Method and system for measuring light guide teeth of automotive lamp silica gel lens sub-assembly

Abstract

The invention discloses a method and a system for measuring light guide teeth of a car lamp silica gel lens sub-assembly. The method comprises the following steps: S1, acquiring three-dimensional data of the car lamp silica gel lens sub-assembly through a 3D laser scanner; s2, carrying out ROI region segmentation on the three-dimensional data of the vehicle lamp silica gel lens sub-assembly to obtain ROI region point cloud data including light guide teeth and a reference surface; s3, performing reference surface fitting on the obtained ROI point cloud data to obtain a mathematical model of a reference surface; s4, correcting the positions of the light guide teeth; and S5, calculating the relative height from the light guide teeth to the reference surface to obtain the extension height of the light guide teeth. The invention provides a method and a system for measuring light guide teeth of a car lamp silica gel lens sub-assembly, which can efficiently and accurately detect the extension height of the light guide teeth of a silica gel lens and ensure the optical performance and quality of a car lamp lens.

Description

TECHNICAL FIELD

[0001] The application relates to a light guide tooth measurement method and system for a silicone lens sub-assembly of a vehicle lamp, and belongs to the technical field of vehicle lamp detection. BACKGROUND

[0002] At present, silicone lenses gradually replace traditional glass and hard plastic lenses in the field of vehicle lamps (especially LED and ADB intelligent vehicle lamps) due to their high light transmittance, flexibility, weather resistance and complex optical structure forming capability, and become the core components of the optical design of a new generation of vehicle lamps. The light guide tooth of a silicone lens is a microstructure designed on the surface of an optical lens, which is usually used to control the direction of light propagation, improve light efficiency uniformity or achieve a specific light distribution effect. The high light guide tooth determines the light deflection angle, and too high light guide tooth leads to light leakage, and too low light guide tooth leads to poor uniformity.

[0003] After the silicone lens is assembled into a sub-assembly, the light guide tooth thereof is usually positioned and fixed by a metal sheet with holes, and the silicone light guide tooth extends out of the hole of the metal sheet. The design requirement for the extension amount of the light guide tooth is 0.2 mm, but due to the small size and high precision requirement (usually the tolerance needs to be controlled within ±0.05 mm), the traditional manual measurement method cannot guarantee the accuracy and repeatability of the measurement results. The traditional measurement methods (such as contact probes or two-dimensional image measuring instruments) have the problems of low efficiency, easy damage to soft materials or inability to obtain three-dimensional topography.

[0004] To solve this measurement problem, the existing technology often uses blue light or structured light three-dimensional scanning technology. However, since the silicone material itself has light transmittance and light guiding characteristics, direct scanning will cause optical signal interference, affecting the measurement accuracy. Therefore, the conventional method needs to spray a developing agent (such as a matte spray) on the surface of the sample before measurement to reduce the light transmittance of the material. However, this method has the following significant defects:

[0005] 1. Long pretreatment time: the process of spraying the developing agent and waiting for drying usually takes 15-30 minutes, which seriously affects the detection efficiency;

[0006] 2. Sample contamination risk: the residual developing agent may penetrate into the silicone microstructure, causing a decrease in optical performance;

[0007] 3. Secondary damage: the subsequent cleaning process easily causes scratches on the surface of the light guide tooth, eventually leading to the scrap of high-value samples.

[0008] Although a line-scan 3D laser scanner can quickly obtain three-dimensional point cloud data of the surface of an object through the principle of laser triangulation, the light transmittance and surface scattering characteristics of the light guide tooth of a silicone lens will cause laser signal attenuation or noise interference, and the existing technology cannot achieve high-precision measurement. These limitations make it difficult for the existing measurement method to meet the needs of modern production lines for high efficiency, non-contact and non-damaging detection, and it is urgent to develop a new measurement scheme.

[0009] Therefore, it is urgent to design a detection scheme to quickly and accurately detect the protruding height of the light guide tooth of the silica gel lens. SUMMARY

[0010] The technical problem to be solved by the present application is to overcome the shortcomings of the prior art and provide a light guide tooth measurement method and system for a vehicle lamp silica gel lens sub-assembly, which can efficiently and accurately detect the protruding height of the light guide tooth of the silica gel lens and ensure the optical performance and quality of the vehicle lamp lens.

[0011] To solve the above technical problems, the technical scheme of the present application is:

[0012] In one aspect, the present application provides a light guide tooth measurement method for a vehicle lamp silica gel lens sub-assembly, which comprises the following steps:

[0013] Step S1: acquiring three-dimensional data of the vehicle lamp silica gel lens sub-assembly by a 3D laser scanner;

[0014] Step S2: performing ROI region segmentation on the three-dimensional data of the vehicle lamp silica gel lens sub-assembly to obtain ROI region point cloud data containing the light guide tooth and the reference surface;

[0015] Step S3: fitting the reference surface to the obtained ROI region point cloud data to obtain a mathematical model of the reference surface;

[0016] Step S4: correcting the position of the light guide tooth;

[0017] Step S5: calculating the relative height of the light guide tooth to the reference surface to obtain the protruding height of the light guide tooth.

[0018] Further, the conditions for ROI region segmentation in step S2 are as follows:

[0019] X L ≤X≤X R ;

[0020] Y L ≤Y≤Y R ;

[0021] Z L ≤Z≤Z R ;

[0022] Wherein, X, Y, Z are the coordinates of the point cloud obtained by ROI region segmentation in three-dimensional space, X L , Y L , Z L are the minimum values of the three-dimensional coordinates of the ROI region, X R , Y R , Z R are the maximum values of the three-dimensional coordinates of the ROI region.

[0023] Further, the step S3 specifically comprises the following steps:

[0024] Step S31, randomly selecting three points which are not coincident and not on the same straight line from the ROI region point cloud data;

[0025] Step S32, performing three-dimensional plane fitting through the three selected points to obtain a fitting plane;

[0026] Step S33, calculating the distance L of the remaining points in the ROI region point cloud data from the fitting plane;

[0027] Step S34, setting a distance threshold H, if L < H, the point is determined as a plane point;

[0028] Step S35, if the number of plane points > 95% of the total number of points in the ROI region point cloud data, the fitting plane is determined as the reference surface; if the number of plane points ≤ 95% of the total number of points in the ROI region point cloud data, steps S31-S34 are repeatedly executed.

[0029] Further, the step S4 specifically comprises the following steps:

[0030] Step S41, obtaining the cube region of the vehicle lamp silica gel lens subassembly by using a fixed region segmentation method;

[0031] Step S42, extracting the contour points of the sample edge;

[0032] Step S43, performing straight line fitting on the extracted contour points by using a RANSAC algorithm to obtain an optimal straight line;

[0033] Step S44, extracting the midpoint of the optimal straight line;

[0034] Step S45, calculating a transformation matrix and a rotation angle, and obtaining a new position of the cube region where the light guide tooth is located after offset in the XY plane according to the transformation matrix and the rotation angle;

[0035] Step S46, obtaining the new position of the cube region where the light guide tooth is located after offset in the XY plane according to the calculated transformation matrix and the rotation angle, so as to correct the position of the light guide tooth.

[0036] Further, the segmentation condition of the cube region of the vehicle lamp silica gel lens subassembly in the step S41 is as follows:

[0037] X' L ≤X'≤X' R ;

[0038] Y' L ≤Y'≤Y' R ;

[0039] Z' L ≤Z'≤Z' R ;

[0040] Where X', Y', and Z' are the coordinates of the point cloud in three-dimensional space within the cubic region of the automotive headlight silicone lens sub-assembly, X' L Y' L Z' L Let X' be the minimum three-dimensional coordinate of the cubic region of the headlight silicone lens sub-assembly. R Y' R Z' R The maximum three-dimensional coordinates of the cubic region of the silicone lens sub-assembly for the vehicle headlight are given.

[0041] Furthermore, step S42 specifically includes the following steps:

[0042] Step S421, normal vector estimation, specifically includes the following steps:

[0043] For each point p in the point cloud within the cubic region of the automotive headlight silicone lens sub-assembly i Use a KD tree to find the relationship between point p and point p. i The nearest K neighborhood points;

[0044] PCA is used to fit the best-fit plane of the neighborhood points. The normal vector of the best-fit plane of the neighborhood points is solved, and the covariance matrix C is calculated. The formula for calculating the covariance matrix C is as follows:

[0045]

[0046] Perform eigenvalue decomposition on the covariance matrix C, and the normal vector n i The eigenvector corresponding to the smallest eigenvalue obtained from the decomposition;

[0047] Orient all normal vectors toward the viewpoint;

[0048] Step S422, Boundary point detection, specifically includes the following steps:

[0049] Find the points where the normal vector differs significantly from the normal vectors of their neighbors. Calculate the difference in normal vectors for each point p. i Calculate each point p i normal vector n i With neighborhood normal vector n j The average included angle θ i The average included angle θ i The calculation formula is as follows:

[0050]

[0051] If θ i is greater than a set threshold value, the point is determined as a contour point.

[0052] Further, the step S43 specifically comprises the following steps:

[0053] Step S431, randomly select two non-coincident contour points to obtain a fitting straight line;

[0054] Step S432, respectively, count the distance L' of each remaining contour point to the fitting straight line, and set a distance threshold H';

[0055] Step S433, if L' < H', it is determined that the contour point is on the fitting straight line, and the contour point is the straight line point;

[0056] Step S434, if the number of straight line points > 95% of the total number of contour points, the fitting straight line is determined as the optimal straight line,

[0057] Step S435, if the number of straight line points ≤ 95% of the total number of contour points, repeat steps S431-S434.

[0058] Further, the step S44 specifically comprises the following steps:

[0059] The maximum and minimum values of the contour points of the optimal straight line are calculated to obtain the contour point (x min ,y min ,z min ) corresponding to the maximum value and the contour point (x max ,y max ,z max ) corresponding to the minimum value, and the midpoint of the optimal straight line is calculated as (x center ,y center ,z center ), and the calculation formula of the midpoint of the optimal straight line is as follows,

[0060] x center =(x max -x min )÷2;

[0061] y center =(y max -y min )÷2;

[0062] z center =(z max -z min )÷2.

[0063] Further, the step S5 specifically comprises the following steps:

[0064] The light guide tooth region point cloud is segmented based on the light guide tooth located in the cubic region;

[0065] The Z value distribution statistics are performed, the value with the most Z value distribution is taken as a reference value, the points within 0.01 mm from the reference value are taken as a set of to-be-calculated points, and then average point calculation is performed on the set of to-be-calculated points;

[0066] Finally, the height of the average point to the reference surface is calculated to obtain the protruding height of the light guide tooth.

[0067] Another aspect of the present application provides a system applying the light guide tooth measurement method of the vehicle lamp silica gel lens subassembly, which comprises:

[0068] A mechanical arm is used to grab the vehicle lamp silica gel lens subassembly and place the vehicle lamp silica gel lens subassembly on a sliding table;

[0069] A sliding table is used to drive the vehicle lamp silica gel lens subassembly to move;

[0070] A 3D laser scanner is used to perform single line scanning on the vehicle lamp silica gel lens subassembly moved to a detection position;

[0071] A host computer is installed with PLC software and vision software, the host computer controls the actions of the mechanical arm and the sliding table through the PLC software, and communicates with the vision software when the vehicle lamp silica gel lens subassembly reaches the detection position; the host computer obtains the scanning result of the 3D laser scanner through the vision software, and then calculates the protruding height of the light guide tooth according to the scanning result; finally, the host computer displays the protruding height result of the light guide tooth on an interface, and records the log in a database.

[0072] By using the above technical scheme, the present application realizes the automatic detection of the light guide tooth of the silica gel lens by combining software and hardware, and realizes the production line detection result recording and statistics by combining the database function. The present application realizes the automation and intelligentization of the height detection of the light guide tooth of the vehicle lamp silica gel lens subassembly by using the 3D laser scanner combined with the sliding table to scan the vehicle lamp silica gel lens subassembly, and by performing algorithm processing on the point cloud data of the vehicle lamp silica gel lens subassembly, which significantly improves the detection efficiency and precision, and provides a powerful guarantee for the quality control of the vehicle lamp silica gel lens subassembly. BRIEF DESCRIPTION OF DRAWINGS

[0073] Figure 1 The flowchart of the light guide tooth measurement method of the vehicle lamp silica gel lens subassembly of the present application is shown;

[0074] Figure 2 The flowchart of step S3 of the present application is shown;

[0075] Figure 3 The flowchart of step S4 of the present application is shown;

[0076] Figure 4 The use schematic diagram of the system for measuring the light guide tooth of the vehicle lamp silica gel lens subassembly is shown. DETAILED DESCRIPTION

[0077] In order to make the content of the present application more easily understood, the present application will be further described in detail below according to specific embodiments and in conjunction with the accompanying drawings.

[0078] Embodiment One

[0079] As shown in the figure, the embodiment provides a method for measuring the light guide tooth of a vehicle lamp silica gel lens subassembly, which comprises the following steps: Figure 1

[0080] Step S1, three-dimensional data of the vehicle lamp silica gel lens subassembly is collected by a 3D laser scanner.

[0081] Step S2, ROI region segmentation is performed on the three-dimensional data of the vehicle lamp silica gel lens subassembly to obtain ROI region point cloud data containing the light guide tooth and the reference surface. Specifically:

[0082] Because the vehicle lamp silica gel lens subassembly contains multiple components, and the embodiment only needs to measure the protruding height of the light guide tooth, only the cubic region data containing the light guide tooth and the reference surface is further processed. The three-dimensional data of the vehicle lamp silica gel lens subassembly is subjected to ROI region segmentation to obtain points meeting the conditions, and the conditions for ROI region segmentation are as follows:

[0083] X L ≤X≤X R ;

[0084] Y L ≤Y≤Y R ;

[0085] Z L ≤Z≤Z R ;

[0086] Wherein, X, Y, Z are the coordinates of the point cloud obtained by ROI region segmentation in three-dimensional space, X L , Y L , Z L are the minimum values of the three-dimensional coordinates of the ROI region, X R , Y R , Z R are the maximum values of the three-dimensional coordinates of the ROI region.

[0087] Step S3, the reference surface fitting is performed on the obtained ROI region point cloud data to obtain the mathematical model of the reference surface; as shown in the figure, specifically: Figure 2 ​​

[0088] Step S31, randomly select three points that do not coincide and are not on the same straight line from the ROI region point cloud data;

[0089] Step S32, perform three-dimensional plane fitting through the three selected points to obtain a fitting plane;

[0090] Step S33, calculate the distance L of the remaining points in the ROI region point cloud data from the fitting plane;

[0091] Step S34, set a distance threshold H, if L < H, then determine the point as a plane point;

[0092] Step S35, if the number of plane points > 95% of the total number of points in the ROI region point cloud data, determine the fitting plane as the reference surface; if the number of plane points ≤ 95% of the total number of points in the ROI region point cloud data, repeat steps S31-S34 until the reference surface is found.

[0093] Step S4, correct the position of the light guide tooth. The light guide tooth region position correction is to deal with the case that the installation position of the vehicle lamp silica gel lens subassembly on the sliding table is offset. The position of the light guide tooth is fixed on the vehicle lamp silica gel lens subassembly, that is, the relative position relationship between the light guide tooth and the vehicle lamp silica gel lens subassembly does not change. However, due to the installation position of the vehicle lamp silica gel lens subassembly on the sliding table may be slightly offset during testing, it is necessary to correct the position of the light guide tooth according to the actual position of the vehicle lamp silica gel lens subassembly, and then obtain the actual position of the light guide tooth region, so as to realize accurate segmentation of the light guide tooth region point cloud data and ensure the accuracy of subsequent light guide tooth height measurement. The linear feature on one side of the vehicle lamp silica gel lens subassembly of the embodiment is relatively obvious, so the correction method adopts a line segment-based position correction method. As shown in the following figure, the specific steps are as follows: Figure 3

[0094] Step S41, obtain the cuboid region of the vehicle lamp silica gel lens subassembly by using the fixed region segmentation method. The edge of the vehicle lamp silica gel lens subassembly of the embodiment has obvious linear features and can be approximated as a cuboid. The size of the cuboid region is required to be large enough to ensure that the edge of the vehicle lamp silica gel lens subassembly is still in this region even after rotation and translation. The segmentation conditions of the cuboid region of the vehicle lamp silica gel lens subassembly are as follows:

[0095] X' L ≤X'≤X' R ;

[0096] Y' L ≤Y'≤Y' R ;

[0097] Z' L ≤Z'≤Z' R ;​

[0098] Where X', Y', and Z' are the coordinates of the point cloud in three-dimensional space within the cubic region of the automotive headlight silicone lens sub-assembly, X' L Y' L Z' L Let X' be the minimum three-dimensional coordinate of the cubic region of the automotive headlight silicone lens subassembly. R Y' R Z' R The maximum three-dimensional coordinates of the cubic region of the automotive headlight silicone lens sub-assembly.

[0099] Step S42: Extract contour points from the sample edge. A contour extraction method based on normal vectors is used because the direction of the point cloud normal vectors at the object contour changes drastically. By analyzing the abrupt changes in the local surface normal vectors, boundary points can be efficiently identified. Specifically:

[0100] Step S421, Normal vector estimation involves calculating the normal vector for each point:

[0101] 1. Neighborhood Search: For each point p in the point cloud within the cubic region of the headlight silicone lens sub-assembly. i Use a KD tree to find the relationship between point p and point p. i The K nearest neighbors (usually K = 20 to 50).

[0102] 2. Local Plane Fitting: PCA (Principal Component Analysis) is used to fit the best-fit plane of the neighborhood points. The normal vector of the best-fit plane is then calculated, and the covariance matrix C is determined. The formula for calculating the covariance matrix C is as follows:

[0103]

[0104] Perform eigenvalue decomposition on the covariance matrix C, and the normal vector n i The eigenvector corresponding to the smallest eigenvalue obtained from the decomposition.

[0105] 3. Normal vector orientation: Orient all normal vectors toward the viewpoint (i.e., the sensor origin, such as a 3D laser scanner) to avoid directional ambiguity.

[0106] Step S422, Boundary Point Detection:

[0107] Find the points where the normal vector differs significantly from the normal vectors of their neighbors. Calculate the difference in normal vectors for each point p. i Calculate each point p i normal vector n i With neighborhood normal vector n j The average included angle θ i average included angle θ i The calculation formula is as follows:

[0108]

[0109] If θ i is greater than a set threshold, the point is determined as a contour point.

[0110] Step S43, using RANSAC algorithm to fit a straight line to the extracted contour points, and get the optimal straight line. Specifically:

[0111] Step S431, randomly select two non-coincident contour points to fit a straight line;

[0112] Step S432, respectively, count the distance L' of each contour point to the fitted straight line, and set a distance threshold H';

[0113] Step S433, if L' < H', it is determined that the contour point is on the fitted straight line, and the contour point is the straight line point;

[0114] Step S434, if the number of straight line points is greater than 95% of the total number of contour points, the fitted straight line is determined as the optimal straight line,

[0115] Step S435, if the number of straight line points is less than or equal to 95% of the total number of contour points, repeat steps S431-S434 until the optimal straight line is found.

[0116] Step S44, extract the midpoint of the optimal straight line. Specifically:

[0117] Calculate the maximum and minimum values of the contour points of the optimal straight line, get the contour point (x min ,y min ,z min ) corresponding to the maximum value and the contour point (x max ,y max ,z max ) corresponding to the minimum value, and the midpoint of the optimal straight line is (x center ,y center ,z center ), then the calculation formula of the midpoint of the optimal straight line is as follows,

[0118] x center =(x max -x min )÷2;

[0119] y center =(y max -y min )÷2;

[0120] z center =(z max -z min )÷2.

[0121] Step S45, calculate the transformation matrix and the rotation angle, and obtain the new position of the light guide tooth region in the XY plane after the light guide tooth region is offset according to the transformation matrix and the rotation angle. Specifically, the transformation matrix T is calculated according to the following formula:

[0122] Since the vehicle lamp silica gel lens subassembly is placed on the sliding table, there is generally no position offset in the Z-axis direction (i.e., the direction perpendicular to the sliding table), so only the position correction is performed on the XY plane (i.e., the horizontal plane).

[0123] Given the state of a line segment in a two-dimensional space before and after transformation;

[0124] Initial state of the line segment:

[0125] The straight line equation is L0: a0x + b0y + c0 = 0;

[0126] The midpoint coordinates are M0 = (x m0 ,y m0 ) ∈ L0.

[0127] State of the line segment after transformation:

[0128] The straight line equation is L1: a1x + b1y + c1 = 0;

[0129] The midpoint coordinates are M1 = (x m1 ,y m1 ) ∈ L1.

[0130] The expression of the transformation matrix T is as follows:

[0131]

[0132]

[0133] M1 = R · M0 + t;

[0134] Where T is the transformation matrix, R is the rotation matrix, t is the translation matrix, the transformation operation includes rotation operation and translation; θ is the rotation angle, t x is the translation distance in the X direction, and t y is the translation distance in the Y direction;

[0135] Step S46, according to the calculated transformation matrix and rotation angle, the new position of the light guide tooth region in the XY plane after the light guide tooth region is offset can be obtained, so that the light guide tooth position is corrected.

[0136] Step S5, calculate the relative height of the light guide tooth to the reference surface, and obtain the protruding height of the light guide tooth, which is the relative height of the light guide tooth to the reference surface. Specifically,

[0137] Based on the light guide tooth region point cloud obtained by segmenting the light guide tooth region in the cubic region;

[0138] Since the point cloud has noise, the height accuracy will be affected by directly calculating the height, so the Z value distribution is first calculated, the value with the most Z value distribution is taken as the reference value, the points within 0.01 mm of the reference value are taken as the set of points to be calculated, and then the average point calculation is performed on the set of points to be calculated;

[0139] Finally, the height of the average point to the reference surface is calculated, and the protruding height of the light guide tooth is obtained.

[0140] Thus, it can be determined whether the protruding height of the light guide tooth is too large or too small, preventing the protruding height of the light guide tooth from affecting the assembly of the vehicle lamp silicone lens subassembly and the module and the lighting effect of the vehicle lamp.

[0141] Example Two

[0142] As shown in Figure 4 , the present embodiment provides a system applying the light guide tooth measurement method of the vehicle lamp silicone lens subassembly of example one, which comprises:

[0143] A mechanical arm is used to grab the vehicle lamp silicone lens subassembly 3 and place it on the sliding table 1.

[0144] A sliding table 1 is used to move the vehicle lamp silicone lens subassembly 3.

[0145] A 3D laser scanner 2 is used to perform a single line scan on the vehicle lamp silicone lens subassembly 3 moved to the detection position. The 3D laser scanner 2 needs to be installed at an angle of 15° to avoid the scanning line emitted by the scanner and the light guide tooth being in a straight line, thereby avoiding the scanning line penetrating the transparent light guide tooth and failing to obtain data.

[0146] A host computer is installed with PLC software and vision software. The host computer controls the actions of the mechanical arm and the sliding table 1 through the PLC software, and communicates with the vision software through the modbus protocol when the vehicle lamp silicone lens subassembly 3 reaches the detection position. The host computer obtains the scanning results of the 3D laser scanner 2 through the vision software, and calculates the protruding height of the light guide tooth according to the scanning results. Finally, the host computer displays the protruding height results of the light guide tooth on the interface, and records the log in the database for subsequent big data analysis.

[0147] The above-described specific embodiments further detail the technical problems solved by the present application, technical solutions, and beneficial effects. It should be understood that the above-described specific embodiments are merely examples of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included within the scope of protection of the present application.

Claims

1. A method for measuring the light guide teeth of a silicone lens sub-assembly for automotive headlights, characterized in that, It comprises the following steps: Step S1, collecting three-dimensional data of the car lamp silica lens subassembly by a 3D laser scanner; Step S2, performing ROI region segmentation on the three-dimensional data of the car lamp silica lens subassembly to obtain ROI region point cloud data containing the light guide teeth and the reference surface; Step S3, fitting the reference surface to the obtained ROI region point cloud data to obtain a mathematical model of the reference surface; Step S4, correcting the position of the light guide teeth; Step S5, calculating the relative height of the light guide teeth to the reference surface to obtain the protruding height of the light guide teeth.

2. The method of claim 1, wherein the method further comprises: The ROI region segmentation condition in the step S2 is as follows: X L ≤ X ≤ X R ; Y L ≤ Y ≤ Y R ; Z L ≤Z≤Z R ; Wherein, X, Y, Z are the coordinates of the point cloud in three-dimensional space obtained by ROI region segmentation, X L , Y L , Z L are the minimum values of the three-dimensional coordinates of the ROI region, X R , Y R , Z R are the maximum values of the three-dimensional coordinates of the ROI region.

3. The method of claim 1, wherein the method further comprises: The step S3 specifically comprises the following steps: Step S31, randomly selecting three points that do not coincide and are not on the same straight line from the ROI region point cloud data; Step S32, performing three-dimensional plane fitting on the three selected points to obtain a fitting plane; Step S33, calculating the distance L of the remaining points in the ROI region point cloud data to the fitting plane; Step S34, setting a distance threshold H, if L < H, the point is determined as a plane point; Step S35, if the number of plane points > 95% of the total number of points in the ROI region point cloud data, the fitting plane is determined as the reference surface, if the number of plane points ≤ 95% of the total number of points in the ROI region point cloud data, steps S31-S34 are repeatedly executed.

4. The method of claim 1, wherein the method further comprises: The step S4 specifically comprises the following steps: Step S41, obtaining a cubic region of the car lamp silica lens subassembly by using a fixed region segmentation method; Step S42, extracting the contour points of the sample edge; Step S43, performing straight line fitting on the extracted contour points by using a RANSAC algorithm to obtain an optimal straight line; Step S44, extracting the midpoint of the optimal straight line; Step S45, calculating a transformation matrix and a rotation angle, and obtaining a new position of the cubic region where the light guide teeth are located after offset in the XY plane according to the transformation matrix and the rotation angle; Step S46, obtaining the new position of the cubic region where the light guide teeth are located after offset in the XY plane according to the calculated transformation matrix and the rotation angle, thereby correcting the position of the light guide teeth.

5. The method of claim 4, wherein the step of measuring the light guide teeth comprises the step of: The segmentation condition of the cubic region of the car lamp silica lens subassembly in the step S41 is as follows: ​ X' L ≤ X' ≤ X' R ; Y' L ≤ Y' ≤ Y' R ; Z' L ≤ Z' ≤ Z' R ; Wherein, X', Y', Z' are the coordinates of the point cloud in the three-dimensional space within the cube region of the vehicle lamp silica gel lens sub-assembly, X' L , Y' L , Z' L are the minimum values of the three-dimensional coordinates of the cube region of the vehicle lamp silica gel lens sub-assembly, X' R , Y' R , Z' R are the maximum values of the three-dimensional coordinates of the cube region of the vehicle lamp silica gel lens sub-assembly.

6. The method of claim 5, wherein the step of measuring the light guide teeth comprises the step of: The step S42 specifically comprises the following steps: ​ Step S421, normal vector estimation, specifically comprising the following steps: For each point p in the point cloud within the cuboid region of the vehicle light silicone lens sub-assembly i KD tree to find the K nearest neighbor points to point p i KD tree to find the K nearest neighbor points to point p Fitting the best fitting plane of the neighborhood points by using PCA, solving the normal vector of the best fitting plane of the neighborhood points, calculating a covariance matrix C, and the calculation formula of the covariance matrix C is as follows: Eigenvalue decomposition of the covariance matrix C, normal vector n i is the eigenvector corresponding to the smallest eigenvalue of the decomposition Unifying all normal vectors to the view point; Step S422, boundary point detection, specifically comprising the following steps: obtaining points with significant difference between the normal vector and the neighborhood normal vector, calculating the normal vector difference of each point p i , calculating the average angle θ i between the normal vector n i of each point p i and the neighborhood normal vector n j , the average angle θ i is calculated as follows: If θ i is greater than a set threshold, the point is determined as a contour point.

7. The method of claim 6, wherein the step of measuring the light guide teeth comprises the step of: The step S43 specifically comprises the following steps: ​ Step S431, randomly selecting two non-coincident contour points to perform fitting to obtain a fitting straight line; Step S432, respectively counting the distance L' of each remaining contour point to the fitting straight line, and setting a distance threshold H'; Step S433, if L' < H', the contour point is determined as a straight line point, and the contour point is the straight line point; Step S434, if the number of straight line points > 95% of the total number of contour points, the fitting straight line is determined as the optimal straight line, Step S435, if the number of straight line points ≤ 95% of the total number of contour points, repeat steps S431-S434.

8. The method of claim 7, wherein the step of measuring the light guide teeth comprises the step of: The step S44 specifically includes the following steps: ​ The maximum value and minimum value of the profile points of the optimal straight line are calculated to obtain the profile points (x min ,y min ,z min ) corresponding to the maximum value and the profile points (x max ,y max ,z max ) corresponding to the minimum value, and the midpoint of the optimal straight line is calculated as (x center ,y center ,z center ), and the calculation formula of the midpoint of the optimal straight line is as follows, x center = (x max - x min ) ÷ 2; y center = (y max - y min ) ÷ 2; z center = (z max - z min ) ÷ 2.

9. The method of claim 4, wherein the method further comprises: The step S5 specifically includes the following steps: Based on the light guide tooth in the cube region segmentation obtained light guide tooth region point cloud; Statistical distribution of Z value, take the most Z value distribution as the reference value, take the points within 0.01mm distance from the reference value as the set of points to be calculated, and then calculate the average point of the set of points to be calculated; Finally, calculate the height of the average point to the reference surface to obtain the protruding height of the light guide tooth.

10. A system for measuring the light guide teeth of a vehicle lamp silicone lens subassembly using the method of any one of claims 1-9, wherein: It comprises: The mechanical arm is used to grab the car lamp silica gel lens sub-assembly and place the car lamp silica gel lens sub-assembly on the sliding table; The sliding table is used to move the car lamp silica gel lens sub-assembly; The 3D laser scanner is used to scan the car lamp silica gel lens sub-assembly moving to the detection position once; The host computer is installed with PLC software and vision software, the host computer controls the action of the mechanical arm and the sliding table through the PLC software, and communicates with the vision software when the car lamp silica gel lens sub-assembly reaches the detection position; The host computer obtains the scanning result of the 3D laser scanner through the vision software, and calculates the protruding height of the light guide tooth according to the scanning result; Finally, the host computer displays the interface of the protruding height of the light guide tooth, and records the log in the database.

Images