Automatic film covering precision detection method and system for automobile lampshade

Abstract

The invention discloses an automatic film covering precision detection method and system for an automobile lampshade, and relates to the technical field of film covering detection, and the method comprises the following steps: obtaining a first real-time two-dimensional coordinate point and a second real-time two-dimensional coordinate point based on a point cloud image before real-time film covering and a point cloud image after real-time film covering; acquiring a first real-time midpoint and a second real-time midpoint based on the first real-time two-dimensional coordinate point and the second real-time two-dimensional coordinate point; acquiring a real-time film covering value based on the first real-time midpoint and the second real-time midpoint; obtaining a historical film covering value based on the normal three-dimensional point cloud atlas before and after film covering; obtaining a first film coating threshold value and a second film coating threshold value based on the historical film coating value; based on the real-time film covering value, the first film covering threshold value and the second film covering threshold value, a film covering precision detection result is obtained; the method is used for solving the problem that the accuracy of film precision detection through a two-dimensional image is poor due to the fact that a film and an automobile lampshade both have strong light reflection and light transmission characteristics in an existing film detection technology.

Description

Technical Field

[0001] This invention relates to the field of coating inspection technology, specifically to an automated method and system for inspecting the accuracy of coating on automotive lamp covers. Background Technology

[0002] To improve the weather resistance, scratch resistance, optical performance, and visual effect of automotive lamp covers, covering the lamp cover surface with a functional film or temporary protective film has become a core process. This process requires the film to achieve high-precision, defect-free, and conformal bonding with the complex three-dimensional curved surface of the lamp cover. Any bonding deviation will directly affect the protection of the product. Therefore, it is necessary to carry out automated film coating precision testing for automotive lamp covers.

[0003] Existing methods for coating accuracy inspection mainly rely on manual visual inspection or two-dimensional image comparison. Manual inspection is inefficient, labor-intensive, and susceptible to the influence of personnel experience, fatigue, and subjective judgment, resulting in poor consistency and making it difficult to meet the stringent efficiency and quality requirements of large-scale automated production. Two-dimensional image inspection is also problematic because both the film and the automotive headlight cover have strong reflective and translucent properties, which can easily lead to overexposure, specular reflection, or transmission interference in the image, affecting the coating accuracy inspection results. The accuracy of coating accuracy inspection using two-dimensional images is poor due to the strong reflective and translucent properties of both the film and the automotive headlight cover. Summary of the Invention

[0004] This invention aims to at least partially solve one of the technical problems in the prior art. It obtains three-dimensional point cloud images of the automotive headlight cover before and after film coating, labeled as real-time pre-coating point cloud image and real-time post-coating point cloud image, respectively. Based on these images, it obtains a first real-time two-dimensional coordinate point and a second real-time two-dimensional coordinate point; it also obtains a first real-time midpoint and a second real-time midpoint; it obtains a real-time coating value based on the first and second real-time two-dimensional coordinate points; it obtains historical coating values ​​based on normal three-dimensional point cloud images before and after coating; it obtains a first coating threshold and a second coating threshold based on the historical coating values; and it obtains coating accuracy detection results based on the real-time coating value, the first coating threshold, and the second coating threshold. This addresses the problem in existing coating detection technologies where the strong reflective and translucent properties of both the film and the automotive headlight cover lead to poor accuracy in coating accuracy detection using two-dimensional images.

[0005] To achieve the above objectives, this application provides an automated method for detecting the accuracy of coating on automotive lamp covers, comprising the following steps:

[0006] Obtain 3D point cloud images of the car headlight cover before and after film coating, and label them as real-time point cloud image before film coating and real-time point cloud image after film coating, respectively.

[0007] The first real-time two-dimensional coordinate point and the second real-time two-dimensional coordinate point are obtained based on the real-time point cloud map before and after film coating.

[0008] The first real-time midpoint and the second real-time midpoint are obtained based on the first real-time two-dimensional coordinate point and the second real-time two-dimensional coordinate point;

[0009] Real-time film covering values ​​are obtained based on the first real-time midpoint and the second real-time midpoint.

[0010] Historical coating values ​​are obtained based on the 3D point cloud images before and after normal coating.

[0011] The first and second coating thresholds are obtained based on historical coating values;

[0012] The coating accuracy detection results are obtained based on the real-time coating value, the first coating threshold, and the second coating threshold.

[0013] Furthermore, obtaining the first real-time two-dimensional coordinate point and the second real-time two-dimensional coordinate point based on the real-time point cloud map before and after film coating includes the following sub-steps:

[0014] Establish a Cartesian coordinate system on the plane where the car headlight cover is placed, and label it as the projected coordinate system;

[0015] Mark the three-dimensional coordinate point in the real-time point cloud image before film covering as the first real-time three-dimensional coordinate point;

[0016] Mark the three-dimensional coordinate points in the real-time point cloud image after film coating as the second real-time three-dimensional coordinate points;

[0017] The first real-time three-dimensional coordinates are projected onto the projected coordinate system to obtain two-dimensional coordinate points, which are then marked as the first real-time two-dimensional coordinate points.

[0018] The second real-time three-dimensional coordinates are projected onto the projected coordinate system to obtain two-dimensional coordinate points, which are then marked as the second real-time two-dimensional coordinate points.

[0019] Furthermore, obtaining the first real-time midpoint and the second real-time midpoint based on the first real-time two-dimensional coordinate point and the second real-time two-dimensional coordinate point includes the following sub-steps:

[0020] The minimum and maximum values ​​of the x-coordinates of the first real-time two-dimensional coordinate point are marked as the first x-coordinate and the second x-coordinate, respectively.

[0021] Get the average of the first and second horizontal values, and mark it as the average of the first horizontal value;

[0022] The minimum and maximum values ​​of the ordinate in the first real-time two-dimensional coordinate point are marked as the first ordinate value and the second ordinate value, respectively.

[0023] Obtain the average of the first and second vertical values, and label it as the average of the first vertical value;

[0024] The coordinate point with the first horizontal average value as the horizontal coordinate value and the first vertical average value as the vertical coordinate value is marked as the first real-time midpoint;

[0025] Treat the second real-time two-dimensional coordinate point as the first real-time two-dimensional coordinate point to obtain the first real-time midpoint, and mark it as the second real-time midpoint.

[0026] Furthermore, obtaining the real-time coating value based on the first real-time midpoint and the second real-time midpoint includes the following sub-steps:

[0027] The region formed by the first real-time two-dimensional coordinate points is marked as the first real-time region; the region formed by the second real-time two-dimensional coordinate points is marked as the second real-time region.

[0028] Move the second real-time region until its midpoint coincides with the first real-time midpoint. Then, rotate the second real-time region around its midpoint until the overlap between the first and second real-time regions is maximized. Obtain the three-dimensional height difference between each second real-time 2D coordinate point and the first real-time 2D coordinate point at the same coordinates, and mark it as the real-time coating value.

[0029] Furthermore, obtaining historical coating values ​​based on normal 3D point cloud images before and after coating includes the following sub-steps:

[0030] The normal 3D point cloud images before and after lamination are treated as real-time point cloud images before lamination and real-time point cloud images after lamination, respectively. Then, the real-time lamination value is obtained and marked as the historical lamination value.

[0031] Furthermore, obtaining the first and second coating thresholds based on historical coating values ​​includes the following sub-steps:

[0032] Get the first number of historical coating values. If there are identical historical coating values, count the number of each identical historical coating value and mark it as the number of identical coating values.

[0033] A Cartesian coordinate system was established with historical film covering values ​​as the horizontal axis data and the number of the same film covering value as the vertical axis data, and this system was marked as the film covering value observation coordinate system.

[0034] The historical coating value and the corresponding number of the same coating value are plotted on the coating value observation coordinate system as the x and y axes of the data points, respectively.

[0035] Mark the data points in the coating value observation coordinate system as coating value coordinate points.

[0036] Furthermore, obtaining the first and second coating thresholds based on historical coating values ​​also includes the following sub-steps:

[0037] Obtain the range length of historical coating values ​​in the coating value coordinate system, and mark it as D1;

[0038] Establish two straight lines parallel to the vertical axis of the coating value observation coordinate system, with a width of D2 between the two lines. Mark the space between the two lines as the real-time search space, which can be moved left and right.

[0039] The distribution threshold is calculated as: K = a × (D2 ÷ D1) × S1; where K is the distribution threshold, a is the set proportion, and S1 is the first quantity.

[0040] Furthermore, obtaining the first and second coating thresholds based on historical coating values ​​also includes the following sub-steps:

[0041] Mark the total number of historical coating values ​​corresponding to all coating value coordinate points in the real-time search space as the real-time search count;

[0042] Move the real-time search space until the minimum x-coordinate of the real-time search space is equal to the x-coordinate of the minimum coating value point. Then move the real-time search space to the right and continuously compare the number of real-time searches with the distribution quantity threshold. When the number of real-time searches is greater than or equal to the distribution quantity threshold, stop moving the real-time search space. Obtain the historical coating value corresponding to the minimum x-coordinate of the real-time search space at this time and mark it as the first coating threshold.

[0043] Move the real-time search space until the x-coordinate of the largest x-coordinate in the real-time search space is equal to the x-coordinate of the largest coating value. Then move the real-time search space to the left and continuously compare the number of real-time searches with the distribution quantity threshold. When the number of real-time searches is greater than or equal to the distribution quantity threshold, stop moving the real-time search space. Obtain the historical coating value corresponding to the largest x-coordinate in the real-time search space at this time and mark it as the second coating threshold.

[0044] Furthermore, obtaining the coating accuracy detection result based on the real-time coating value, the first coating threshold, and the second coating threshold includes the following steps:

[0045] If the real-time coating value is in the area between the first coating threshold and the second coating threshold, it is marked as a normal coating area.

[0046] The area where the real-time coating value is less than the first coating threshold is marked as the first abnormal area;

[0047] Areas with real-time coating values ​​greater than the second coating threshold are marked as the second abnormal area.

[0048] This application also provides an automated coating accuracy detection system for automotive lamp covers, including: a real-time data acquisition module, a coordinate point acquisition module, a midpoint acquisition module, a real-time coating value acquisition module, a historical coating value acquisition module, a threshold acquisition module, and a detection result acquisition module;

[0049] The real-time data acquisition module is used to acquire three-dimensional point cloud images of the car headlight cover before and after film coating, which are respectively labeled as real-time point cloud image before film coating and real-time point cloud image after film coating.

[0050] The coordinate point acquisition module is used to acquire a first real-time two-dimensional coordinate point and a second real-time two-dimensional coordinate point based on the real-time point cloud map before and after film covering.

[0051] The midpoint acquisition module is used to acquire the first real-time midpoint and the second real-time midpoint based on the first real-time two-dimensional coordinate point and the second real-time two-dimensional coordinate point;

[0052] The real-time film coating value acquisition module is used to acquire the real-time film coating value based on the first real-time midpoint and the second real-time midpoint.

[0053] The historical coating value acquisition module is used to acquire historical coating values ​​based on the three-dimensional point cloud maps before and after normal coating.

[0054] The threshold acquisition module is used to acquire a first coating threshold and a second coating threshold based on historical coating values;

[0055] The detection result acquisition module is used to obtain the coating accuracy detection result based on the real-time coating value, the first coating threshold, and the second coating threshold.

[0056] The beneficial effects of this invention are as follows: This invention acquires three-dimensional point cloud images of the automotive headlight cover before and after film coating, respectively labeled as real-time pre-coating point cloud image and real-time post-coating point cloud image; obtains a first real-time two-dimensional coordinate point and a second real-time two-dimensional coordinate point based on the real-time pre-coating point cloud image and real-time post-coating point cloud image; obtains a first real-time midpoint and a second real-time midpoint based on the first real-time two-dimensional coordinate point and the second real-time two-dimensional coordinate point; obtains a real-time coating value based on the first real-time midpoint and the second real-time midpoint; obtains historical coating values ​​based on the normal three-dimensional point cloud images before and after film coating; obtains a first coating threshold and a second coating threshold based on the historical coating values; and obtains the film coating accuracy detection result based on the real-time coating value, the first coating threshold, and the second coating threshold. The advantage is that the accuracy of film coating accuracy detection is improved through three-dimensional point cloud image analysis.

[0057] The present invention obtains real-time coating values ​​based on a first real-time midpoint and a second real-time midpoint. Its advantage is that it can determine whether the coating is properly adhered by using the real-time coating value, thereby improving the accuracy of coating precision detection. Attached Figure Description

[0058] Figure 1 This is a schematic diagram of the system of the present invention;

[0059] Figure 2 This is a schematic diagram of the first coating threshold and the second coating threshold of the present invention;

[0060] Figure 3 This is a flowchart of the steps of the method of the present invention. Detailed Implementation

[0061] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0062] Example 1, please refer to Figure 1 As shown, this application provides an automated coating accuracy detection system for automotive lamp covers, including: a real-time data acquisition module, a coordinate point acquisition module, a midpoint acquisition module, a real-time coating value acquisition module, a historical coating value acquisition module, a threshold acquisition module, and a detection result acquisition module;

[0063] The real-time data acquisition module is used to acquire three-dimensional point cloud images of the car headlight cover before and after film coating, which are labeled as real-time point cloud image before film coating and real-time point cloud image after film coating, respectively; the headlight cover is placed with the film coating side facing up, and the three-dimensional point cloud image is mainly acquired from the top view.

[0064] The coordinate point acquisition module is used to acquire the first real-time two-dimensional coordinate point and the second real-time two-dimensional coordinate point based on the real-time point cloud map before and after film covering.

[0065] The coordinate point acquisition module is configured with a coordinate point acquisition strategy, which includes:

[0066] Establish a Cartesian coordinate system on the plane where the car headlight cover is placed, and label it as the projected coordinate system;

[0067] Mark the three-dimensional coordinate point in the real-time point cloud image before film covering as the first real-time three-dimensional coordinate point;

[0068] Mark the three-dimensional coordinate points in the real-time point cloud image after film coating as the second real-time three-dimensional coordinate points;

[0069] The first real-time three-dimensional coordinates are projected onto the projected coordinate system to obtain two-dimensional coordinate points, which are then marked as the first real-time two-dimensional coordinate points.

[0070] The second real-time three-dimensional coordinates are projected onto the projection coordinate system to obtain two-dimensional coordinate points, which are then marked as the second real-time two-dimensional coordinate points. The three-dimensional coordinate points are then converted into two-dimensional coordinate points to facilitate image analysis.

[0071] The midpoint acquisition module is used to acquire the first real-time midpoint and the second real-time midpoint based on the first real-time two-dimensional coordinate point and the second real-time two-dimensional coordinate point;

[0072] The midpoint acquisition module is configured with a midpoint acquisition strategy, which includes:

[0073] The minimum and maximum values ​​of the x-coordinates of the first real-time two-dimensional coordinate point are marked as the first x-coordinate and the second x-coordinate, respectively.

[0074] Get the average of the first and second horizontal values, and mark it as the average of the first horizontal value;

[0075] The minimum and maximum values ​​of the ordinate in the first real-time two-dimensional coordinate point are marked as the first ordinate value and the second ordinate value, respectively.

[0076] Obtain the average of the first and second vertical values, and label it as the average of the first vertical value;

[0077] The coordinate point with the first horizontal mean as the x-coordinate value and the first vertical mean as the y-coordinate value is marked as the first real-time midpoint; the first real-time midpoint is regarded as the center point of all first real-time two-dimensional coordinate points;

[0078] The second real-time two-dimensional coordinate point is regarded as the first real-time two-dimensional coordinate point to obtain the first real-time midpoint, which is marked as the second real-time midpoint; the second real-time midpoint is regarded as the center point of all second real-time two-dimensional coordinate points; this facilitates the detection of the three-dimensional point cloud map before and after coating.

[0079] The real-time film coating value acquisition module is used to acquire real-time film coating values ​​based on the first real-time midpoint and the second real-time midpoint.

[0080] The real-time coating value acquisition module is configured with a real-time coating value acquisition strategy, which includes:

[0081] The region formed by the first real-time two-dimensional coordinate points is marked as the first real-time region; the region formed by the second real-time two-dimensional coordinate points is marked as the second real-time region.

[0082] The second real-time region is moved until its midpoint coincides with the midpoint of the first real-time region. Then, the second real-time region is rotated around the midpoint of the second real-time region until the overlap area between the first and second real-time regions is maximized. The three-dimensional height difference between the second and first real-time two-dimensional coordinate points with the same coordinates at this time is obtained and marked as the real-time coating value. The real-time coating value represents the height change of the film before and after coating, which is convenient for judging the change of the film. At the same time, since the position of the car headlight cover changes during the coating process, this method can be used to roughly adjust it to the same position for easy comparison.

[0083] The historical coating value acquisition module is used to acquire historical coating values ​​based on the normal 3D point cloud maps before and after coating.

[0084] The historical coating value acquisition module is configured with a historical coating value acquisition strategy, which includes:

[0085] The normal 3D point cloud images before and after lamination are treated as real-time point cloud images before lamination and real-time point cloud images after lamination, respectively. Then, the real-time lamination value is obtained and marked as the historical lamination value.

[0086] The threshold acquisition module is used to obtain the first coating threshold and the second coating threshold based on historical coating values;

[0087] The threshold acquisition module is configured with a threshold acquisition strategy, which includes:

[0088] Obtain a first number of historical coating values. If there are identical historical coating values, count the number of each identical historical coating value and mark it as the number of identical coating values. The first number of historical coating values ​​is to obtain the range of normal historical coating values, so the larger the first number is, the better. For example, the first number is 500.

[0089] A Cartesian coordinate system was established with historical film covering values ​​as the horizontal axis data and the number of the same film covering value as the vertical axis data, and this system was marked as the film covering value observation coordinate system.

[0090] The historical coating value and the corresponding number of the same coating value are plotted on the coating value observation coordinate system as the x and y axes of the data points, respectively.

[0091] The data points in the coating value observation coordinate system are marked as coating value coordinate points; these coordinate points are used to facilitate observation of the distribution range of historical coating values.

[0092] Obtain the range length of historical coating values ​​in the coating value coordinate system, labeled as D1; ​​see [link / reference]. Figure 2 As shown, D1 has a length from 0.4 to 1.8, which is 1.4.

[0093] Establish two straight lines parallel to the vertical axis of the coating value observation coordinate system. The width between the two lines is D2. Mark the space between the two lines as the real-time search space, which can be moved left and right. The real-time search space is convenient for observing the distribution range of historical coating values, so D2 should be smaller than D1, for example, D1 is 0.1.

[0094] The distribution number threshold is calculated as: K = a × (D2 ÷ D1) × S1; where K is the distribution number threshold, a is the set ratio, and S1 is the first quantity; the distribution number threshold is set relatively small in order to obtain the range of historical coating values ​​with a small number of distributions, because (D2 ÷ D1) × S1 represents the average number of historical coating values ​​in the real-time search space, so K must be less than (D2 ÷ D1) × S1, that is, a is less than 1, for example, a is 0.2;

[0095] In practical applications, the distribution threshold is calculated as: K = 0.2 × (0.1 ÷ 1.4) × 500 = 7.1, and the result is rounded to one decimal place.

[0096] Mark the total number of historical coating values ​​corresponding to all coating value coordinate points in the real-time search space as the real-time search count;

[0097] The real-time search space is moved until the x-coordinate of the minimum x-coordinate in the real-time search space is equal to the x-coordinate of the minimum coating value. Then, the real-time search space is moved to the right, and the number of real-time searches and the distribution quantity threshold are compared in real time. When the number of real-time searches is greater than or equal to the distribution quantity threshold, the movement of the real-time search space is stopped. The historical coating value corresponding to the minimum x-coordinate of the real-time search space at this time is obtained and marked as the first coating threshold. Abnormally small historical coating values ​​are deleted, thereby obtaining the accurate minimum value of the historical coating value.

[0098] The real-time search space is moved until the x-coordinate of the largest x-coordinate in the real-time search space is equal to the x-coordinate of the largest coating value. Then, the real-time search space is moved to the left, and the number of real-time searches and the distribution quantity threshold are compared in real time. When the number of real-time searches is greater than or equal to the distribution quantity threshold, the movement of the real-time search space is stopped. The historical coating value corresponding to the largest x-coordinate in the real-time search space at this time is obtained and marked as the second coating threshold. Abnormally large historical coating values ​​are deleted, thereby obtaining the accurate minimum value of the historical coating value.

[0099] For practical applications, please refer to Figure 2 As shown, the real-time search space is moved until the x-coordinate of the minimum x-coordinate of the real-time search space is equal to the x-coordinate of the minimum coating value point. Then, the real-time search space is moved to the right. The movement of the real-time search space stops when the number of real-time searches is greater than or equal to the distribution number threshold of 7.1. Please refer to [link to relevant documentation]. Figure 2 The real-time search space stopping position is shown, and the first coating threshold is 0.5. Similarly, please refer to [link to relevant documentation]. Figure 2 As shown, the second coating threshold obtained is 1.7.

[0100] The detection result acquisition module is used to obtain the film coating accuracy detection result based on the real-time film coating value, the first film coating threshold, and the second film coating threshold;

[0101] The detection result acquisition module is configured with a detection result acquisition strategy, which includes:

[0102] If the real-time coating value falls within the range between the first coating threshold and the second coating threshold, it is marked as a normal coating area, indicating that the coating thickness is normal.

[0103] Areas with real-time film coating values ​​less than the first film coating threshold are marked as first abnormal areas; this indicates that the film coating thickness is less than the normal film thickness, and can be initially located as uncoated areas; this may be due to the film coating not being aligned with the lampshade, or the film coating being damaged;

[0104] Areas with real-time coating values ​​greater than the second coating threshold are marked as second abnormal areas; this indicates that the coating thickness is greater than the normal film thickness, possibly because the coating failed to adhere tightly to the car headlight cover, resulting in air bubbles; different processing and alarms are applied based on different detection results.

[0105] In practical applications, for example, if a real-time coating value of 1.1 is obtained, and the real-time coating value of 1.1 is between the first coating threshold of 0.5 and the second coating value of 0.7, it means that the position corresponding to the real-time coating value is a normal coating area.

[0106] Example 2, please refer to Figure 3 As shown, this application provides an automated method for detecting the accuracy of coating on automotive lamp covers, comprising the following steps: Step S1: Obtain the three-dimensional point cloud images of the car headlight cover before and after film coating, and label them as the real-time point cloud image before film coating and the real-time point cloud image after film coating, respectively.

[0108] Step S2 involves obtaining the first and second real-time two-dimensional coordinate points based on the real-time point cloud map before and after lamination. Step S2 includes the following sub-steps:

[0109] Step S201: Establish a Cartesian coordinate system on the plane where the car headlight cover is placed, and mark it as the projected coordinate system;

[0110] Step S202: Mark the three-dimensional coordinate point in the real-time point cloud map before film covering as the first real-time three-dimensional coordinate point;

[0111] Step S203: Mark the three-dimensional coordinate points in the real-time coated point cloud map as the second real-time three-dimensional coordinate points;

[0112] Step S204: Project the first real-time three-dimensional coordinates onto the projection coordinate system to obtain two-dimensional coordinate points, and mark them as the first real-time two-dimensional coordinate points;

[0113] Step S205: Project the second real-time three-dimensional coordinates onto the projection coordinate system to obtain two-dimensional coordinate points, and mark them as the second real-time two-dimensional coordinate points.

[0114] Step S3: Obtain the first real-time midpoint and the second real-time midpoint based on the first real-time two-dimensional coordinate point and the second real-time two-dimensional coordinate point; Step S3 includes the following sub-steps:

[0115] Step S301: The minimum and maximum values ​​of the abscissa in the first real-time two-dimensional coordinate point are marked as the first abscissa and the second abscissa, respectively.

[0116] Step S302: Obtain the average of the first horizontal value and the second horizontal value, and mark it as the average of the first horizontal value;

[0117] Step S303: The minimum and maximum values ​​of the ordinate in the first real-time two-dimensional coordinate point are marked as the first ordinate value and the second ordinate value, respectively.

[0118] Step S304: Obtain the mean of the first vertical value and the second vertical value, and mark it as the mean of the first vertical value;

[0119] Step S305: Mark the coordinate point with the first horizontal average value as the horizontal coordinate value and the first vertical average value as the vertical coordinate value as the first real-time midpoint.

[0120] Step S306: Treat the second real-time two-dimensional coordinate point as the first real-time two-dimensional coordinate point to obtain the first real-time midpoint, and mark it as the second real-time midpoint.

[0121] Step S4: Obtain the real-time coating value based on the first real-time midpoint and the second real-time midpoint; Step S4 includes the following sub-steps:

[0122] Step S401: Mark the region composed of the first real-time two-dimensional coordinate points as the first real-time region; mark the region composed of the second real-time two-dimensional coordinate points as the second real-time region;

[0123] Step S402: Move the second real-time region so that the midpoint of the second real-time region coincides with the midpoint of the first real-time region. Then, rotate the second real-time region with the midpoint of the second real-time region as the rotation center. Stop rotating when the overlapping area between the first real-time region and the second real-time region is the largest. Obtain the three-dimensional height difference between the second real-time two-dimensional coordinate point and the first real-time two-dimensional coordinate point at each of the same coordinates at this time, and mark it as the real-time coating value.

[0124] Step S5: Obtain historical coating values ​​based on the normal 3D point cloud images before and after coating; Step S5 includes the following sub-steps:

[0125] Step S501: Treat the normal three-dimensional point cloud maps before and after lamination as real-time point cloud maps before lamination and real-time point cloud maps after lamination, respectively, and then obtain the real-time lamination value and mark it as the historical lamination value.

[0126] Step S6: Obtain the first coating threshold and the second coating threshold based on historical coating values; Step S6 includes the following sub-steps:

[0127] Step S601: Obtain the first number of historical coating values. If there are identical historical coating values, count the number of each identical historical coating value and mark it as the number of identical coating values.

[0128] Step S602: Establish a Cartesian coordinate system with historical film covering values ​​as the horizontal axis data and the number of the same film covering value as the vertical axis data, and mark it as the film covering value observation coordinate system;

[0129] Step S603: Plot the historical coating value and the corresponding number of the same coating value as the abscissa and ordinate of the data point on the coating value observation coordinate system, respectively.

[0130] Step S604: Mark the data points in the film coating value observation coordinate system as film coating value coordinate points;

[0131] Step S605: Obtain the range length of historical coating values ​​in the coating value coordinate system and mark it as D1;

[0132] Step S606: Establish two straight lines parallel to the vertical axis of the coating value observation coordinate system. The width between the two lines is D2. Mark the space between the two lines as the real-time search space, which can be moved left and right.

[0133] Step S607, calculate the distribution quantity threshold as: K=a×(D2÷D1)×S1; where K is the distribution quantity threshold, a is the set ratio, and S1 is the first quantity;

[0134] Step S608: Mark the total number of historical coating values ​​corresponding to all coating value coordinate points in the real-time search space as the real-time search count;

[0135] Step S609: Move the real-time search space so that the minimum x-coordinate of the real-time search space is equal to the x-coordinate of the minimum coating value coordinate point. Then move the real-time search space to the right and continuously judge the size of the real-time search count and the distribution quantity threshold. When the real-time search count is greater than or equal to the distribution quantity threshold, stop moving the real-time search space. Obtain the historical coating value corresponding to the minimum x-coordinate of the real-time search space at this time and mark it as the first coating threshold.

[0136] Step S610: Move the real-time search space so that the maximum x-coordinate of the real-time search space is equal to the x-coordinate of the maximum coating value coordinate point. Then move the real-time search space to the left and continuously judge the size of the real-time search count and the distribution quantity threshold. When the real-time search count is greater than or equal to the distribution quantity threshold, stop moving the real-time search space. Obtain the historical coating value corresponding to the maximum x-coordinate of the real-time search space at this time and mark it as the second coating threshold.

[0137] Step S7: Obtain the coating accuracy detection result based on the real-time coating value, the first coating threshold, and the second coating threshold; Step S7 includes the following sub-steps:

[0138] Step S701: If the real-time coating value is in the area between the first coating threshold and the second coating threshold, it is marked as a normal coating area.

[0139] Step S702: Mark the area where the real-time coating value is less than the first coating threshold as the first abnormal area;

[0140] Step S703: The area where the real-time coating value is greater than the second coating threshold is marked as the second abnormal area.

[0141] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0142] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings or direct couplings or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

Claims

1. A method for automatically detecting the accuracy of coating on automotive lamp covers, characterized in that, Includes the following steps: Obtain 3D point cloud images of the car headlight cover before and after film coating, and label them as real-time point cloud image before film coating and real-time point cloud image after film coating, respectively. The first real-time two-dimensional coordinate point and the second real-time two-dimensional coordinate point are obtained based on the real-time point cloud map before and after film coating. The first real-time midpoint and the second real-time midpoint are obtained based on the first real-time two-dimensional coordinate point and the second real-time two-dimensional coordinate point; Real-time film covering values ​​are obtained based on the first real-time midpoint and the second real-time midpoint. Historical coating values ​​are obtained based on the 3D point cloud images before and after normal coating. The first and second coating thresholds are obtained based on historical coating values; The coating accuracy detection results are obtained based on the real-time coating value, the first coating threshold, and the second coating threshold.

2. The method for automatically detecting the coating accuracy of automotive lamp covers according to claim 1, characterized in that, Obtaining the first and second real-time two-dimensional coordinate points based on the real-time point cloud map before and after lamination includes the following sub-steps: Establish a Cartesian coordinate system on the plane where the car headlight cover is placed, and label it as the projected coordinate system; Mark the three-dimensional coordinate point in the real-time point cloud image before film covering as the first real-time three-dimensional coordinate point; Mark the three-dimensional coordinate points in the real-time point cloud image after film coating as the second real-time three-dimensional coordinate points; The first real-time three-dimensional coordinates are projected onto the projected coordinate system to obtain two-dimensional coordinate points, which are then marked as the first real-time two-dimensional coordinate points. The second real-time three-dimensional coordinates are projected onto the projected coordinate system to obtain two-dimensional coordinate points, which are then marked as the second real-time two-dimensional coordinate points.

3. The method for automatically detecting the coating accuracy of automotive lamp covers according to claim 2, characterized in that, Obtaining the first real-time midpoint and the second real-time midpoint based on the first real-time two-dimensional coordinate point includes the following sub-steps: The minimum and maximum values ​​of the x-coordinates of the first real-time two-dimensional coordinate point are marked as the first x-coordinate and the second x-coordinate, respectively. Get the average of the first and second horizontal values, and mark it as the average of the first horizontal value; The minimum and maximum values ​​of the ordinate in the first real-time two-dimensional coordinate point are marked as the first ordinate value and the second ordinate value, respectively. Obtain the average of the first and second vertical values, and label it as the average of the first vertical value; The coordinate point with the first horizontal average value as the horizontal coordinate value and the first vertical average value as the vertical coordinate value is marked as the first real-time midpoint; Treat the second real-time two-dimensional coordinate point as the first real-time two-dimensional coordinate point to obtain the first real-time midpoint, and mark it as the second real-time midpoint.

4. The automated coating accuracy detection method for automotive lamp covers according to claim 3, characterized in that, Obtaining the real-time coating value based on the first real-time midpoint and the second real-time midpoint includes the following sub-steps: The region formed by the first real-time two-dimensional coordinate points is marked as the first real-time region; the region formed by the second real-time two-dimensional coordinate points is marked as the second real-time region. Move the second real-time region until its midpoint coincides with the first real-time midpoint. Then, rotate the second real-time region around its midpoint until the overlap between the first and second real-time regions is maximized. Obtain the three-dimensional height difference between each second real-time 2D coordinate point and the first real-time 2D coordinate point at the same coordinates, and mark it as the real-time coating value.

5. The automated coating accuracy detection method for automotive lamp covers according to claim 4, characterized in that, Obtaining historical coating values ​​based on normal 3D point cloud images before and after coating includes the following sub-steps: The normal 3D point cloud images before and after lamination are treated as real-time point cloud images before lamination and real-time point cloud images after lamination, respectively. Then, the real-time lamination value is obtained and marked as the historical lamination value.

6. The method for automatically detecting the coating accuracy of automotive lamp covers according to claim 5, characterized in that, Obtaining the first and second coating thresholds based on historical coating values ​​includes the following sub-steps: Get the first number of historical coating values. If there are identical historical coating values, count the number of each identical historical coating value and mark it as the number of identical coating values. A Cartesian coordinate system was established with historical film covering values ​​as the horizontal axis data and the number of the same film covering value as the vertical axis data, and this system was marked as the film covering value observation coordinate system. The historical coating value and the corresponding number of the same coating value are plotted on the coating value observation coordinate system as the x and y axes of the data points, respectively. Mark the data points in the coating value observation coordinate system as coating value coordinate points.

7. The automated coating accuracy detection method for automotive lamp covers according to claim 6, characterized in that, Obtaining the first and second coating thresholds based on historical coating values ​​also includes the following sub-steps: Obtain the range length of historical coating values ​​in the coating value coordinate system, and mark it as D1; Establish two straight lines parallel to the vertical axis of the coating value observation coordinate system, with a width of D2 between the two lines. Mark the space between the two lines as the real-time search space, which can be moved left and right. The distribution threshold is calculated as: K = a × (D2 ÷ D1) × S1; where K is the distribution threshold, a is the set proportion, and S1 is the first quantity.

8. The method for automatically detecting the coating accuracy of automotive lamp covers according to claim 7, characterized in that, Obtaining the first and second coating thresholds based on historical coating values ​​also includes the following sub-steps: Mark the total number of historical coating values ​​corresponding to all coating value coordinate points in the real-time search space as the real-time search count; Move the real-time search space until the minimum x-coordinate of the real-time search space is equal to the x-coordinate of the minimum coating value point. Then move the real-time search space to the right and continuously compare the number of real-time searches with the distribution number threshold. Stop moving the real-time search space when the number of real-time searches is greater than or equal to the distribution number threshold. Obtain the historical coating value corresponding to the minimum horizontal coordinate in the real-time search space at this time, and mark it as the first coating threshold; Move the real-time search space until the x-coordinate of the largest x-coordinate in the real-time search space is equal to the x-coordinate of the largest film value point. Then move the real-time search space to the left and continuously compare the number of real-time searches with the distribution number threshold. Stop moving the real-time search space when the number of real-time searches is greater than or equal to the distribution number threshold. Obtain the historical coating value corresponding to the largest horizontal coordinate in the real-time search space at this time, and mark it as the second coating threshold.

9. The method for automatically detecting the coating accuracy of automotive lamp covers according to claim 8, characterized in that, Obtaining the coating accuracy detection result based on the real-time coating value, the first coating threshold, and the second coating threshold includes the following steps: If the real-time coating value is in the area between the first coating threshold and the second coating threshold, it is marked as a normal coating area. The area where the real-time coating value is less than the first coating threshold is marked as the first abnormal area; Areas with real-time coating values ​​greater than the second coating threshold are marked as the second abnormal area.

10. An automated coating accuracy testing system for automotive lamp covers, used to implement the automated coating accuracy testing method for automotive lamp covers as described in any one of claims 1-9, characterized in that, It includes a real-time data acquisition module, a coordinate point acquisition module, a midpoint acquisition module, a real-time coating value acquisition module, a historical coating value acquisition module, a threshold acquisition module, and a detection result acquisition module; The real-time data acquisition module is used to acquire three-dimensional point cloud images of the car headlight cover before and after film coating, which are respectively labeled as real-time point cloud image before film coating and real-time point cloud image after film coating. The coordinate point acquisition module is used to acquire a first real-time two-dimensional coordinate point and a second real-time two-dimensional coordinate point based on the real-time point cloud map before and after film covering. The midpoint acquisition module is used to acquire the first real-time midpoint and the second real-time midpoint based on the first real-time two-dimensional coordinate point and the second real-time two-dimensional coordinate point; The real-time film coating value acquisition module is used to acquire the real-time film coating value based on the first real-time midpoint and the second real-time midpoint. The historical coating value acquisition module is used to acquire historical coating values ​​based on the three-dimensional point cloud maps before and after normal coating. The threshold acquisition module is used to acquire a first coating threshold and a second coating threshold based on historical coating values; The detection result acquisition module is used to obtain the coating accuracy detection result based on the real-time coating value, the first coating threshold, and the second coating threshold.

Images