A contour-based 3D glue abnormality detection method and device

Abstract

The application provides a contour-based 3D glue coating anomaly detection method and device, wherein the method comprises the following steps: projecting three-dimensional point cloud data of a glue coating contour on a plane to obtain a projection data set, calculating a unit direction vector of two points based on random two points in the projection data set, dividing a maximum inner point set according to the normalized distance of all points in the projection data set to the unit direction vector, fitting a direction vector according to the maximum inner point set to obtain a reference straight line, correcting the projection data set through rotation and translation of the reference straight line to obtain a target point cloud, measuring the glue coating contour based on the target point cloud to obtain a measurement sequence, detecting the glue coating contour according to the measurement sequence to obtain a glue coating anomaly detection result. The application can correct the contour data, improve the accuracy and robustness of the detection result, and detect the contour data with multiple curve characteristics, thereby expanding the application range of the detection method.

Description

Technical Field

[0001] This invention relates to the field of adhesive coating anomaly detection technology, and in particular to a contour-based 3D adhesive coating anomaly detection method and apparatus. Background Technology

[0002] Adhesive application technology is widely used by major automobile manufacturers and auto parts suppliers. It is commonly seen in the body-in-white and hood application processes in automotive welding workshops, and in windshield applications in final assembly workshops. Its quality directly affects the vehicle's shock absorption, noise reduction, and windproof performance. In the automotive manufacturing industry, adhesive application quality inspection requirements mainly include adhesive width, adhesive height, and adhesive breaks. Some manufacturers also focus on detecting abnormal adhesive application trajectories and positional deviations (such as abnormalities in the reverse-engineered robotic arm), and have certain requirements for the inspection duration and accuracy.

[0003] Currently, the mainstream non-contact adhesive application inspection solution based on machine vision uses laser triangulation for 3D contour measurement. This allows for real-time, continuous, high-speed, and high-precision reconstruction of the 3D contour data of the adhesive application, which is then used to meet subsequent inspection requirements. The inspection solution based on a 3D line laser contour measuring instrument utilizes laser triangulation, supplemented by a high-precision camera and optical plane calibration method, to obtain highly accurate adhesive application contour data, which is then used to detect height, width, and adhesive breaks. While this solution offers high accuracy, the limitations of the space around the glue gun and the arbitrary spraying direction result in blind spots in some areas, making inspection impossible. Therefore, multiple laser contour measuring instruments are typically used to form a multi-view adhesive application inspection system to completely cover all adhesive application areas.

[0004] However, although multi-array laser profile measuring system can solve the problem of blind spots, the shapes and uneven surfaces of the coated body, hood, and windshield in the final assembly workshop cause the profiles scanned by the line laser profile measuring instrument to have a certain tilt angle in space. Furthermore, the same three-dimensional profile has multiple curve features, such as ellipse, straight line, and right angle, which makes it difficult to detect the height and width of the coated profile, resulting in an increase in abnormal detection.

[0005] Therefore, there is an urgent need for a method to detect adhesive coating anomalies that can correct contour data, measure contour data with multi-curve features, and improve detection accuracy. Summary of the Invention

[0006] Therefore, it is necessary to provide a contour-based 3D adhesive coating anomaly detection method and apparatus to address the aforementioned technical problems.

[0007] A contour-based 3D adhesive coating anomaly detection method includes the following steps: projecting the 3D point cloud data of the adhesive coating contour onto a plane to obtain a projection dataset; calculating the unit direction vector passing through two random points in the projection dataset; dividing the projection dataset into a maximum interior point set based on the standardized distance from all points in the projection dataset to the unit direction vector; obtaining a reference line based on the direction vector fitted to the maximum interior point set; performing rotation and translation correction on the projection dataset using the reference line to obtain a target point cloud; measuring the adhesive coating contour based on the target point cloud to obtain a measurement sequence; and detecting the adhesive coating contour based on the measurement sequence to obtain an adhesive coating anomaly detection result.

[0008] In one embodiment, the step of projecting the 3D point cloud data of the adhesive coating contour onto a plane to obtain a projection dataset, and calculating the unit direction vector passing through two random points in the projection dataset, includes: randomly selecting two distinct points p1 and p2 in the projection dataset P of the adhesive coating contour, with corresponding coordinates (x1, z1) and (x2, z2) respectively, and then calculating the direction vector of the line passing through p1 and p2. for:

[0009]

[0010] Calculate the direction vector unit direction vector for:

[0011]

[0012] In one embodiment, the step of dividing the projection dataset into a maximum inlier set based on the standardized distances from all points in the projection dataset to the unit direction vector, and obtaining a reference line based on the direction vector fitted by the maximum inlier set, includes: calculating the standardized distances from all points in the projection dataset to the unit direction vector, using the formula:

[0013]

[0014] The number of points in the projection dataset whose standardized distances to the unit direction vector satisfy the preset conditions is counted. Specifically, when the standardized distance is less than the preset distance, the corresponding points are divided into an interior set; when the standardized distance is greater than or equal to the preset distance, the corresponding points are divided into an exterior set. The size of the current interior set is denoted as n. i :

[0015] n i =|P inner |=|{l i <τ|i∈N +}| (4)

[0016] Repeated statistics are performed to obtain multiple interior point sets of sizes n1, n2, n3, ..., n. k The largest interior set n is obtained by selecting it. max ,for:

[0017]

[0018] Calculate the maximum interior set n max Corresponding direction vector The direction vector of the optimal fitted line is obtained as follows:

[0019]

[0020] The reference line is obtained based on the direction vector of the optimal fitted line.

[0021] In one embodiment, the step of rotating and translating the projection dataset using the reference line to obtain the target point cloud includes: obtaining the angle between the reference line and the coordinate axis by inversely solving the tangent of the slope of the reference line, which is:

[0022]

[0023] Based on the included angle, construct a two-dimensional rotation matrix for clockwise rotation as follows:

[0024]

[0025] Rotate the projected dataset in space around the coordinate axes by the same angle to obtain the rotated dataset:

[0026] P′=RP (9)

[0027] The translation coefficient is calculated using the following formula:

[0028]

[0029] The reference line is translated according to the translation coefficient until it coincides with the coordinate axis, resulting in the target point cloud:

[0030] P target =RP+T (11)

[0031] In one embodiment, the step of measuring the adhesive coating contour based on the target point cloud and obtaining a measurement sequence includes: measuring the height and width of the adhesive coating contour according to the target point cloud, and calculating the area of ​​the adhesive coating contour using numerical integration; and obtaining the corresponding measurement sequence based on the number of contours reconstructed by laser scanning during the adhesive coating process and the measurement data of the height, width and area of ​​the adhesive coating contour.

[0032] In one embodiment, obtaining the corresponding measurement sequence based on the measurement data of the number of contours reconstructed by laser scanning during the adhesive application process and the height, width, and area of ​​the adhesive contours includes: assuming the complete 3D adhesive application trajectory includes M contours, then three measurement sequences of size M are obtained respectively:

[0033] In one embodiment, the step of detecting the adhesive coating contour based on the measurement sequence to obtain an adhesive coating anomaly detection result includes: calculating the difference sequence between the two measurement sequences before and after correction, using the following formula:

[0034]

[0035] Based on the law of large numbers and the central limit theorem, the difference sequence is assumed to follow a Gaussian distribution. According to the distribution characteristics of the Gaussian distribution, the probability value of data in the difference sequence within the interval (u-3δ, u+3δ) is calculated using the following formula:

[0036]

[0037] Transform Summarized as follows:

[0038]

[0039] Solving for:

[0040]

[0041] The detection result of adhesive coating anomaly is obtained based on the magnitude of the probability value.

[0042] A contour-based 3D adhesive coating anomaly detection device is used to implement the contour-based 3D adhesive coating anomaly detection method described above. The device includes: a direction vector calculation module for projecting the 3D point cloud data of the adhesive coating contour onto a plane to obtain a projection dataset, and calculating a unit direction vector passing through two random points in the projection dataset; a reference line acquisition module for dividing the projection dataset into a maximum interior point set based on the standardized distance from all points in the projection dataset to the unit direction vector, and obtaining a reference line based on the direction vector fitted to the maximum interior point set; a projection dataset correction module for performing rotation and translation correction on the projection dataset using the reference line to obtain a target point cloud; an adhesive coating contour measurement module for measuring the adhesive coating contour based on the target point cloud to obtain a measurement sequence; and an adhesive coating anomaly detection module for detecting the adhesive coating contour according to the measurement sequence to obtain an adhesive coating anomaly detection result.

[0043] Compared with existing technologies, the advantages and beneficial effects of this invention are as follows: The three-dimensional point cloud data of the adhesive coating contour is projected onto a plane to obtain a projection dataset. The unit direction vector passing through two random points in the projection dataset is calculated. Based on the standardized distance from all points in the projection dataset to the unit direction vector, a maximum interior point set is obtained. A reference line is obtained based on the direction vector fitted to the maximum interior point set. The projection dataset is then corrected by rotation and translation using the reference line to obtain the target point cloud. This correction ensures the accuracy of the contour data. The adhesive coating contour is measured based on the target point cloud to obtain a measurement sequence. This enables the detection of adhesive coating trajectories with multi-curve features, expanding the applicability of the detection method. The adhesive coating contour is detected based on the measurement sequence to obtain anomaly detection results. Anomaly detection of the adhesive coating contour is performed based on the corrected contour data, improving the accuracy and robustness of the detection results. Attached Figure Description

[0044] Figure 1 This is a flowchart illustrating a contour-based 3D adhesive coating anomaly detection method in one embodiment.

[0045] Figure 2 This is a screenshot of the detection result for uncorrected contour data in one embodiment;

[0046] Figure 3 This is a diagram showing the detection result after contour data correction in one embodiment;

[0047] Figure 4 This is a schematic diagram of a contour-based 3D adhesive coating anomaly detection device in one embodiment. Detailed Implementation

[0048] Before describing the specific embodiments of the present invention, the overall concept of the present invention will be explained as follows:

[0049] This invention is mainly based on the development of 3D coating process. Currently, the coating process has problems such as uneven coating surface and many curved features in the contour, which make it difficult to detect the coating height and width.

[0050] Therefore, this invention proposes a contour-based 3D adhesive coating anomaly detection method. The method involves projecting the 3D point cloud data of the adhesive coating contour onto the xoz plane to obtain a projection dataset. The unit direction vector passing through two random points in the projection dataset is calculated. Based on the standardized distance from all points in the projection dataset to the unit direction vector, a maximum interior point set is obtained. A reference line is obtained based on the direction vector fitted to the maximum interior point set. The projection dataset is then corrected for rotation and translation using the reference line to obtain the target point cloud. The adhesive coating contour is measured based on the target point cloud to obtain a measurement sequence. The target point cloud is then detected based on the measurement sequence to obtain the adhesive coating anomaly detection result. Correcting the adhesive coating contour data facilitates measurement of the adhesive coating contour and improves the accuracy and reliability of the adhesive coating anomaly detection result.

[0051] Having introduced the overall concept of the present invention, to make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below through specific embodiments in conjunction with the accompanying drawings. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0052] In one embodiment, such as Figure 1 As shown, a contour-based 3D adhesive coating anomaly detection method is provided, including the following steps:

[0053] Step S101: Project the 3D point cloud data of the adhesive coating contour onto the plane to obtain the projection dataset, and calculate the unit direction vector passing through the two random points in the projection dataset.

[0054] Specifically, since the detection of 3D adhesive coating contours mainly uses width and height data, the 3D point cloud data of the adhesive coating contours can be projected onto the xoz plane to obtain a projection dataset, which facilitates subsequent processing. Two points are randomly selected in the projection dataset, and the unit direction vectors passing through the two points are calculated. This allows the normalized distances of the remaining points in the projection dataset to be calculated based on the unit direction vectors.

[0055] Step S101 includes: randomly selecting two distinct points p1 and p2 from the projection dataset P of the adhesive coating contour, with corresponding coordinates (x1, z1) and (x2, z2) respectively; then the direction vector of the line passing through p1 and p2... for:

[0056]

[0057] Calculate the direction vector unit direction vector for:

[0058]

[0059] Specifically, two distinct points p1 and p2 are randomly selected from the projection dataset of the glue coating contour, and their corresponding coordinates (x1, z1) and (x2, z2) are obtained. The direction vector of the straight line passing through the two points is calculated based on the coordinates of the two points, and the corresponding unit direction vector is calculated based on the direction vector, so as to calculate the standardized distance of the points in the projection dataset according to the direction vector.

[0060] Step S102: Based on the standardized distance from all points in the projection data to the unit direction vector, the maximum interior point set is obtained, and the reference line is obtained based on the direction vector fitted by the maximum interior point set.

[0061] Specifically, although the unevenness in the coating scenario results in a large number of curved features in the scanned contour data, all coating contour curves contain a reference straight line. Even if the position of the coating part is uncertain and the surface is uneven, causing this straight line to present different postures in three-dimensional space, it can basically reflect the spatial directional characteristics of all curves in all contours. Therefore, the reference straight line can be obtained from the various curves of the contour, so as to correct the coating contour according to the reference straight line.

[0062] After obtaining the unit direction vector, the normalized distance from the other points in the projection dataset to the unit direction vector is calculated sequentially based on the unit direction vector. The points in the projection dataset can be classified based on the normalized distance, dividing them into an interior point set and an exterior point set. After dividing all points, the maximum interior point set is obtained. The direction vector is obtained by fitting the maximum interior point set, and the line corresponding to the direction vector is used as the reference line. The reference line can be used to correct the glue coating contour, thereby facilitating the detection of glue coating height and width.

[0063] Step S102 includes: calculating the standardized distance from all points in the projected dataset to the unit direction vector, using the formula:

[0064]

[0065] In the statistical projection dataset, count the number of points whose standardized distances to the unit direction vector satisfy a preset condition. Specifically, if the standardized distance is less than the preset distance, the corresponding points are assigned to an interior set; if the standardized distance is greater than or equal to the preset distance, the corresponding points are assigned to an exterior set. Let the size of the current interior set be n. i :

[0066] n i =|P inner |=|{l i <τ|i∈N +}| (4)

[0067] Repeatedly count the numbers to obtain n1, n2, n3, ..., n k The largest interior set n is obtained by selecting it. max ,for:

[0068]

[0069] Calculate the maximum interior set n max Corresponding direction vector The direction vector of the optimal fitted line is obtained as follows:

[0070]

[0071] The reference line is obtained based on the direction vector of the best-fit line.

[0072] Specifically, when acquiring the baseline line, the standardized distance from all points in the projection dataset to the unit direction vector is calculated. The number of points whose standardized distance meets a preset requirement is that the standardized distance is less than a preset distance, which can be set according to actual conditions. When the standardized distance of a point is less than the preset distance, the point is assigned to the inlier set; when the standardized distance is greater than or equal to the preset distance, the point is assigned to the outer set, and the current size of the inlier set is recorded. This statistical process is repeated to obtain several inlier sets. The largest inlier set is selected, and a direction vector is fitted based on the largest inlier set. This is the direction vector of the optimal fitted line. The equation of the baseline line is obtained based on this direction vector, which facilitates subsequent correction of the projection dataset based on the baseline line, ensuring the accuracy of the contour data.

[0073] Step S103: Rotate and translate the projection dataset using the reference line to obtain the target point cloud.

[0074] Specifically, after obtaining the reference line from the projection dataset, the projection dataset is corrected using the reference line. The correction includes rotation and translation. The main purpose is to transform the projection dataset so that the reference line is parallel and coincident with the coordinate axis, ensuring that the corrected projection data conforms to the adhesive coating contour and improving the accuracy of anomaly detection results.

[0075] Step S103 includes: obtaining the angle between the reference line and the coordinate axis by inversely solving for the tangent of the slope of the reference line, which is:

[0076]

[0077] Construct a two-dimensional rotation matrix for clockwise rotation based on the included angle, as follows:

[0078]

[0079] Rotate the projected dataset in space around the coordinate axes by the same angle to obtain the rotated dataset:

[0080] P′=RP (9)

[0081] The translation coefficient is calculated using the following formula:

[0082]

[0083] The reference line is translated to coincide with the coordinate axes based on the translation coefficient, resulting in the target point cloud:

[0084] P target =RP+T (11)

[0085] Specifically, during the correction of the projected dataset, the angle between the reference line and the coordinate axes is obtained by inversely solving for the tangent of the slope of the reference line, for example, the angle with the x-axis is α. A two-dimensional rotation matrix is ​​constructed based on this angle; note that this is a clockwise rotation. The projected dataset P is rotated by α around the x-axis in space to obtain P′, completing the translation operation. At this point, the reference line is rotated to be parallel to the x-axis. By calculating the translation coefficients, the reference line is moved to coincide with the x-axis, obtaining the target point cloud. The target point cloud can improve the accuracy of subsequent anomaly detection results.

[0086] The unevenness and tilt of the workpiece surface can lead to significant errors when detecting the characteristics of the adhesive strip, such as its width and height. Figure 2 As shown, when performing height and width detection, the gray point is the highest point found and used to calculate the height of the adhesive strip, and the two black points are the reference points for calculating the width of the adhesive strip. Obviously, due to the tilt of the contour, the height detection is incorrect, which leads to the width detection being incorrect as well. Such erroneous data will also cause a large error in the subsequent anomaly detection based on the adhesive strip features, interfering with the final detection result.

[0087] like Figure 3 As shown, after performing the above-mentioned rotation and translation contour correction, the tilted contour of the adhesive strip is corrected to a relatively horizontal state. The gray point used to detect the highest point of height and the two reference points for calculating the width are both correctly searched, which greatly improves the detection accuracy of the adhesive strip width and height.

[0088] Step S104: Measure the adhesive coating contour based on the target point cloud and obtain a measurement sequence.

[0089] Specifically, after obtaining the transformed target point cloud, the height and width of the adhesive coating contour are measured based on the target point cloud, and the area of ​​the contour is calculated using numerical integration. Based on the measurement data such as the height, width, and area of ​​the adhesive coating contour, the corresponding measurement sequence is obtained, thereby enabling the detection of adhesive coating scenes with multiple contours, improving the accuracy of the detection results, and expanding the applicability of the detection method.

[0090] Step S104 includes: measuring the height and width of the adhesive coating contour based on the target point cloud, and calculating the area of ​​the adhesive coating contour using numerical integration; and obtaining the corresponding measurement sequence based on the number of contours reconstructed by laser scanning during the adhesive coating process and the measurement data of the height, width and area of ​​the adhesive coating contour.

[0091] Specifically, after obtaining the target point cloud through rotation and translation transformation, the height and width of the adhesive coating contour are measured based on the target point cloud, and the area of ​​the contour is calculated based on the height and width of the contour. Combining the number of contours obtained through 3D reconstruction by laser scanning during the adhesive coating process with the measurement data such as the height, width, and area of ​​the adhesive coating contour, measurement sequences corresponding to the height, width, and area are obtained respectively. Through the measurement sequences, multi-dimensional detection of height anomalies, width anomalies, and area anomalies of adhesive coating trajectories with multi-curve features can be achieved, ensuring the accuracy of the detection results and the applicability of the detection method.

[0092] Assuming the complete 3D adhesive application trajectory comprises M contours, three measurement sequences of size M are obtained:

[0093]

[0094] Specifically, when the 3D adhesive application trajectory includes M contours, there are M corresponding heights, widths, and areas, forming three measurement sequences of size M, which facilitates the detection of anomalies in the adhesive application height, width, and area.

[0095] Step S105: Detect the adhesive coating contour according to the measurement sequence to obtain the adhesive coating anomaly detection result.

[0096] Specifically, during the glue coating inspection, the height, width, and area of ​​the glue coating contour are measured according to the obtained measurement sequence to obtain the glue coating anomaly detection results, ensuring the reliability of the detection results and effectively improving the accuracy and robustness of glue coating anomaly detection.

[0097] The steps for obtaining the glue coating anomaly detection results are as follows: Calculate the difference sequence between the two measurement sequences before and after correction, using the following formula:

[0098]

[0099] Combining the law of large numbers and the central limit theorem, the difference sequence is assumed to follow a Gaussian distribution. Based on the distribution characteristics of the Gaussian distribution, the probability value of data in the difference sequence within the interval (u-3δ, u+3δ) is calculated using the following formula:

[0100]

[0101] Transform Summarized as follows:

[0102]

[0103] Solving for:

[0104]

[0105] The detection result of adhesive coating anomaly is obtained based on the magnitude of the probability value.

[0106] Specifically, taking the corrected height data as an example, the height difference sequence is calculated based on the two height sequences before and after correction. Considering the distribution pattern of the data in the difference sequence, the normal data is distributed near the mean, and a small number of outlier abnormal data are distributed outside a certain range of the mean. Based on this consideration, it is only necessary to solve for this range threshold to detect the abnormal data that occurs during glue application.

[0107] Therefore, by combining the probability distribution of the Gaussian distribution, the probability of the sequence falling within the corresponding interval can be calculated. Based on the calculated probability value, it can be compared with a preset probability threshold. If the probability value is less than the threshold, the adhesive application is considered normal; if the probability value is greater than or equal to the threshold, the adhesive application is considered abnormal. Furthermore, considering the influence of noise, a highly abnormal adhesive application can be determined when multiple consecutive abnormal data points are detected.

[0108] The above methods are used to detect adhesive application anomalies in width and area, and obtain corresponding test results. Through multi-dimensional detection methods, the robustness of the test results and applicability to multiple scenarios are ensured.

[0109] In this embodiment, the 3D point cloud data of the adhesive coating contour is projected onto a plane to obtain a projection dataset. The unit direction vector passing through two random points in the projection dataset is calculated. Based on the standardized distance from all points in the projection dataset to the unit direction vector, a maximum interior point set is obtained. A reference line is obtained based on the direction vector fitted to the maximum interior point set. The projection dataset is then corrected by rotation and translation using the reference line to obtain the target point cloud. This correction ensures the accuracy of the contour data. The adhesive coating contour is measured based on the target point cloud to obtain a measurement sequence. This enables the detection of adhesive coating trajectories with multi-curve features, expanding the applicability of the detection method. The adhesive coating contour is then detected based on the measurement sequence to obtain anomaly detection results. Anomaly detection of the adhesive coating contour can be performed based on the corrected contour data, improving the accuracy and robustness of the detection results.

[0110] like Figure 4 As shown, a contour-based 3D adhesive coating anomaly detection device 40 is provided to implement the contour-based 3D adhesive coating anomaly detection method described above. It includes: a direction vector calculation module 41, a reference line acquisition module 42, a projection dataset correction module 43, an adhesive coating contour measurement module 44, and an adhesive coating anomaly detection module 45, wherein:

[0111] The direction vector calculation module 41 is used to project the three-dimensional point cloud data of the adhesive outline onto a plane, obtain the projection dataset, and calculate the unit direction vector passing through the two random points in the projection dataset.

[0112] The reference line acquisition module 42 is used to divide the maximum interior point set according to the standardized distance from all points in the projection dataset to the unit direction vector, and to obtain the reference line according to the direction vector fitted by the maximum interior point set.

[0113] The projection dataset correction module 43 is used to perform rotation and translation correction on the projection dataset using a reference line to obtain the target point cloud.

[0114] The adhesive coating contour measurement module 44 is used to measure the adhesive coating contour based on the target point cloud and obtain a measurement sequence;

[0115] The adhesive coating anomaly detection module 45 is used to detect the adhesive coating contour according to the measurement sequence and obtain the adhesive coating anomaly detection result.

[0116] In one embodiment, the device further includes: a measurement calculation module, used to measure the height and width of the adhesive coating contour based on the target point cloud, and calculate the area of ​​the adhesive coating contour using numerical integration; and a measurement sequence acquisition module, used to acquire the corresponding measurement sequence based on the measurement data of the number of contours reconstructed by laser scanning during the adhesive coating process and the height, width, and area of ​​the adhesive coating contour.

[0117] It is obvious to those skilled in the art that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a computer storage medium (ROM / RAM, magnetic disk, optical disk) for execution by the computing device. In some cases, the steps shown or described can be performed in a different order than those described herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Therefore, the present invention is not limited to any particular hardware and software combination.

[0118] The above description, in conjunction with specific embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the inventive concept, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A method for detecting 3D adhesive coating anomalies based on contours, characterized in that, Includes the following steps: The three-dimensional point cloud data of the adhesive coating contour is projected onto a plane to obtain a projection dataset. The unit direction vector passing through the two random points in the projection dataset is calculated. Based on the standardized distances from all points in the projection dataset to the unit direction vector, a maximum interior point set is obtained. Based on the direction vector fitted to the maximum interior point set, a reference line is obtained. The target point cloud is obtained by rotating and translating the projection dataset using the reference line; The adhesive coating contour is measured based on the target point cloud to obtain a measurement sequence, including: measuring the height and width of the adhesive coating contour according to the target point cloud, and calculating the area of ​​the adhesive coating contour using numerical integration; obtaining the corresponding measurement sequence based on the number of contours reconstructed by laser scanning during the adhesive coating process and the measurement data of the height, width, and area of ​​the adhesive coating contour. Assuming that the complete 3D adhesive coating trajectory includes M contours, three measurement sequences of size M are obtained respectively: ； The adhesive coating profile is detected based on the measurement sequence to obtain adhesive coating anomaly detection results, including: The formula for calculating the difference between the two measurement sequences before and after correction is: ， Based on the law of large numbers and the central limit theorem, the difference sequence is assumed to follow a Gaussian distribution. According to the distribution characteristics of the Gaussian distribution, the data in the difference sequence within the interval [missing information] are calculated. The probability value within is given by the formula: ， Transform After sorting, we get: ， Solving for: ， The detection result of adhesive coating anomaly is obtained based on the magnitude of the probability value.

2. The contour-based 3D coating anomaly detection method according to claim 1, characterized in that, The step of projecting the 3D point cloud data of the adhesive coating contour onto a plane to obtain a projection dataset, and calculating the unit direction vector passing through two random points in the projection dataset, includes: Randomly select two distinct points from the projection dataset P of the glue coating contour. and The corresponding coordinates are respectively and Then it is too late. and The direction vector of the straight line ,for: ， Calculate the direction vector unit direction vector ,for: 。 3. The contour-based 3D adhesive coating anomaly detection method according to claim 2, characterized in that, The step of partitioning the projection dataset into a maximum interior point set based on the standardized distances from all points in the projection dataset to the unit direction vector, and obtaining the reference line based on the direction vector fitted by the maximum interior point set, includes: The standardized distance from all points in the projected dataset to the unit direction vector is calculated using the following formula: ， The number of points in the projected dataset whose standardized distances to the unit direction vector satisfy a preset condition is counted. Specifically, when the standardized distance is less than a preset distance, the corresponding points are divided into an interior set; when the standardized distance is greater than or equal to the preset distance, the corresponding points are divided into an exterior set. The current size of the interior set is denoted as . : ， Repeated statistics were performed to obtain the sizes of multiple interior point sets. The largest interior set is obtained by selecting it. ,for: ， Calculate the maximum interior set Corresponding direction vector That is, the direction vector of the optimal fitted line is obtained as follows: ， The reference line is obtained based on the direction vector of the optimal fitted line.

4. The contour-based 3D adhesive coating anomaly detection method according to claim 3, characterized in that, The step of rotating and translating the projection dataset using the reference line to obtain the target point cloud includes: By solving for the tangent of the slope of the reference line, the angle between the reference line and the coordinate axis is obtained as follows: ， Based on the included angle, construct a two-dimensional rotation matrix for clockwise rotation as follows: ， Rotate the projected dataset in space around the coordinate axes by the same angle to obtain the rotated dataset: ， The translation coefficient is calculated using the following formula: ， The reference line is translated according to the translation coefficient until it coincides with the coordinate axis, resulting in the target point cloud: 。 5. A contour-based 3D adhesive coating anomaly detection device, characterized in that, A method for implementing a contour-based 3D adhesive coating anomaly detection as described in any one of claims 1-4 includes: The direction vector calculation module is used to project the three-dimensional point cloud data of the adhesive coating contour onto a plane, obtain the projection dataset, and calculate the unit direction vector passing through the two random points in the projection dataset. The reference line acquisition module is used to divide the maximum interior point set based on the standardized distance from all points in the projection dataset to the unit direction vector, and to obtain the reference line based on the direction vector fitted by the maximum interior point set. The projection dataset correction module is used to perform rotation and translation correction on the projection dataset using the reference line to obtain the target point cloud. The adhesive coating contour measurement module is used to measure the adhesive coating contour based on the target point cloud and obtain a measurement sequence, including: measuring the height and width of the adhesive coating contour according to the target point cloud, and calculating the area of ​​the adhesive coating contour using numerical integration; obtaining the corresponding measurement sequence based on the number of contours reconstructed by laser scanning during the adhesive coating process and the measurement data of the height, width, and area of ​​the adhesive coating contour. Assuming that the complete 3D adhesive coating trajectory includes M contours, three measurement sequences of size M are obtained respectively: ； The adhesive coating anomaly detection module is used to detect the adhesive coating contour according to the measurement sequence and obtain the adhesive coating anomaly detection result, including: The formula for calculating the difference between the two measurement sequences before and after correction is: ， Based on the law of large numbers and the central limit theorem, the difference sequence is assumed to follow a Gaussian distribution. According to the distribution characteristics of the Gaussian distribution, the data in the difference sequence within the interval [missing information] are calculated. The probability value within is given by the formula: ， Transform After sorting, we get: ， Solving for: ， The detection result of adhesive coating anomaly is obtained based on the magnitude of the probability value.

Images