance matrix: Calculate the covariance matrix using the centralized point cloud data. The covariance matrix can reflect the relationship between various features of the data. The specific principle and calculation method are as follows:

[0051]

[0052] Step S513: Eigenvalue decomposition: perform eigenvalue decomposition on the covariance matrix. The eigenvalue reflects the degree of variation in different directions, while the eigenvector represents the corresponding direction. The principle and formula for performing eigenvalue decomposition on the covariance matrix are as follows:

[0053] Cv i =λ i v i

[0054] Among them, λ i is the eigenvalue, v i is the corresponding eigenvector.

[0055] Step S514: Select the principal component: select the first few eigenvectors as the principal component according to the size of the eigenvalue, and select the eigenvector corresponding to the largest eigenvalue;

[0056] S52: minimum bounding box calculation;

[0057] Using the main direction and centroid obtained in the above steps, the point cloud is converted into coordinates based on the origin, and the coordinate axis is made to coincide with the main direction; then, the three-axis coordinate maximum values ​​Max(x,y,z) and Min(x,y,z) of the transformed point cloud are calculated; 8 coordinate points are arranged and combined according to the maximum values, and these 8 coordinate points are the vertices of the minimum bounding box;

[0058] S53: Geometric parameter analysis;

[0059] Step S531: Geometric parameter analysis: By analyzing the size (length, width, height) of the bounding box, it is determined whether the dispensing coating meets the predetermined specification standards; the size calculation method is as follows:

[0060]

[0061] Step S532: Compare and judge: Compare the size of the bounding box with the preset coating parameters (l h ,w h ,h h ) to check whether it is within the compliance range. If the following conditions are met at the same time, it meets the specifications:

[0062]

[0063] The present invention also provides a computer storage medium, wherein the computer storage medium stores computer instructions, and when the computer instructions are called, they are used to execute the OLED dispensing coating specification detection method.

[0064] The present invention also provides an electronic device, wherein the electronic device comprises: a processor; and a memory arranged to store computer executable instructions, wherein the executable instructions, when executed, cause the processor to execute the OLED dispensing coating specification detection method.

[0065] The present invention arranges 3D linear array cameras in the OLED production line, and combines with the automated production line to collect high-precision 3D point cloud data without modifying the equipment, thereby laying a foundation for automated detection; adopts a method based on through-filtering and Euclidean clustering to remove the point cloud of the conveyor belt and the point cloud of the flexible circuit board, and removes redundant point clouds while retaining the product point cloud, and segmenting the product panel point cloud, and then adopts Euclidean clustering to segment the product's glue coating area; solves the minimum bounding box of the glue coating area, and compares the size of the minimum bounding box with the set threshold value, and the product that meets the three conditions of length, width and height meets the specification requirements; the present invention designs a related algorithm to realize efficient and accurate OLED glue coating specification detection, thereby improving production efficiency and quality control level. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 The present invention is a flow chart of a method for detecting specifications of OLED dispensing coating.

[0067] Figure 2 This is a diagram of the equipment layout for applying the present invention. DETAILED DESCRIPTION

[0068] In order to facilitate the understanding of those skilled in the art, the present invention is further described below in conjunction with the embodiments and drawings, and the contents mentioned in the embodiments are not intended to limit the present invention. The present invention is described in detail below in conjunction with the drawings.

[0069] Embodiment 1

[0070] The first embodiment of the present application provides a method for detecting the specification of OLED dispensing coating, such as Figure 1-Figure 2 As shown, the following steps are included:

[0071] Step S10, point cloud data acquisition: Point cloud data acquisition is to obtain the three-dimensional point cloud information of the product through a 3D linear array scanning camera. This process is as follows Figure 2 As shown, it involves a conveyor belt and products moving at a constant speed, capturing detailed features of the surface of the scanned area at high resolution. The machine is able to record millions of three-dimensional points in a short period of time, forming a point cloud data set containing shape, contour and surface details;

[0072] Step S20, conveyor belt point cloud elimination based on through-filtering; conveyor belt point cloud elimination based on through-filtering is a commonly used point cloud processing technology. This method sets a spatial range so that only point cloud data that meets specific conditions are retained, thereby achieving effective redundant point cloud elimination; this application uses this method to remove unnecessary conveyor belt background information from the collected point cloud data;

[0073] Step S30, segmenting product point cloud regions based on point cloud registration;

[0074] Step S40, segmenting the glue coating area based on Euclidean clustering;

[0075] Step S50, the glue dispensing specification detection based on the minimum bounding box of the point cloud has been used to determine whether it has passed the detection;

[0076] The specific method of product point cloud region segmentation based on point cloud registration in step S30 is as follows: step S31, point cloud template preparation; step S32, point cloud rough registration based on SAC-IA; step S33, point cloud fine registration based on ICP; step S34, obtaining a new point cloud; based on the registration result, the boundary range is set with the point cloud template, and the point cloud contained outside the boundary is removed, and the obtained point cloud is the product point cloud region;

[0077] The specific method of the glue dispensing coating area segmentation based on Euclidean clustering in step S40 is as follows: step S41, setting the search radius; step S42, selecting the size interval of the cluster; step S43, selecting KD-tree as the search method.

[0078] Among them, the method of point cloud data collection in step S10 is: point cloud data collection is to obtain three-dimensional point cloud information of the product through a 3D linear array scanning camera, record millions of three-dimensional points in a short time, and form a point cloud data set containing shape, contour and surface details.

[0079] The method for removing conveyor belt point cloud based on through-filtering in step S20 includes the following steps:

[0080] Step S21, set filtering parameters: First, define the spatial range of the through-filter; this usually includes setting the minimum and maximum values ​​of the X, Y and Z axes to limit the point cloud area you want to retain. The actual application scenario of the present invention is the assembly line of products, and the point cloud data outside the product mainly includes the point cloud of the conveyor belt and the point cloud of the area on both sides of the X axis outside the conveyor belt. Therefore, the filtering parameters only need to set the X axis and Z axis. The conveyor belt point cloud is eliminated by collecting the Z-axis minimum value zmin when the conveyor belt is running stably, and the point cloud with a z value greater than zmin is eliminated. To eliminate the point cloud on both sides of the X axis outside the conveyor belt, it is only necessary to collect the X-axis interval where the conveyor belt point cloud is located when the conveyor belt is running stably, and those that are not in the interval are eliminated; Step S22, apply the filter: Use the through-filter algorithm to filter the point cloud data; eliminate all points that are not in the specified range, and only retain the points that meet the conditions. Through the above steps, point cloud elimination based on through-filtering can not only improve the efficiency of data processing, but also improve the accuracy and reliability of subsequent analysis.

[0081] Specifically, the main principle of the embodiment of the present application is that, since OLED dispensing coating defects have the characteristic of being transparent, it is difficult for traditional 2D visual images to identify and measure the specifications of dispensing coating, and 3D point cloud information is used instead of traditional 2D visual images as the input data for detection; the OLED panel without dispensing coating process is flat and smooth, while the OLED panel after dispensing coating has a certain height of glue; therefore, based on the difference before and after coating, corresponding equipment and methods can be designed to realize automatic detection of dispensing coating defects; the equipment layout of the embodiment of the present application is as follows Figure 2 As shown, after the glue dispensing and UV curing process, the OLED panel flows into the detection area of ​​the present invention via a conveyor belt, and the product information is collected by a 3D linear array camera.

[0082] The present invention arranges 3D linear array cameras in the OLED production line, and combines it with the automated production line without modifying the equipment, so as to collect high-precision 3D point cloud data, thereby laying a foundation for automated detection; adopts the method based on through-filtering and Euclidean clustering to remove the point cloud of the conveyor belt and the point cloud of the flexible circuit board, and removes the redundant point cloud while retaining the product point cloud, and segmenting the product panel point cloud, and then adopts Euclidean clustering to segment the product's glue coating area; solves the minimum bounding box of the glue coating area, and compares the size of the minimum bounding box with the set threshold value, and the product that meets the three conditions of length, width and height meets the specification requirements.

[0083] The embodiment of the present application has made relevant designs in terms of hardware equipment layout design and hardware selection, and uses a 3D linear array camera to collect 3D point cloud data of OLED products. Relevant algorithms are designed to achieve efficient and accurate OLED dispensing and coating specification detection, thereby improving production efficiency and quality control level.

[0084] In the embodiment of the present application, the point cloud actually collected by the same sensor in the production line is used in step S31, so that the point cloud template and the registered point cloud have the same size and point cloud density; then, the point cloud processing software is used to crop the glue coating area of ​​the point cloud data to obtain the point cloud template.

[0085] In the embodiment of the present application, the point cloud coarse registration based on SAC-IA (Sample Consensus Initial Alignment) is an effective point cloud alignment method, which aims to solve the registration problem between 3D point clouds collected from different perspectives or at different times, and can effectively deal with noise and mismatching problems, thereby obtaining a more accurate initial transformation matrix; however, due to the limited number of sampling point pairs, the initial transformation matrix may have certain errors; the method of point cloud coarse registration based on SAC-IA in step S32 is:

[0086] Step S321, solving the rotation matrix:

[0087] The rotation matrix R can be obtained by selecting the point pair (p s ,p t ) is calculated; assuming p s and p t are points in the source point cloud and the target point cloud respectively. The rotation matrix can be calculated by the following steps:

[0088] Calculate the centroid:

[0089]

[0090] Calculate the covariance matrix H:

[0091]

[0092] The rotation matrix R is calculated by singular value decomposition (SVD):

[0093] [U,S,V T ]=SVD(H),R=VU T

[0094] Step S322, solving the translation matrix t:

[0095] The translation matrix calculation formula is as follows:

[0096] t=c t -R·c s .

[0097] The method for accurate point cloud registration based on ICP in step S33 is:

[0098] Step S331, setting the maximum number of iterations and the convergence threshold;

[0099] Step S332, inputting the pose matrix obtained by the rough registration as the initial registration matrix;

[0100] Step S333: nearest point matching, finding the nearest point from the target point cloud, and matching the two;

[0101] Step S334, calculate the covariance matrix H:

[0102]

[0103] Step S335: Calculate R * =VU T ,t * =u q -Ru q ,R=R * R,t=t+t * ;

[0104] Step S336, perform rigid transformation on the source point cloud until the maximum number of iterations set in the first step or the convergence threshold is met, and the final rotation matrix and translation matrix are output;

[0105] Step S337: Get a new point cloud:

[0106] P s '=RP s +t;

[0107] Based on the registration results, the boundary range is set with the point cloud template, and the point cloud outside the boundary is removed. The obtained point cloud is the product point cloud area.

[0108] S40: Dispensing coating area segmentation based on Euclidean clustering:

[0109] By analyzing the point cloud of the product panel and the coating point cloud, it can be clearly observed that there is a significant difference in the values ​​of the two on the Z axis. In order to effectively segment the dispensing coating area, the Euclidean clustering algorithm can be used to segment the coating point cloud. This method measures the Euclidean distance between each point in the point cloud and clusters points with similar Z values ​​together, thereby accurately extracting the point cloud data of the coating area. The specific process is as follows:

[0110] S41: Search radius setting:

[0111] In the dispensing coating area segmentation based on Euclidean clustering, the setting of the search radius is an important step, which directly affects the clustering effect and the accuracy of segmentation. Due to the high precision of the linear array 3D camera, the density of the collected point cloud is large, and there is a significant difference in the value of the point cloud of the product panel and the coating point cloud on the Z axis. Therefore, the present invention uses a cross-validation method to perform multiple clustering through different radius values ​​to evaluate the stability and accuracy of the clustering results. The specific radius value setting principle uses the camera's theoretical point cloud distance m and the theoretical dispensing height n as reference standards, performs multiple clustering multiple times within the range of m±n, and selects the value with the best effect as the search radius value.

[0112] S42: Cluster size interval selection:

[0113] Select products that meet the dispensing coating specifications, and count the interval Q where the point cloud number of their point cloud coating area is located. The cluster size interval selects 10% of the extreme value of Q as the redundancy, that is, the minimum number of cluster points is Qmin-10%Qmin, and the maximum number of cluster points is Qmax+10%Qmax.

[0114] S43: Select KD-tree as the search method.

[0115] In the embodiment of the present application, the glue coating specification detection based on the minimum bounding box of the point cloud is a method for detecting the glue coating using point cloud data. By analyzing the size (length, width, height) of the bounding box and comparing it with the set value, it is determined whether the glue coating meets the predetermined specification standard. This method can effectively analyze the specification parameters of the glue coating; the glue coating specification detection based on the minimum bounding box of the point cloud in step S50 has been used to determine whether the method of passing the detection is:

[0116] Step S511: Based on PCA principal component analysis;

[0117] The basic idea of ​​PCA is to find the direction with the greatest variation, which can be achieved by following the steps below:

[0118] Data centering: The coordinates of the data points are first averaged to facilitate centering. The specific method is to subtract the mean of all points in the point cloud. The specific principle and calculation method are as follows:

[0119] Assume that there are n points in the point cloud X = {x1, x2, ..., x n}, each point contains d dimensions, then the mean is:

[0120]

[0121] The centered point cloud is:

[0122] X centered =X-μ

[0123] Step S512: Calculate the covariance matrix: Calculate the covariance matrix using the centralized point cloud data. The covariance matrix can reflect the relationship between various features of the data. The specific principle and calculation method are as follows:

[0124]

[0125] Step S513: Eigenvalue decomposition: perform eigenvalue decomposition on the covariance matrix. The eigenvalue reflects the degree of variation in different directions, while the eigenvector represents the corresponding direction. The principle and formula for performing eigenvalue decomposition on the covariance matrix are as follows:

[0126] Cv i =λ i v i

[0127] Among them, λ i is the eigenvalue, v i is the corresponding eigenvector.

[0128] Step S514: Select the principal component: select the first few eigenvectors as the principal component according to the size of the eigenvalue, and select the eigenvector corresponding to the largest eigenvalue;

[0129] S52: minimum bounding box calculation;

[0130] Using the main direction and centroid obtained in the above steps, the point cloud is converted into coordinates based on the origin, and the coordinate axis is made to coincide with the main direction; then, the three-axis coordinate maximum values ​​Max(x,y,z) and Min(x,y,z) of the transformed point cloud are calculated; 8 coordinate points are arranged and combined according to the maximum values, and these 8 coordinate points are the vertices of the minimum bounding box;

[0131] S53: Geometric parameter analysis;

[0132] Step S531: Geometric parameter analysis: By analyzing the size (length, width, height) of the bounding box, it is determined whether the dispensing coating meets the predetermined specification standards; the size calculation method is as follows:

[0133]

[0134] Step S532: Compare and judge: Compare the size of the bounding box with the preset coating parameters (l h ,w h ,h h ) to check whether it is within the compliance range. If the following conditions are met at the same time, it meets the specifications:

[0135]

[0136] The key points of the embodiments of this application are as follows:

[0137] Key point 1: By deploying 3D line array cameras in the OLED production line and combining them with the automated production line, high-precision 3D point cloud data can be collected without modifying the equipment, laying the foundation for automated inspection.

[0138] Key point 2: The point cloud of the conveyor belt and the point cloud of the flexible circuit board are eliminated by using the methods based on straight-through filtering and Euclidean clustering. While retaining the product point cloud, the redundant point cloud is eliminated and the product panel point cloud is segmented. Then, Euclidean clustering is used to segment the product's glue coating area.

[0139] Key point 3: Find the minimum bounding box of the dispensing area, and compare the size of the minimum bounding box with the set threshold. Products that meet the three conditions of length, width, and height meet the specification requirements.

[0140] Embodiment 2

[0141] Embodiment 2 of the present application provides a computer storage medium, wherein the computer storage medium stores computer instructions, and when the computer instructions are called, they are used to execute the OLED dispensing coating specification detection method.

[0142] Embodiment 3

[0143] Embodiment 3 of the present application provides an electronic device, wherein the electronic device includes: a processor; and a memory arranged to store computer executable instructions, wherein the executable instructions, when executed, cause the processor to execute the OLED dispensing coating specification detection method.

[0144] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention is disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technician familiar with the profession can make some changes or modifications to equivalent embodiments of equivalent changes by using the technical contents disclosed above without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent changes and modifications made to the above embodiments according to the technology of the present invention are within the scope of the technical solution of the present invention without departing from the content of the technical solution of the present invention.

Claims

1. A method for detecting OLED dispensing coating specifications, characterized in that: The following steps are involved: Step S10, point cloud data collection; Step S20, removing conveyor belt point cloud based on through-filtering; Step S30, segmenting product point cloud regions based on point cloud registration; Step S40, segmenting the glue coating area based on Euclidean clustering; Step S50, the glue dispensing specification detection based on the minimum bounding box of the point cloud has been used to determine whether it has passed the detection; The specific method of product point cloud region segmentation based on point cloud registration in step S30 is as follows: step S31, point cloud template preparation; step S32, point cloud rough registration based on SAC-IA; step S33, point cloud fine registration based on ICP; step S34, obtaining a new point cloud; based on the registration result, the boundary range is set with the point cloud template, and the point cloud contained outside the boundary is removed, and the obtained point cloud is the product point cloud region; The specific method of the glue dispensing coating area segmentation based on Euclidean clustering in step S40 is as follows: step S41, setting the search radius; step S42, selecting the size interval of the cluster; step S43, selecting KD-tree as the search method.

2. The method for detecting OLED dispensing coating specifications according to claim 1, characterized in that: The method for collecting point cloud data in step S10 is as follows: point cloud data collection is to obtain three-dimensional point cloud information of the product through a 3D linear array scanning camera, record millions of three-dimensional points in a short time, and form a point cloud data set containing shape, contour and surface details.

3. The method for detecting OLED dispensing coating specifications according to claim 1, characterized in that: The method for removing conveyor belt point cloud based on through-filtering in step S20 comprises the following steps: Step S21, setting filtering parameters: first, defining the spatial range of the through-filter; Step S22, apply filter: use the straight-through filtering algorithm to filter the point cloud data; remove all points that are not within the specified range and only keep the points that meet the conditions.

4. The method for detecting OLED dispensing coating specifications according to claim 1, characterized in that: In step S31, the point cloud actually collected by the same sensor in the production line is used to make the point cloud template and the registered point cloud have the same size and point cloud density; Afterwards, point cloud processing software is used to crop the glue coating area of ​​the point cloud data to obtain a point cloud template.

5. The method for detecting OLED dispensing coating specifications according to claim 1, characterized in that: The method for rough registration of point clouds based on SAC-IA in step S32 is: Step S321, solving the rotation matrix: The rotation matrix R can be obtained by selecting the point pair (p s ,p t ) is calculated; assuming p s and p t are points in the source point cloud and the target point cloud respectively. The rotation matrix can be calculated by the following steps: Calculate the centroid: Calculate the covariance matrix H Calculate the rotation matrix R by singular value decomposition (SVD) [U,S,V T ]=SVD(H),R=VU T Step S322, solving the translation matrix t: The translation matrix calculation formula is as follows: t=c t -R·c s 。 6. The method for detecting OLED dispensing coating specifications according to claim 1, characterized in that: The method for accurate point cloud registration based on ICP in step S33 is: Step S331, setting the maximum number of iterations and the convergence threshold; Step S332, inputting the pose matrix obtained by the rough registration as the initial registration matrix; Step S333: nearest point matching, finding the nearest point from the target point cloud, and matching the two; Step S334, calculate the covariance matrix H: Step S335: Calculate R * =VU T ,t * =u q -Ru q ,R=R * R,t=t+t * ; Step S336, perform rigid transformation on the source point cloud until the maximum number of iterations set in the first step or the convergence threshold is met, and the final rotation matrix and translation matrix are output; Step S337: Get a new point cloud: P s '=RP s +t; Based on the registration results, the boundary range is set with the point cloud template, and the point cloud outside the boundary is removed. The obtained point cloud is the product point cloud area.

7. The method for detecting OLED dispensing coating specifications according to claim 1, characterized in that: The method for determining whether the glue dispensing specification detection based on the minimum bounding box of the point cloud in step S50 has been passed is: Step S511: based on PCA principal component analysis; The basic idea of ​​PCA is to find the direction with the greatest variation, which can be achieved by following the steps below: Data centering: The coordinates of the data points are first averaged to facilitate centering. The specific method is to subtract the mean of all points in the point cloud. The specific principle and calculation method are as follows: Assume that there are n points in the point cloud X = {x1, x2, ..., x n }, each point contains d dimensions, then the mean is: The centered point cloud is: X centered =X-μ Step S512: Calculate the covariance matrix: Calculate the covariance matrix using the centralized point cloud data. The covariance matrix can reflect the relationship between various features of the data. The specific principle and calculation method are as follows: Step S513: Eigenvalue decomposition: perform eigenvalue decomposition on the covariance matrix. The eigenvalue reflects the degree of variation in different directions, while the eigenvector represents the corresponding direction. The principle and formula for performing eigenvalue decomposition on the covariance matrix are as follows: Cv i =λ i v i Among them, λ i is the eigenvalue, v i is the corresponding eigenvector. Step S514: Select the principal component: select the first few eigenvectors as the principal component according to the size of the eigenvalue, and select the eigenvector corresponding to the largest eigenvalue; S52: minimum bounding box calculation; Using the main direction and centroid obtained in the above steps, the point cloud is converted into coordinates based on the origin, and the coordinate axis is made to coincide with the main direction; then, the three-axis coordinate maximum values ​​Max(x,y,z) and Min(x,y,z) of the transformed point cloud are calculated; 8 coordinate points are arranged and combined according to the maximum values, and these 8 coordinate points are the vertices of the minimum bounding box; S53: Geometric parameter analysis; Step S531: Geometric parameter analysis: By analyzing the size (length, width, height) of the bounding box, it is determined whether the dispensing coating meets the predetermined specification standards; the size calculation method is as follows: Step S532: Compare and judge: Compare the size of the bounding box with the preset coating parameters (l h ,w h ,h h ) to check whether it is within the compliance range. If the following conditions are met at the same time, it meets the specifications:

8. A computer storage medium, wherein the computer storage medium stores computer instructions, and when the computer instructions are called, they are used to execute the OLED dispensing coating specification detection method as described in any one of claims 1-7.

9. An electronic device, wherein: The electronic device comprises: a processor; and a memory arranged to store computer executable instructions, wherein when the executable instructions are executed, the processor executes a method for detecting specifications of OLED dispensing coating as described in any one of claims 1 to 7.

Images