A Countersunk Hole Automatic Detection System Based on 3D Line Laser Scanning

Abstract

The present invention discloses a countersunk hole automatic detection system based on three-dimensional line laser scanning, including a numerical control machining device with a line laser profiler and a supporting control system; the control system includes a data acquisition module, a line laser profiler calibration module, a point cloud data preprocessing module, a countersunk hole feature point extraction module, and a countersunk hole geometric parameter calculation module; the machining dimensions of the countersunk hole are generated through the control system. The system provided by the present invention can realize the real-time acquisition and processing of three-dimensional point cloud data of the countersunk hole morphology, thereby effectively improving the detection efficiency and accuracy of the countersunk hole in the assembly and machining process of large mechanical product components.

Description

Technical Field

[0001] The present invention belongs to the technical field of automated machining, and particularly relates to a counterbore hole automatic detection system based on three-dimensional line laser scanning. Background Art

[0002] During the assembly process of large mechanical products, it is necessary to fixedly connect the contacting components. To ensure the appearance of mechanical products, currently, counterbore holes are mainly made on the surface of the components and fixed connections are made using rivets. During the machining process of counterbore holes, the geometric parameters of the counterbore holes will deviate due to factors such as tool wear and tool positioning, thus affecting the assembly connection quality and surface appearance quality of large mechanical products. At present, the assembly of large mechanical products has gradually moved towards automated and digital assembly. However, for the detection of counterbore hole geometric parameters, plug gauges for manual contact detection or monocular camera two-dimensional detection are often used. The detection efficiency and detection accuracy are easily affected by the accessibility of the measured target object, ambient light, or the color distinguishability of the measured target object, making it difficult to meet the detection requirements of mechanical products for automation, high efficiency, and robustness. Three-dimensional line laser scanning has the advantages of non-contact non-destructive measurement and higher measurement accuracy of the component surface. In addition, three-dimensional line laser scanning obtains a large amount of morphological point cloud data of the component surface. The automated processing of this data and the automatic calculation of counterbore hole geometric parameters can further improve the analysis level of counterbore hole machining quality and provide important guiding significance for subsequent rivet connections. In summary, there is an urgent need to develop a non-contact, highly adaptable, and high-precision three-dimensional line laser scanning automatic detection system for the automated assembly of large mechanical products to achieve the detection of the machining quality of counterbore holes in large mechanical product components.

[0003] Patent document CN209588980U discloses a laser measurement target seat for counterbore holes, including a target seat body. The upper part of the target seat body is provided with a groove for fixing a standard seat. The bottom side of the target seat body is set as a conical surface, and the cone angle of the conical surface is the same as the counterbore angle. An exhaust hole is provided in the middle of the target seat body, and the exhaust hole runs through the entire target seat body along the length direction. This device only provides the layout method and usage method of the laser measurement equipment, but the counterbore hole itself is an uneven machining surface, and there are errors in the actual measurement process.

[0004] Patent document CN111531203B discloses an online visual inspection method and system for the countersinking depth of aircraft panels. The method includes: automatically creating countersinking holes on the aircraft panel; obtaining the first image of the countersinking hole at a first set position; obtaining the second image of the countersinking hole at a second set position; processing the above two images to obtain the ellipse equations of the inner and outer edges of the first countersinking hole image and the ellipse equations of the inner and outer edges of the second countersinking hole image; determining the inner and outer ellipse feature points of the first countersinking hole image and the inner and outer ellipse feature points of the second countersinking hole image according to the above ellipse equations; determining the countersinking depth according to the above feature points; obtaining the set countersinking depth; and correcting the countersinking depth according to the countersinking depth and the set countersinking depth. This method directly uses image recognition, but in view of the actual situation of inconsistent flatness of the countersinking holes, it may affect the recognition process. Summary of the Invention

[0005] The object of the present invention is to provide an automatic inspection system for countersinking holes, which can realize real-time acquisition and processing of three-dimensional point cloud data of the morphology of countersinking holes, thereby effectively improving the inspection efficiency and accuracy of countersinking holes in the assembly and processing of large mechanical product components.

[0006] To achieve the object of the present invention, an automatic inspection system for countersinking holes based on three-dimensional line laser scanning is provided, which includes a numerical control processing device with a line laser profiler and a supporting control system.

[0007] The control system includes a data acquisition module, a line laser profiler calibration module, a point cloud data preprocessing module, a countersinking hole feature point extraction module, and a countersinking hole geometric parameter calculation module.

[0008] The data acquisition module is used to acquire the scanned contour line data of the processed workpiece and the pose data of the numerical control processing device.

[0009] The line laser profiler calibration module is used to calibrate the conversion relationship between the coordinate system of the line laser profiler and the coordinate system of the numerical control processing device.

[0010] Based on the calibrated conversion relationship, the point cloud data preprocessing module converts the scanned contour line data of the processed workpiece into a contour line point cloud data set in the numerical control processing device, and the contour line point cloud data set includes a smooth surface point cloud data set and a non-smooth surface point cloud data set.

[0011] The countersinking hole feature point extraction module calculates the neighborhood geometric features between the contour lines according to the smooth surface point cloud data set and the non-smooth surface point cloud data set to construct a corresponding countersinking hole contour line set, and then calculates the neighborhood geometric features within the contour line to obtain a countersinking hole feature point set, and the countersinking hole feature point set includes a hole diameter feature point, an aperture feature point, and a local surface feature point of the object to be measured.

[0012] The countersunk hole geometric parameter calculation module generates corresponding machining dimensions according to the obtained set of countersunk hole feature points.

[0013] The present invention realizes efficient and complete three-dimensional point cloud data acquisition of the countersunk hole morphology through a combination of software and hardware, and performs real-time and accurate detection of the machining quality of the countersunk hole, thus solving the problems of low coverage rate, low efficiency, and poor robustness in the detection of the machining quality of countersunk holes during the assembly of large mechanical products in industrial production.

[0014] Specifically, the numerical control machining equipment includes a machine tool or a robotic arm for numerically controlled drilling.

[0015] Specifically, on the basis of the completion of the installation of the online laser profiler in the data acquisition module, the working parameters of the line laser profiler and the numerical control machining equipment are respectively configured, and then the line laser profiler is carried by the numerical control machining equipment for automatic movement to collect the point cloud data of the line laser profiler and the pose data of the numerical control machining equipment in real time, obtaining a massive set of point cloud data p a =(L1, L2, L3, … L i ) of the measured object, which is composed of the point cloud data L i of i contour lines in sequence according to the scanning order, where L i =(p1, p2, p3, …, p j ) is composed of j points p j (x j , y j , z j ) in sequence according to their positions on the contour line, where x j , y j and z j represent the positions of the point p j on the three coordinate axes, i is the number of contour lines included in the set of point cloud data p a obtained in one acquisition, j represents the number of points included in the set of point cloud data L i of one contour line, and the pose data M t =(m1, m2, m3, …, m i ) is composed of i pose data m i in sequence according to the acquisition order.

[0016] Specifically, after the line laser profiler installation module of the line laser profiler calibration module installs and fixes the line laser profiler on the numerical control machining equipment, by collecting the point cloud data on the surface of the calibration ball obtained by the line laser profiler and the pose data of the numerical control machining equipment at different positions and postures, through a certain calibration method, according to the motion relationship matrix and the result of circle fitting processing, the acquisition pose matrix between the coordinate system of the line laser profiler and the coordinate system of the numerical control machining equipment is obtained. where r is the rotation transformation matrix and t is the translation transformation matrix.

[0017] Specifically, before outputting the contour line point cloud data set, the point cloud data preprocessing module also needs to perform preprocessing, including removing invalid points and noise points, and downsampling, so as to ensure the point cloud density in the area near the countersunk hole, reduce the point cloud density in other non-key areas, improve the calculation speed of the method, and then the surface point cloud data of the measured target object containing the countersunk hole l P a According to the acquired pose matrix T, the point cloud data is fused to obtain the coordinates of the surface point cloud data of the measured target object in the global coordinate system g P a , where l represents the point cloud data set P a is located in the coordinate system of the three-dimensional line laser scanning device, and g represents the point cloud data set P a is located in the global coordinate system.

[0018] Specifically, the point cloud data preprocessing module randomly samples each contour line point cloud data in the scanned contour line data to obtain the corresponding contour line, and uses the contour line as the reference line of the corresponding contour point cloud data for geometric feature statistics, and classifies the contour line point cloud data based on the obtained geometric feature statistics results to obtain a contour line point cloud data set containing smooth surface point cloud data and non-smooth surface point cloud data.

[0019] Specifically, according to the geometric feature statistics results, the contour line point cloud data is classified, and the classification expression is as follows:

[0020]

[0021] In the formula, represents the cumulative sum of the absolute values of the distances from all the point clouds p j (x j , y j , z j ) on the i-th contour line point cloud data to the corresponding contour line l i , and β represents the set threshold.

[0022] When the cumulative sum of the absolute values of the distances is greater than the threshold, it is classified as non-smooth surface point cloud data.

[0023] When the cumulative sum of the absolute values of the distances is less than the threshold, it is classified as smooth surface point cloud data.

[0024] Specifically, the countersunk hole feature point extraction module calculates the neighborhood geometric features between the contour lines of the smooth surface point cloud data and the non-smooth surface point cloud data according to the scanning order of the line laser profiler to construct a countersunk hole contour line set containing all the countersunk holes.

[0025] Specifically, based on the neighborhood geometric features between contour lines, calculate in sequence to obtain the contour line set P cs =(H1, H2, H3, …, H α ), and the process is as follows:

[0026] The expression of the neighborhood geometric features between contour lines is When a certain contour line satisfies , N m =1, otherwise N m =0. Where n is the number of contour lines in the neighborhood, and the value is generally determined according to the movement speed of the numerical control machining equipment and the scanning frequency of the line laser profiler.

[0027] When represents the surface smooth area of the measured target corresponding to this contour line, and when represents the countersunk hole area of the measured target corresponding to this contour line.

[0028] According to the order of acquisition of the contour lines, calculate the neighborhood geometric features between contour lines for each contour line in sequence. When a certain contour line has and several adjacent contour lines immediately following it all have , then these contour lines are considered to belong to the same countersunk hole and added to H α , until a contour line has and then stop adding.

[0029] Specifically, the countersunk hole feature point extraction module processes according to the contour line set H α obtained after a countersunk hole is scanned by a line laser. Sequentially, for each contour line L α contained in H, calculate the neighborhood geometric features within the contour line to obtain the countersunk hole feature point set p i belonging to the countersunk hole diameter and aperture, and the set p u and the local surface feature point set p b of the measured target around the countersunk hole. s .

[0030] Specifically, the extraction process of the countersunk hole diameter feature points is as follows:

[0031] Based on any one contour line in the countersunk hole contour line set, calculate the neighborhood geometric features within the contour line, and start iterating from the (K1 + 1)-th point cloud data at both the beginning and the end of it.

[0032] If starting to iterate from the beginning of the contour line, k q1 <0, and k1 < k q0 < k2, E k1 < Δk < Ek2 , k q1 > k3k q2 > k4k q3 When this occurs, the q-th point is determined as the hole diameter feature point of the countersunk hole. At the same time, the points from the q - K3 to the q - 1 are added to the local surface feature point set of the object under test in the countersunk hole area, and the iterative calculation is stopped.

[0033] If starting the iteration from the end of the contour line, when k q1 > 0, and k1 < k q0 < k2, E k1 < Δk < E k2 , k q1 < k3k q2 < k4k q3 When this occurs, the q-th point is considered as the hole diameter feature point of the countersunk hole. At the same time, the points from the q - K3 to the q - 1 are added to the local surface feature point set of the object under test in the countersunk hole area, and the iterative calculation is stopped.

[0034] Among them, Δk = |k q0 - k q |, k represents the slope of the straight line fitted by several adjacent points on the contour line, k q0 represents the slope of the straight line fitted from the (q - K1 + 1)-th point to the q-th point at the q-th point on the contour line, k q1 represents the slope of the straight line fitted from the q-th point to the (q + K2 - 1)-th point at the q-th point on the contour line, k q2 represents the slope of the straight line fitted from the (q + K2 - 1)-th point to the (q + 2K2 - 2)-th point at the q-th point on the contour line, k q3 represents the slope of the straight line fitted from the (q + 2K2 - 2)-th point to the (q + 3K2 - 3)-th point at the q-th point on the contour line, k1, k2, k3, k4, K1, K2, K3, E k1 , E k2 are all constant terms determined based on the countersunk hole cone angle.

[0035] Specifically, the extraction process of the hole diameter feature point is as follows:

[0036] Based on any one of the contour lines in the countersunk hole contour line set, calculate the neighborhood geometric features within the contour line, and start the iteration from the (q + 3K5)-th point at both the beginning and the end.

[0037] If starting the iteration from the beginning of the contour line, when there is k p1 < 0, k p2 < 0, k p3 < 0 and 4k a < k p0 < -4k aIf so, the q-th point is considered as the aperture feature point of the countersunk hole.

[0038] When iterating from the end of the contour line, when there are k p1 > 0, k p2 > 0, k p3 > 0 and 4k a < k p0 < -4k a If so, the q-th point is considered as the aperture feature point of the countersunk hole.

[0039] Among them, k represents the slope of the straight line fitted by several adjacent points on the contour line, k p0 represents the slope of the straight line fitted from the q-th point to the q + K4 - 1-th point at the q-th point on the contour line, k p1 represents the slope of the straight line fitted from the q - K5 + 1-th point to the q-th point at the q-th point on the contour line, k p2 represents the slope of the straight line fitted from the q - 2K5 + 2-th point to the q - K5 + 1-th point at the q-th point on the contour line, k p3 represents the slope of the straight line fitted from the q - 3K5 + 3-th point to the q - 2K5 + 2-th point at the q-th point on the contour line, k a = (k p1 + k p2 + k p3 ) / 3, and K4 and K5 are constant terms determined based on the countersunk hole cone angle.

[0040] Specifically, the specific calculation process of the countersunk hole geometric parameter calculation module is as follows:

[0041] Process the data of the set of local surface feature points of the target countersunk hole to obtain the expression of the local smooth surface of the target countersunk hole:

[0042] A α x + B α y + C α z + D α = 0

[0043] According to the set of countersunk hole diameter feature points and the set of aperture feature points of the target countersunk hole, perform circle fitting processing respectively to obtain the corresponding countersunk hole diameter and hole diameter;

[0044] And based on the local smooth surface expression, the countersunk hole diameter feature points and the aperture feature points, calculate the depth of the countersunk hole, and its expression is as follows:

[0045]

[0046]

[0047] In the formula, Denote the distance from the m-th point cloud to the plane, and (x m , y m , z m ) denote the coordinates of the m-th point cloud, h α denotes the countersink depth, and n′ denotes the total number of point clouds in the set of countersink feature points.

[0048] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0049] (1) By using a three-dimensional scanning and measuring device to obtain the three-dimensional point cloud data of the measured target object containing the countersink, the surface contour of the measured target object can be visually and comprehensively characterized in the form of three-dimensional point cloud data.

[0050] (2) Only one scanning and measuring task is required, and the automatic identification and high-precision measurement of the countersink hole including the countersink diameter, hole diameter, and countersink depth can be efficiently and accurately completed. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 FIG. is a schematic structural diagram of a countersink hole automatic detection system based on three-dimensional line laser scanning provided in this embodiment;

[0052] Figure 2 FIG. is a schematic structural diagram of the end of a numerical control machining device in the countersink hole automatic detection system provided in this embodiment;

[0053] Figure 3 FIG. is a working flow chart of the countersink hole automatic detection system provided in this embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0054] It should be noted that, without conflict, the embodiments in this application and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the drawings and in combination with the embodiments.

[0055] As Figure 1 shown, a countersink hole automatic detection system based on three-dimensional line laser scanning includes an upper computer S100 of the line laser scanning automatic detection system, a line laser profiler S102, a numerical control system PCL module S104, a numerically controlled drilling machine S106, an industrial Ethernet S108, a database server S110, and an upper computer S112 of the numerically controlled machining tool control system. Each part of the system communicates and transmits data through the industrial Ethernet S108.

[0056] More specifically, the system can be divided into a data acquisition module, a line laser profiler calibration module, a point cloud data preprocessing module, a countersink hole feature point extraction module, and a countersink hole geometric parameter calculation module.

[0057] A data acquisition module, which is used to acquire the scanned contour line data of the workpiece to be machined and the pose data of the numerical control machining equipment.

[0058] A line laser profiler calibration module, which is used to calibrate the conversion relationship between the coordinate system of the line laser profiler and the coordinate system of the machining machine tool.

[0059] A point cloud data preprocessing module, based on the calibrated conversion relationship, converts the scanned contour line data of the workpiece to be machined into a contour line point cloud data set in the numerical control machining equipment, and the point cloud data set includes a smooth surface point cloud data set and a non-smooth surface point cloud data set.

[0060] A countersunk hole feature point extraction module, which calculates the neighborhood geometric features between the contour lines according to the smooth surface point cloud data set and the non-smooth surface point cloud data set to construct a corresponding countersunk hole contour line set, and then calculates the neighborhood geometric features within the contour line to obtain a countersunk hole feature point set, and the countersunk hole feature point set includes a hole diameter feature point, a hole diameter feature point and a local surface feature point of the object to be measured.

[0061] A countersunk hole geometric parameter calculation module, which generates corresponding machining dimensions according to the obtained countersunk hole feature point set.

[0062] The working relationship of each device in the system is as follows: The line laser profiler S102 is installed on the end effector of the numerical control drilling machine S106, powered by a 24V power supply, and connected to the industrial Ethernet S108 and the upper computer S100 of the line laser scanning automatic detection system through a network cable. The upper computer S100 of the line laser scanning automatic detection system controls the line laser profiler S102 to acquire the contour line point cloud data of the object to be measured by adjusting the parameters of the line laser profiler controller.

[0063] Before the start of the countersunk hole automatic detection process, using the data acquisition module, first confirm the successful network connection of the numerical control system PCL module S104, the numerical control drilling machine S106 and the line laser profiler S102 on the upper computer S100 of the line laser scanning automatic detection system. Then configure the working parameters of the line laser profiler S102, including but not limited to sampling frequency, exposure time, measurement range, number of contour line points and photosensitive sensitivity, so that the collected laser contour line is clear, stable and has fewer noise points. Then set the correct address of each motion axis of the numerical control system PCL module S104 in the upper computer S100 of the line laser scanning automatic detection system to ensure that the machine tool pose data can be correctly obtained.

[0064] During the automatic detection process of countersunk holes, it is first necessary to calibrate the relative position relationship between the line laser profiler S102 and the numerically controlled hole-making machine tool S106. By collecting the surface contour point cloud data of the calibration ball obtained by the line laser profiler S102 and the pose data of the numerically controlled hole-making machine tool S106 at different positions and postures, through a certain calibration method, according to the motion relationship matrix and the result of circle fitting processing, the acquisition pose matrix between the coordinate system of the line laser profiler S102 and the coordinate system of the numerically controlled hole-making machine tool S106 is obtained.

[0065] During the automatic detection process of countersunk holes, using the data acquisition module, the upper computer S100 of the line laser scanning automatic detection system first selects the target object to be scanned from the database server S110, and at the same time configures the motion path and motion parameters of the numerically controlled hole-making machine tool S106. Then, through the point cloud data preprocessing module, the countersunk hole feature point extraction module, and the countersunk hole geometric parameter calculation module, the automatic detection of the countersunk holes on the target object to be measured is realized, and finally the detection results are saved to the database server S110.

[0066] As Figure 2 shown, the fixed rod 4 of the line laser profiler S102 and the connecting plate 3 are connected and fixed by screws. The connecting plate 3 is fixedly connected to the line laser profiler 2 by screws. By changing the position of the threaded through holes on the connecting plate 3, the connection and fixation with different types of line laser profilers can be achieved. The fixed rod 4 is clamped and connected to the spindle tool holder of the end effector 1 of the numerical control processing equipment. By replacing the tool holder, the clamping connection between the fixed rod 4 and the spindles of different types of numerical control processing equipment can be achieved. This method can not only realize the rapid installation and fixation of the line laser profiler on the numerical control processing equipment for automatic countersunk hole processing, without the need to transform the structure of the numerical control processing equipment for installation, but also adapt to different types of numerical control processing equipment, with wide installation applicability.

[0067] As Figure 3 shown, a working process for machining countersunk holes is realized based on the countersunk hole automatic detection system provided in the above embodiment.

[0068] Initialization of the processing equipment:

[0069] First, install the line laser profiler. The fixed rod and the connecting plate are connected and fixed by screws. The connecting plate is fixedly connected to the line laser profiler by screws. By changing the position of the threaded through holes on the connecting plate, the connection and fixation with different types of line laser profilers can be achieved. The fixed rod is clamped and connected to the spindle tool holder of the end effector of the numerical control processing equipment. By replacing the tool holder, the clamping connection between the fixed rod 1 and the spindles of different types of numerical control processing equipment can be achieved.

[0070] On the basis of the completion of the installation of the online laser profiler, the working parameters of the line laser profiler and the numerical control processing equipment are configured respectively. Then, the line laser profiler is carried by the numerical control processing equipment to perform automatic movement, and the point cloud data of the line laser profiler and the pose data of the numerical control processing equipment are collected in real time, obtaining a massive point cloud data set p of the object to be measured a =(L1,L2,L3,…L i ), which is composed of the point cloud data L i of i contour lines in sequence according to the scanning order, where L i =(p1,p2,p3,…,p j ), which is composed of j points p j (x j ,y j ,z j ) in sequence according to their positions on the contour line. Among them, x j , y j and z j represent the positions of the point p j on the three coordinate axes. i is the number of contour lines included in the point cloud data set p a obtained by one acquisition, and j represents the number of points included in the point cloud data set L i of one contour line. The pose data M t =(m1,m2,m3,…,m i ), which is composed of i pose data m i in sequence according to the acquisition order.

[0071] Formal work:

[0072] By collecting the point cloud data of the calibration ball surface and the pose data of the numerical control processing equipment obtained by the line laser profiler at different positions and postures of the numerical control processing equipment, through a certain calibration method, according to the motion relationship matrix and the result of circle fitting processing, the acquisition pose matrix between the coordinate system of the line laser profiler and the coordinate system of the numerical control processing equipment is obtained where r is the rotation transformation matrix and t is the translation transformation matrix, which are used for subsequent point cloud data analysis and calculation.

[0073] Then, preprocessing of the point cloud data is carried out. First, the point cloud data set P a obtained by the line laser profiler is processed to remove invalid points, noise points and downsampling. Since the point cloud data obtained by line laser contour scanning contains invalid points beyond the range and outlier noise points, removing them can reduce the error of subsequent processing. In addition, by performing downsampling during the measurement process, the point cloud density in the area near the countersunk hole can be guaranteed, the point cloud density in other non-key areas can be reduced, and the calculation speed of the method can be improved. After that, the point cloud data set l P aPerform point cloud data fusion according to the acquired pose matrix T to obtain the coordinates gP of the point cloud data on the surface of the target object to be measured in the global coordinate system a , where l represents the point cloud data set P a Located in the device coordinate system of the line laser profiler, g represents the point cloud data set P a Located in the global coordinate system.

[0074] The point cloud data preprocessing module sequentially performs statistical analysis on each contour line point cloud data L i Sequentially, that is, randomly sample a2 points from the first and last a1 points of each contour line point cloud data L i , and then perform linear fitting on the obtained a2 points to obtain a straight line l i , calculate the distance from each point p i on the contour line L j (x j , y j , z j ) to the straight line l i , and then classify the contour line according to the geometric feature statistic , and sequentially add it to the laser contour line point cloud data set containing the smooth surface of the target object to be measured and the laser contour line point cloud data set containing the non-smooth surface of the target object to be measured respectively. The classification criteria for a single contour line are as follows:

[0075]

[0076] In the formula, represents the cumulative sum of the absolute values of the distances from all the point clouds p j (x j , y j , z j ) on the i-th contour line point cloud data to the corresponding contour straight line l i , β represents the set threshold. When the cumulative sum of the absolute values of the distances from all the points in the i-th contour line point cloud data L i to the straight line l j is greater than or equal to the threshold β, it can be considered that the contour line L i contains the point cloud data of the non-smooth surface of the target object to be measured. The contour line L i is added to the point cloud data set , otherwise it is considered that the contour line contains the point cloud data of the smooth surface of the target object to be measured, and the contour line L i is added to the point cloud data set In this case, a1 and a2 are generally determined according to the number j of points included in a contour line collected by the line laser profiler. When j = 3200, a1 = 400, a2 = 100, and β = 1.5 are taken.

[0077] The countersunk hole feature point extraction module processes the set of line laser contour line point cloud data containing the smooth surface of the target object to be measured and the set of line laser contour line point cloud data containing the non-smooth surface of the target object to be measured to sequentially calculate, according to the neighborhood geometric features between the contour lines, a set of contour lines P cs =(H1, H2, H3, …, H α ), where α is the number of countersunk holes included in one scan, and H α is the set of contour lines obtained after a countersunk hole is scanned by the line laser. The neighborhood geometric features between the contour lines are as follows:

[0078] When a certain contour line satisfies , N m = 1; otherwise, N m = 0.

[0079] where n is the number of contour lines in the neighborhood. When represents the smooth surface area of the target object corresponding to this contour line, and when represents the countersunk hole area of the target object corresponding to this contour line. According to the order of the collected contour lines, the neighborhood geometric features between each contour line are calculated in sequence. When a certain contour line has and several adjacent contour lines immediately following it all have , then these contour lines are considered to belong to the same countersunk hole and added to H α , until a contour line has and then the addition stops. n is generally determined according to the movement speed of the machine tool or robotic arm and the scanning frequency of the line laser profiler, and generally n = 8 is taken.

[0080] The countersunk hole feature point extraction module processes the set of contour lines H α obtained after a countersunk hole is scanned by the line laser, and sequentially calculates the neighborhood geometric features within each contour line L α included in H i to obtain the set of countersunk hole feature points p u belonging to the countersunk hole diameter and hole diameter and the set p b and the set of local surface feature points p s of the target object around the countersunk hole. Finally, the set of countersunk hole diameter feature points included in the point cloud data set P a is obtained. Set of aperture feature points and the set of local surface feature points of the object to be measured where is the countersink diameter feature of a single countersink hole, is the aperture feature of a single countersink hole, is the local surface feature around a single countersink hole.

[0081] Geometric feature calculation of the countersink diameter feature points: Restore the contour line to the coordinate system of the line laser profiler. At this time, the contour line is in the xz plane, and the neighborhood geometric feature within the contour line is k q0 、k q1 、k q2 、k q3 and Δk = |k q0 -k q |.

[0082] where k represents the slope of the straight line fitted by several adjacent points on the contour line, and k q0 represents the slope of the straight line fitted from the (q - K1 + 1)-th point to the q-th point at the q-th point on the contour line, and k q1 represents the slope of the straight line fitted from the q-th point to the (q + K2 - 1)-th point at the q-th point on the contour line, and k q2 represents the slope of the straight line fitted from the (q + K2 - 1)-th point to the (q + 2K2 - 2)-th point at the q-th point on the contour line, and k q3 represents the slope of the straight line fitted from the (q + 2K2 - 2)-th point to the (q + 3K2 - 3)-th point at the q-th point on the contour line.

[0083] The calculation of the neighborhood geometric feature within the contour line is divided into two parts:

[0084] a) Start iterating from the (K1 + 1)-th point at the beginning of the contour line. If k q1 < 0, and k1 < k q0 < k2, E k1 < Δk < E k2 , k q1 > k3k q2 > k4k q3 > k4k At the same time, add the points from the (q - K3) to the (q - 1) points to the local surface feature of the object to be measured and stop the iterative calculation. Otherwise, q + 1 and recalculate;

[0085] b) Start iterating from the (K1 + 1)-th point at the end of the contour line. When k q1 > 0, and k1 < k q0<k2, E k1 <Δk < E k2 , k q1 <k3k q2 <k4k q3 When it is the case, the q-th point is considered as the feature point of the countersink hole diameter, and this point is added to the countersink hole features belonging to the hole diameter At the same time, the points from the q - K3-th point to the (q - 1)-th point are added to the local surface features of the measured object in the countersink hole area And stop the iterative calculation, otherwise q + 1 and recalculate.

[0086] Among them, k1, k2, k3, k4, K1, K2, K3, E k1 , E k2 The values are determined according to the cone angle of the countersink hole. Take K1 = 15, K2 = 5, K3 = 50, k1 = -0.2, k2 = 0.2, k3 = 1.1, k4 = 1.2, E k1 = 0.55, E k2 = 1.05.

[0087] Calculation of the geometric features of the hole diameter feature points:

[0088] When it is found that the q-th point is the feature point of the countersink hole diameter, a new iteration starts from the q-th point, and the neighborhood geometric features within the contour line can be calculated to obtain H α The countersink hole features belonging to the hole diameter on each contour line within this The neighborhood geometric features within this contour line include k p0 , k p1 , k p2 , k p3 and k a .

[0089] Among them, k represents the slope of the straight line fitted by several adjacent points on the contour line, k p0 represents the slope of the straight line fitted from the q-th point to the (q + K4 - 1)-th point at the q-th point on the contour line, k p1 represents the slope of the straight line fitted from the (q - K5 + 1)-th point to the q-th point at the q-th point on the contour line, k p2 represents the slope of the straight line fitted from the (q - 2K5 + 2)-th point to the (q - K5 + 1)-th point at the q-th point on the contour line, k p3 represents the slope of the straight line fitted from the (q - 3K5 + 3)-th point to the (q - 2K5 + 2)-th point at the q-th point on the contour line, k a =(k p1 + k p2 + k p3 ) / 3, and K4 and K5 are constant terms determined based on the cone angle of the countersink hole.

[0090] The calculation of the neighborhood geometric features within the contour line is divided into two parts:

[0091] a) Start iterating from the (q + 3K5)-th point at the start end of the contour line. If there is a k p1 <0, k p2 <0, k p3 <0 and 4k a <k p0 <-4k a at this time, then the q-th point is considered as the aperture feature point of the countersunk hole, and add this point to the countersunk hole feature belonging to the aperture and stop the iterative calculation, otherwise q - 1 and recalculate;

[0092] b) When starting to iterate from the end of the contour line, if there is a k p1 >0, k p2 >0, k p3 >0 and 4k a <k p0 <-4k a at this time, then the q-th point is considered as the aperture feature point of the countersunk hole, and add this point to the countersunk hole feature belonging to the aperture and stop the iterative calculation, otherwise q + 1 and recalculate.

[0093] Among them, K4 and K5 are determined based on the countersunk hole cone angle, and generally K4 = 2, K5 = 4.

[0094] After that, the countersunk hole geometric parameter calculation module processes the nest diameter feature aperture feature and local surface feature of a single countersunk hole respectively, and determines the nest diameter hole diameter and the countersunk hole depth h α .

[0095] The countersunk hole geometric parameter calculation module processes the set of local surface feature points around a single countersunk hole to handle outliers and noise points and perform plane fitting processing, and obtains the plane equation Ax + By + Cz + D = 0 of the locally smooth surface of the measured target object; α x + B α y + C α z + D α = 0;

[0096] The countersunk hole geometric parameter calculation module processes the nest diameter feature and the aperture feature of a single countersunk hole to handle outliers and noise points respectively, and then perform circle fitting processing respectively, obtains the circle equations of the nest diameter circle and the aperture circle, and determines the nest diameter hole diameter

[0097] According to the plane equation A of the locally smooth surface of the target object to be measured α x + B α y + C α z + D α = 0 and the aperture feature points of a single countersunk hole Calculate the distance d from each point in its set to the plane s , and find its average value, which is the countersunk hole depth h α , where the calculation methods of the distance and the average value are as follows:

[0098]

[0099] where (x m , y m , z m ) are the coordinates of any point in the aperture feature point set ,

[0100]

[0101] where n' is the number of points in the aperture feature point set .

[0102] Finally, by repeatedly calling the countersunk hole geometric parameter calculation module, the countersunk hole diameter feature of the countersunk hole diameter feature point set aperture feature point set and the local surface feature points of the target object to be measured in the single countersunk hole are processed respectively, and all the geometric parameters of the countersunk holes included in the point cloud data P obtained by one scan are obtained, including the countersunk hole diameter aperture and the countersunk hole depth H a =(h1, h2, h3,..., h aperture and the countersunk hole depth H a =(h1, h2, h3,..., h α ).

Claims

1. An automatic detection system for countersunk holes based on three-dimensional line laser scanning, characterized in that, It includes a numerically controlled machining device with a line laser profiler and a control system used in combination; The control system includes a data acquisition module, a line laser profiler calibration module, a point cloud data preprocessing module, a countersunk hole feature point extraction module, and a countersunk hole geometric parameter calculation module; The data acquisition module is used to acquire the scanned contour line data of the workpiece to be machined and the pose data of the numerically controlled machining device; The line laser profiler calibration module is used to calibrate the conversion relationship between the line laser profiler coordinate system and the numerically controlled machining device coordinate system; The point cloud data preprocessing module, based on the calibrated conversion relationship, converts the scanned contour line data of the workpiece to be machined into a contour line point cloud data set in the numerically controlled machining device, and the contour line point cloud data set includes a smooth surface point cloud data set and a non-smooth surface point cloud data set; The countersunk hole feature point extraction module calculates the neighborhood geometric features between the contour lines according to the smooth surface point cloud data set and the non-smooth surface point cloud data set to construct a corresponding countersunk hole contour line set, and then calculates the neighborhood geometric features within the contour line to obtain a countersunk hole feature point set, and the countersunk hole feature point set includes a hole diameter feature point, a hole diameter feature point, and a local surface feature point of the object to be measured; The countersunk hole geometric parameter calculation module generates corresponding machining dimensions according to the obtained countersunk hole feature point set.

2. The automatic detection system for countersunk holes based on three-dimensional line laser scanning according to claim 1, characterized in that, The numerically controlled machining device includes a machine tool or a robotic arm for numerically controlled hole drilling.

3. The automatic detection system for countersunk holes based on three-dimensional line laser scanning according to claim 1, characterized in that, The point cloud data preprocessing module sequentially performs random sampling on each contour line point cloud data in the scanned contour line data set to obtain a corresponding contour straight line, and uses the contour straight line as the reference line of the corresponding contour point cloud data for geometric feature statistics, and classifies the contour line point cloud data based on the obtained geometric feature statistics results to obtain a contour line point cloud data set including smooth surface point cloud data and non-smooth surface point cloud data.

4. The automatic detection system for countersunk holes based on three-dimensional line laser scanning according to claim 3, characterized in that, Classify the contour line point cloud data according to the geometric feature statistics results, and the classification expression is as follows: Where D i l represents the cumulative sum of the absolute values of the distances from all the point clouds p j (x j , y j , z j ) on the point cloud data of the i-th contour line to the corresponding contour line l i , and β represents the set threshold value; When the cumulative sum of the absolute values of the distances is greater than the threshold, it is classified as non-smooth surface point cloud data; When the cumulative sum of the absolute values of the distances is less than the threshold, it is classified as smooth surface point cloud data.

5. The automatic detection system for countersunk holes based on three-dimensional line laser scanning according to claim 1 or 3, characterized in that, Before outputting the contour line point cloud data set, the point cloud data preprocessing module also needs to perform preprocessing, including removing invalid points and noise points, and downsampling processing.

6. The automatic detection system for countersunk holes based on three-dimensional line laser scanning according to claim 1, characterized in that, The countersunk hole feature point extraction module calculates the neighborhood geometric features between the contour lines of the smooth surface point cloud data set and the non-smooth surface point cloud data set according to the scanning order of the line laser profiler to construct a countersunk hole contour line set including all countersunk holes.

7. The automatic detection system for countersunk holes based on three-dimensional line laser scanning according to claim 1, characterized in that, The extraction process of the hole diameter feature point is as follows: Based on any one contour line in the countersunk hole contour line set, perform neighborhood geometric feature calculation within the contour line, and start iteration from the (K1 + 1)-th point cloud data at both the beginning and the end; If starting the iteration from the leading end of the contour line, when k q1 < 0, and k1 < k q0 < k2, E k1 < Δk < E k2 , k q1 < k3k q2 < k4k q3 at this time, then the q-th point is determined as the hole diameter feature point of the countersunk hole, and at the same time, the points from the q - K3 to q - 1 are added to the local surface feature point set of the measured target object in the countersunk hole area and the iterative calculation is stopped; If starting the iteration from the end of the contour line, when k q1 >0, and k1 < k q0 < k2, E k1 < Δk < E k2 , k q1 < k3k q2 < k4k q3 at this time, then the q-th point is considered as the hole diameter feature point of the countersunk hole, and at the same time, the points from the q - K3 to q - 1 are added to the local surface feature point set of the measured target object in the countersunk hole area and the iterative calculation is stopped; where Δk = |k q0 - k q |, k represents the slope of the straight line fitted by several adjacent points on the contour line, and k q0 represents the slope of the straight line fitted from the (q - K1 + 1)-th point to the q-th point at the q-th point on the contour line, and k q1 represents the slope of the straight line fitted from the q-th point to the (q - K2 + 1)-th point at the q-th point on the contour line, and k q2 represents the slope of the straight line fitted from the (q + K2 - 1)-th point to the (q + 2K2 - 2)-th point at the q-th point on the contour line, and k q3 represents the slope of the straight line fitted from the (q + 2K2 - 2)-th point to the (q + 3K2 - 3)-th point at the q-th point on the contour line, and k1, k2, k3, k4, K1, K2, K3, E k1 , E k2 are all constant terms determined based on the countersink hole cone angle.

8. The automatic detection system for countersunk holes based on three-dimensional line laser scanning according to claim 1, characterized in that, The extraction process of the hole diameter feature point is as follows: Based on any one contour line in the countersunk hole contour line set, perform neighborhood geometric feature calculation within the contour line, and start iteration from the (q + 3K5)-th point at both the beginning and the end; If starting the iteration from the leading end of the contour line, when there is a k p1 <0, k p2 <0, k p3 <0 and 4k a <k p0 <-4k a then the q-th point is considered as the hole diameter feature point of the countersunk hole; If starting iteration from the end of the contour line, when there is k p1 > 0, k p2 > 0, k p3 > 0 and 4k a < k p0 < -4k a at this time, then the q-th point is considered as the aperture feature point of the countersunk hole; Among them, k represents the slope of the straight line fitted by several adjacent points on the contour line, k p0 represents the slope of the straight line fitted from the q-th point to the q+K4-1-th point at the q-th point on the contour line, k p1 represents the slope of the straight line fitted from the q-K5+1-th point to the q-th point at the q-th point on the contour line, k p2 represents the slope of the straight line fitted from the q-2K5+2-th point to the q-K5+1-th point at the q-th point on the contour line, k p3 represents the slope of the straight line fitted from the q-3K5+3-th point to the q-2K5+2-th point at the q-th point on the contour line, k a =(k p1 +k p2 +k p3 ) / 3, and K4 and K5 are constant terms determined based on the countersink hole cone angle.

9. The counterbore hole automatic detection system based on three-dimensional line laser scanning according to claim 1, wherein, The specific calculation process of the countersunk hole geometric parameter calculation module is as follows: Perform data processing on the set of local surface feature points of the target countersunk hole to obtain the local smooth surface expression of the target countersunk hole: A α x + B α y + C α z + D α = 0 According to the set of countersink diameter feature points and the set of hole diameter feature points of the target countersunk hole, perform circle fitting processing respectively to obtain the corresponding countersink diameter and hole diameter; And based on the local smooth surface expression, the countersink diameter feature points and the hole diameter feature points, calculate and obtain the depth of the countersunk hole, and its expression is as follows: In the formula, represents the distance from the m-th point cloud to the plane, and (x m , y m , z m ) represents the coordinates of the m-th point cloud, h α represents the countersink depth, and n' represents the total number of point clouds in the set of countersink feature points.

Images