Product size and form and location tolerance measuring device and method based on machine vision

Abstract

The invention provides a product dimension and form and location tolerance measuring device and method based on machine vision, an external dimension detection module of the measuring device comprises a detection rotating mechanism, one side of the detection rotating mechanism is provided with a linear array camera and a telecentric lens which move axially, and the other side of the detection rotating mechanism is provided with a parallel light source which moves axially; the internal dimension detection module comprises an axial moving mechanism, the axial moving mechanism is arranged at the end part of the detection rotating mechanism, and a laser displacement sensor rotating in the circumferential direction is arranged on the axial moving mechanism; the image and data processing system comprises an image acquisition module, an image positioning module, an image contour processing module, a size calculation module and a data analysis processing module which are connected in sequence, the image acquisition module is connected with the linear array camera, the size calculation module is further connected with the data acquisition module, and the data acquisition module is connected with the laser displacement sensor. Non-contact automatic measurement is adopted, the appearance of a pipe product is prevented from being damaged, and the detection precision and efficiency are improved.

Description

Technical Field

[0001] The invention relates to the technical field of measuring the size and form and position tolerance of pipe parts, and in particular to a device and method for measuring the size and form and position tolerance of products based on machine vision. Background Art

[0002] The current measurement methods for pipe parts are mainly divided into manual self-inspection and automatic detection by intelligent equipment. Automatic detection by intelligent equipment is mainly contact sensor measurement. The measurement data includes process dimensions, form and position tolerances, threads, gaps, etc., which are mainly contact measurements. It is easy to cause damage to the appearance of high-end products. Therefore, manual measurement is mainly used for high-end products.

[0003] However, manual measurement mainly uses vernier calipers, micrometers, internal diameter gauges, special measuring tools, etc., which are labor-intensive and require manual recording of measurement results, with a high error rate. In addition, the detection speed is slow, the accuracy is insufficient, the consistency is poor, only some dimensions can be detected, and there are problems such as missed dimensions. Summary of the invention

[0004] The present invention provides a device and method for measuring product size and form and position tolerance based on machine vision, which adopts non-contact automatic measurement to avoid damaging the appearance of pipe products and improve detection accuracy and efficiency.

[0005] The technical solution of the present invention is implemented as follows: a product size and form tolerance measuring device based on machine vision includes an image and data processing system and an operating table, an external size detection module and an internal size detection module are arranged on the operating table, the external size detection module includes a detection rotation mechanism, an axially movable linear array camera and a telecentric lens are arranged on one radial side of the detection rotation mechanism, and an axially movable parallel light source is arranged on the other side, the internal size detection module includes two axial moving mechanisms, the axial moving mechanism is arranged at the axial end of the detection rotation mechanism, and a circumferentially rotating laser displacement sensor is arranged on the axial moving mechanism;

[0006] The image and data processing system includes an image acquisition module, an image positioning module, an image contour processing module, a size calculation module and a data analysis and processing module which are connected in sequence. The image acquisition module is connected to the linear array camera, the size calculation module is also connected to the data acquisition module, and the data acquisition module is connected to the laser displacement sensor.

[0007] Furthermore, it also includes a feeding mechanism, which includes a feeding workbench, on which a feeding rotating mechanism is arranged, and above the end of the feeding rotating mechanism a scanner and a strip light are arranged, and the scanner is connected to the data analysis and processing module.

[0008] Furthermore, the feeding mechanism also includes an XZ-axis linear module, which is located above the feeding workbench and the operating table, and a gripping mechanism is provided on the XZ-axis linear module.

[0009] Furthermore, a material unloading conveying mechanism is also provided on the loading workbench, and the material unloading conveying mechanism is located below the loading rotating mechanism. A material unloading port is provided between the loading workbench and the operating table. The loading end of the material unloading conveying mechanism is located at the material unloading port, and a second product proximity switch is provided at the loading end.

[0010] Furthermore, the detection rotation mechanism includes two support seats arranged at intervals along the Y-axis, and an active roller driven by a motor is arranged on the support seat. Driven rollers are arranged on both sides of the upper end of the active roller. The driven rollers are driven to rotate by the friction of the active roller. A V-shaped support groove is formed between the two driven rollers, and a first product proximity switch is arranged between the two support seats.

[0011] Furthermore, the grasping mechanism includes a connecting seat, on which a rotating plate driven by a motor is arranged, and on which a pneumatic gripper is arranged in a circumferential direction.

[0012] Furthermore, both axial ends of the feeding rotating mechanism are provided with gear assemblies, and the gear assembly at at least one end can move axially.

[0013] Furthermore, positioning components are provided at both axial ends of the detection rotating mechanism, and the positioning component at at least one end can move axially; the positioning component includes a support plate, on which an axially sliding support rod is provided, and a support spring and a positioning roller are provided at one end of the support rod close to the detection rotating mechanism, and the end of the tube product is against the positioning roller.

[0014] Furthermore, the axial moving mechanism includes a Y-axis linear module, a Y-axis slide is arranged on the Y-axis linear module, a motor-driven sensor mounting rod is arranged on the Y-axis slide, and the laser displacement sensor is arranged on the sensor mounting rod.

[0015] The method for measuring product size and shape and position tolerance based on machine vision, using the measuring device, includes the following steps:

[0016] (1) Place the tube product to be tested on the detection rotating mechanism;

[0017] (2) The linear array camera, the telecentric lens, and the parallel light source are moved axially from one end of the tube product to the other end, and the image acquisition module acquires an axial cross-sectional image of the tube product through the linear array camera;

[0018] (3) The image positioning module locates and binds the cross-sectional image according to the data model of the tube product, and automatically summarizes and finds the number of pixels corresponding to the area to be detected and the size to be detected;

[0019] (4) The image contour processing module fits the contour edge of the area to be detected to obtain the precise number of pixels. The size calculation module calculates the size to be detected based on the precise number of pixels after the camera pixel mapping distance value is calibrated;

[0020] (5) The detection rotating mechanism drives the tube product to rotate a set angle for multiple times, and repeats steps (2) to (4) after each rotation to obtain the dimensions to be detected of multiple cross-sectional images, thereby obtaining the external dimensions of the tube product;

[0021] (6) The axial moving mechanism drives the laser displacement sensor to extend into the specified position inside the tube product. The laser displacement sensor rotates circumferentially for one circle. The data acquisition module collects data of a circumferential section through the laser displacement sensor. The size calculation module uses the three-point method to calculate the inner diameter value of the specified position.

[0022] Furthermore, the method further includes the following steps: the data analysis and processing module obtains the product identification of the tube product to be tested through a scanner, matches the external dimensions and inner diameter values ​​obtained in steps (5) and (6) with the product identification, and analyzes and classifies the external dimensions and inner diameter values ​​and stores them in a database.

[0023] Beneficial effects of the present invention:

[0024] The present invention detects the external dimensions of tube products in a non-contact visual detection manner through the cooperation of an axially movable linear array camera, a telecentric lens and a parallel light source, detects the internal dimensions of tube products in a non-contact manner through the cooperation of a circumferentially rotating laser displacement sensor and a Y-axis linear module, and calculates, analyzes and classifies the dimensions, wall thickness and wall thickness difference, and form and position tolerances of tube products through an image and data processing system.

[0025] The measuring device of the present invention automatically inputs information of tube products through the cooperation of the feeding rotating mechanism, the scanner and the strip light, and realizes automatic loading and unloading of tube products through the cooperation of the XZ linear module and the grabbing mechanism. The measuring device of the present invention realizes the measurement of multiple dimensions of tube parts in a stable, efficient and standard consistent manner, and sorts them according to the classification standards, strictly controls the product quality, and improves the efficiency of part size detection while ensuring the detection accuracy.

[0026] The measuring device of the present invention realizes the rapid and accurate detection of the size and related of small and medium-caliber engine thin-walled pipes, and achieves the recognition workpiece detection speed: ≤60s / piece, workpiece detection accuracy: ≤0.01mm, equipment detection accuracy: ≤0.008mm, equipment detection resolution: ≤0.005mm, and equipment false detection rate: ≤2‰. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0028] Figure 1 It is a front view of the measuring device of the present invention;

[0029] Figure 2 for Figure 1 Left view of

[0030] Figure 3 for Figure 1 A top view of

[0031] Figure 4 is a three-dimensional diagram of the measuring device of the present invention;

[0032] Figure 5 It is a schematic diagram of the internal structure of the measuring device of the present invention;

[0033] Figure 6 It is a structural schematic diagram of the internal dimension detection module;

[0034] Figure 7 It is a structural schematic diagram of the feeding rotating mechanism;

[0035] Figure 8 It is a top view of the feeding rotating mechanism;

[0036] Fig. 9 It is a structural diagram of the grasping mechanism;

[0037] Fig.10 It is a structural schematic diagram of the external dimension detection module;

[0038] Fig.11 It is a schematic diagram of the structure of the positioning component;

[0039] Fig.12 This is a schematic diagram of the calculation principle for the rotation radius r calibration;

[0040] Fig.13 It is a structural diagram of the image and data processing system.

[0041] Operating table 1, detection rotation mechanism 2, support seat 3, driven roller 4, active roller 5, first product proximity switch 6, linear array camera 7, telecentric lens 8, parallel light source 9, positioning assembly 10, support plate 11, support rod 12, support spring 13, positioning roller 14, axial telescopic member 15, Y-axis linear module 16, sensor mounting rod 17, laser displacement sensor 18, image acquisition module 19, image positioning module 20, image contour processing module 21, size calculation module 22, data analysis and processing module 23, data acquisition module 24, pipe products 25, loading workbench 26, loading rotation mechanism 27, gear assembly 28, scanner 29, strip light 30, X-axis linear module 31, Z-axis linear module 32, grabbing mechanism 33, connecting seat 34, rotating plate 35, pneumatic clamp 36, unloading conveying mechanism 37, unloading port 38, second product proximity switch 39, baffle 40, contour photography route 41. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] Example 1

[0044] like Figure 1 , 5 As shown in FIG. 6 , a device for measuring product dimensions and geometric tolerances based on machine vision comprises an operating table 1, on which an external dimension detection module and an internal dimension detection module are arranged, and the external dimension detection module comprises a detection rotation mechanism 2, and the detection rotation mechanism 2 comprises two support seats 3 arranged at intervals along the Y axis, the support seat 3 is fixed on the operating table 1, and a motor-driven active roller 5 is installed on the support seat 3, and driven rollers 4 are installed on both sides of the upper end of the active roller 5, and the driven rollers 4 are driven to rotate by the friction of the active roller 5, and a V-shaped support groove is formed between the two driven rollers 4, and the V-shaped support groove is used to support the end of the tube product 25, and the two driven rollers 4 drive the tube product 25 to rotate, and through the rotation of the tube product 25, multiple axial sections of the tube product 25 can be measured, and images of multiple axial sections are collected in sequence, and finally the outer diameter value is obtained, and the multi-section measurement ensures the accuracy of the measurement. A first product proximity switch 6 is fixed on the operating table 1 between the two support seats 3, which is used to detect whether the tube product 25 to be detected is installed.

[0045] like Figure 4 , 5As shown in FIG. 10 , a linear array camera 7 and a telecentric lens 8 are disposed on one radial side of the detection rotating mechanism 2, and a parallel light source 9 is disposed on the other side. The linear array camera 7, the telecentric lens 8 and the parallel light source 9 can all move along the axial direction of the tube product 25. The linear array camera 7 includes two, corresponding to the upper and lower sides of the tube product 25, respectively. The contour photographing route 41 is shown in FIG. Figure 5 As shown, it is convenient to capture the image contour of the tube product 25 in real time through the line array camera 7. The line array camera 7 is connected to the telecentric lens 8 in a one-to-one correspondence, and the parallel light source 9 is symmetrically arranged with the line array camera 7. Protective covers are fixed to the outer sides of the line array camera 7 and the telecentric lens 8 and the outer sides of the parallel light source 9. The use of the telecentric lens 8 and the parallel light source 9 avoids the deformation of the image, the error caused by the viewing angle, the distortion problem of the ordinary lens, and the uncertainty of the boundary caused by the interference of the inappropriate light source, thereby greatly improving the detection accuracy. Since the line array camera 7 is used, the line array camera 7 only captures the image of a line, so it is necessary to cooperate with the axial movement to drive the line array camera 7 to move, so that the contour of the tube product 25 can be captured.

[0046] like Figure 5 , 6 As shown in Figure 11, the operating table 1 at both axial ends of the detection rotating mechanism 2 is provided with a positioning assembly 10, and the positioning assembly 10 at at least one end can move axially; the positioning assembly 10 includes a support plate 11, and an axially sliding support rod 12 is connected to the support plate 11, and the end of the support rod 12 close to the detection rotating mechanism 2 is successively covered with a support spring 13, and a positioning roller 14 is rotatably connected, and the support plate 11 is connected to an axial telescopic member 15, and the axial telescopic member 15 is fixed on the operating table 1, and the positioning roller 14 is driven to move axially by the axial telescopic member 15, and the axial telescopic member 15 can be a telescopic cylinder, etc. The pipe product 25 is placed on the detection rotating mechanism 2, and the first product proximity switch 6 detects the pipe product 25, and the axial telescopic member 15 moves, so that the positioning roller 14 is against the end of the pipe product 25, and the position of the pipe product 25 is positioned.

[0047] like Figure 5 and 6As shown, the internal dimension detection module includes two axial moving mechanisms, which are arranged at the axial end of the detection rotating mechanism 2. The axial moving mechanism includes a Y-axis linear module 16. When the tube product 25 is placed on the detection rotating mechanism 2, its axial direction is the Y-axis direction. A Y-axis slide is arranged on the Y-axis linear module 16, and a motor is fixed on the Y-axis slide. The output end of the motor is connected to the sensor mounting rod 17, and a laser displacement sensor 18 is fixed at the end of the sensor mounting rod 17 away from the motor. The Y-axis linear module 16 drives the laser displacement sensor 18 to move axially and enter the specified position in the tube product 25. The motor drives the laser displacement sensor 18 to rotate circumferentially through the sensor mounting rod 17. In view of the fact that the tube product 25 is relatively long, a group of laser displacement sensors at each end is used to realize the detection of the internal dimensions of the product, which can reduce the length of the sensor mounting rod 17, thereby reducing the error caused by the long sensor mounting rod 17. During simultaneous detection, two groups of laser displacement sensors 18 can be operated simultaneously to improve the detection efficiency.

[0048] like Fig.13 As shown, the measuring device also includes an image and data processing system, which includes an image acquisition module 19, an image positioning module 20, an image contour processing module 21, a size calculation module 22 and a data analysis processing module 23 connected in sequence. The image acquisition module 19 is connected to the linear array camera 7. The image acquisition module 19 obtains the axial cross-sectional image of the tube product 25 through the linear array camera 7. The image positioning module 20 positions and binds the tube product 25 according to the data model established for the product to be detected. If the tube product 25 is placed out of alignment, the positioning and binding can automatically find the position and direction of the area to be detected, and automatically summarize and find the area and size to be detected. The tube product 25 is preliminarily positioned by the detection rotation mechanism 2 and the positioning component 10, and then combined with the positioning and binding of the image positioning module 20, the detection positioning and target locking of the tube product 25 are fully guaranteed, such as finding the edge lines of two corresponding diameters in the image from the image, and obtaining the number of pixels between the two edge lines.

[0049] The image contour processing module 21 performs filtering processing on the axial cross-sectional image, segments the edge line image according to the sub-pixel technology, fits the edge of the product outline, and obtains the precise value of the number of pixels between the two edge lines. The size calculation module 22 calculates the size to be detected, such as the length and outer diameter of the tube product 25, after calibrating the camera pixel mapping distance value. The specific method is as follows:

[0050] Calibration: Camera pixel mapping distance value calibration, the real distance value corresponding to one pixel on the image. According to the principle of camera detection size: Lobject (object length) = N (number of pixels) * e (camera pixel mapping distance value), it is known that in order to calculate the actual length of the object, N (number of pixels) and e (camera pixel mapping distance value) must be known. N (number of pixels) can be directly counted from the collected image, and e (camera pixel mapping distance value) needs to be obtained through calibration.

[0051] The calibration process is as follows: the length L of a standard object is known, and the number of pixels N corresponding to the length L is obtained by directly counting the image of the standard object. Therefore, the mapping pixel mapping distance value e=L / N can be obtained.

[0052] The size calculation module 22 is also connected to the data acquisition module 24, and the data acquisition module 24 is connected to the laser displacement sensor 18. The data acquisition module 24 collects data of a circumferential section through the laser displacement sensor 18 and transmits it to the size calculation module 22. The size calculation module 22 uses the three-point method to calculate the inner diameter value of the circumferential section. The specific method is as follows:

[0053] The rotation radius calibration of the inner diameter measurement sensor (laser displacement sensor 18) is known from the three-point method of the inner diameter measurement principle: as long as the rotation radius r of the sensor is known (the rotation radius r refers to the distance between the rotation center O of the laser displacement sensor 18 and the emission point of the laser displacement sensor), the inner diameter value of the measured pipe product 25 can be calculated, such as Fig.12 As shown, the calibration process of the rotation radius r of the laser displacement sensor 18 is as follows:

[0054] The XY coordinate system is established with the rotation center O of the laser displacement sensor 18 as the coordinate origin. (X0, Y0) is the coordinate of the inner circle O0 of the tube product 25, L i Point A on the inner circle of the tube product 25 i The distance from the laser sensor emission point, ɑ i For line OA i The angle between the line OA and the X axis i is the rotation center O and point A i The connection;

[0055] Given the inner diameter D of the calibration piece, solve the equation [(L i +r)cosɑ i -X0] 2 +[(L i +r)sinɑ i -Y0] 2 =(D / 2) 2 (i=1, 2, 3), the rotation radius r of the emitting point of the laser displacement sensor is obtained;

[0056] Based on the circumferential cross-sectional data measured by the laser displacement sensor 18 and the rotation radius r of the laser displacement sensor emission point, the inner circle O0 coordinates and inner diameter of the pipe product 25 to be measured can be calculated, and the internal coaxiality can be further obtained. In combination with the outer diameter value, dimensions such as wall thickness can also be obtained.

[0057] The size calculation module 22 transmits the calculated values ​​to the data analysis and processing module 23, which analyzes and classifies them and stores them in a database.

[0058] The method for measuring product size and shape and position tolerance based on machine vision, using the measuring device, includes the following steps:

[0059] (1) Place the tube product 25 to be tested on the detection rotating mechanism 2;

[0060] (2) The linear array camera 7, the telecentric lens 8 and the parallel light source 9 are moved axially from one end to the other end of the tube product 25, and the image acquisition module 19 acquires an axial cross-sectional image of the tube product 25 through the linear array camera 7;

[0061] (3) The image positioning module 20 positions and binds the cross-sectional image according to the data model of the tube product 25, and automatically summarizes and finds the area and size to be detected;

[0062] (4) The image contour processing module 21 fits the contour edge of the area to be detected, and the size calculation module 22 calculates the size to be detected after calibration;

[0063] (5) The detection rotating mechanism 2 drives the tube product 25 to rotate a set angle for multiple times, and repeats steps (2) to (4) after each rotation to obtain the dimensions to be detected of multiple cross-sectional images, thereby obtaining the external dimensions of the tube product 25;

[0064] (6) The axial movement mechanism drives the laser displacement sensor 18 to extend into a designated position in the tube product 25, and the laser displacement sensor 18 rotates circumferentially for one circle. The data acquisition module 24 collects data of a circumferential cross section through the laser displacement sensor 18, and the size calculation module 22 calculates the inner diameter value of the designated position using a three-point method;

[0065] The data analysis and processing module 23 analyzes the external dimensions and inner diameter values ​​obtained in steps (5) and (6), such as further obtaining the wall thickness and wall thickness difference, geometric tolerance, etc. from the detected dimensions, classifies and collects them according to the set classification standards, and stores them in the database.

[0066] Example 2

[0067] This embodiment is basically the same as embodiment 1, except that: Figure 2-5 As shown in 7, it also includes a feeding mechanism, which includes a feeding workbench 26, on which a feeding rotating mechanism 27 is arranged, and the feeding rotating mechanism 27 includes two end seats, on which a rotating driving wheel and a driven wheel are installed, the driving wheel is driven to rotate by a motor, and the tube product is driven to rotate by the driving wheel, and both ends of the axial direction of the feeding rotating mechanism 27 are provided with a gear assembly 28, and the gear assembly 28 at at least one end can move axially, the gear assembly 28 at one end is an end roller fixed on the feeding workbench 26, and the gear assembly 28 at the other end is a baffle 40 driven by a telescopic cylinder, and the tube product 25 to be tested is positioned by the cooperation of the end roller and the baffle 40.

[0068] like Figure 5 and 7 As shown, a scanner 29 and a strip light 30 are fixed above the end of the feeding rotating mechanism 27, and the scanner 29 is connected to the data analysis and processing module 23. The scanner 29 is used to automatically input the product identification of the tube product 25 to be tested, and the strip light 30 is used to ensure the ambient brightness of the scanner 29. The tube product 25 to be tested is placed on the feeding rotating mechanism 27, and the feeding rotating mechanism 27 drives the tube product 25 to rotate. The scanner 29 inputs the product identification on the tube product 25, and the data analysis and processing module 23 obtains the product identification of the tube product 25 to be tested through the scanner 29, and the external dimension and inner diameter value calculated by the size calculation module 22 correspond to the product identification.

[0069] like Figure 5 As shown, the feeding mechanism also includes an XZ-axis linear module, which is located above the feeding workbench 26 and the operating table 1. The XZ-axis linear module includes an X-axis linear module 31 and a Z-axis linear module 32. The X-axis linear module 31 is perpendicular to the Y-axis linear module 16, and the Z-axis linear module 32 is arranged in the vertical direction and moves along the X-axis linear module 31. A gripping mechanism 33 is arranged on the Z-axis linear module 32, and the gripping mechanism 33 is driven to move up and down by the Z-axis linear module 32.

[0070] like Fig. 9 As shown, the grabbing mechanism 33 includes a connecting seat 34 arranged on the Z-axis linear module 32, a motor is fixed on the connecting seat 34, the output end of the motor is connected to the rotating plate 35, and a pneumatic gripper 36 is fixed on the rotating plate 35 along the circumference. The pneumatic gripper 36 is used to grip the tube component, and then the motor drives the rotating plate 35 to rotate, so that the empty pneumatic gripper 36 is downward to grip the next tube product 25 again, so as to facilitate the transportation of multiple products at one time.

[0071] like Figure 5 and 8As shown, a material unloading conveying mechanism 37 is also provided on the loading workbench 26, and the material unloading conveying mechanism 37 is located below the loading rotating mechanism 27. The material unloading conveying mechanism 37 is a conveyor belt driven by a motor, and push plates are evenly distributed on the outer side of the conveyor belt along the circumferential direction. A material unloading port 38 is provided between the loading workbench 26 and the operating table 1, and the loading end of the material unloading conveying mechanism 37 is located at the material unloading port 38, and a second product proximity switch 39 is provided at this end.

[0072] The Z-axis linear module 32 drives the grabbing mechanism 33 to move along the X-axis linear module 31 to the loading workbench 26, the grabbing mechanism 33 moves down, grabs the tube product 25 to be tested with the coded information entered, then moves up and returns along the X-axis linear module 31, and places the tube product 25 to be tested on the detection rotating mechanism 2. After the detection of the tube product 25 to be tested is completed, the grabbing mechanism 33 moves down to remove the tube product 25 that has been removed, then moves up to reset, and the motor drives the rotating plate 35 to rotate so that another tube product 25 to be tested faces downward and is placed on the detection rotating mechanism 2; after all the tube products 25 on the grabbing mechanism 33 have been detected, the Z-axis linear module 32 drives the grabbing mechanism 33 to move along the X-axis linear module 31 to the unloading port 38, and places the inspected tube product 25 on the unloading conveying mechanism 37.

[0073] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A device for measuring product dimensions and geometric tolerances based on machine vision, including an image and data processing system and an operating table, characterized in that: The operating table is provided with an external dimension detection module and an internal dimension detection module. The external dimension detection module includes a detection rotation mechanism. An axially movable linear array camera and a telecentric lens are arranged on one radial side of the detection rotation mechanism, and an axially movable parallel light source is arranged on the other side. The internal dimension detection module includes two axial moving mechanisms. The axial moving mechanisms are arranged at the axial ends of the detection rotation mechanism, and a circumferentially rotating laser displacement sensor is arranged on the axial moving mechanisms. The image and data processing system includes an image acquisition module, an image positioning module, an image contour processing module, a size calculation module and a data analysis and processing module which are connected in sequence. The image acquisition module is connected to the linear array camera, the size calculation module is also connected to the data acquisition module, and the data acquisition module is connected to the laser displacement sensor.

2. The device for measuring product size and geometric tolerance based on machine vision according to claim 1, characterized in that: It also includes a feeding mechanism, which includes a feeding workbench, a feeding rotating mechanism is arranged on the feeding workbench, a scanner and a strip light are arranged above the end of the feeding rotating mechanism, and the scanner is connected to the data analysis and processing module.

3. The device for measuring product size and geometric tolerance based on machine vision according to claim 2, characterized in that: The feeding mechanism also includes an XZ-axis linear module, which is located above the feeding workbench and the operating table, and a gripping mechanism is arranged on the XZ-axis linear module.

4. The device for measuring product size and geometric tolerance based on machine vision according to claim 2 or 3, characterized in that: A material unloading conveying mechanism is also provided on the loading workbench, which is located below the loading rotating mechanism. A material unloading port is provided between the loading workbench and the operating table. The loading end of the material unloading conveying mechanism is located at the material unloading port, and the loading end is provided with a second product proximity switch.

5. The device for measuring product size and geometric tolerance based on machine vision according to claim 1, characterized in that: The detection rotation mechanism includes two support seats arranged at intervals along the Y-axis, and an active roller driven by a motor is arranged on the support seat. Driven rollers are arranged on both sides of the upper end of the active roller. The driven rollers are driven to rotate by the friction of the active roller. A V-shaped support groove is formed between the two driven rollers, and a first product proximity switch is arranged between the two support seats.

6. The device for measuring product size and geometric tolerance based on machine vision according to claim 3, characterized in that: The grabbing mechanism comprises a connecting seat, on which a rotating plate driven by a motor is arranged, and on the rotating plate, a pneumatic gripper is arranged in a circumferential direction.

7. The device for measuring product size and geometric tolerance based on machine vision according to claim 2, characterized in that: Both ends of the axial direction of the feeding rotating mechanism are provided with a gear assembly, and the gear assembly at at least one end can move axially.

8. The device for measuring product size and geometric tolerance based on machine vision according to claim 1, characterized in that: Positioning assemblies are provided at both axial ends of the detection rotating mechanism, and the positioning assembly at at least one end can move axially; the positioning assembly includes a support plate, on which an axially sliding support rod is provided, and a support spring and a positioning roller are provided at one end of the support rod close to the detection rotating mechanism, and the end of the tube product is against the positioning roller.

9. A method for measuring product dimensions and geometric tolerances based on machine vision, characterized in that: The measuring device according to any one of claims 1 to 8 comprises the following steps: (1) Place the tube product to be tested on the detection rotating mechanism; (2) The linear array camera, the telecentric lens, and the parallel light source are moved axially from one end of the tube product to the other end, and the image acquisition module acquires an axial cross-sectional image of the tube product through the linear array camera; (3) The image positioning module locates and binds the cross-sectional image according to the data model of the tube product, and automatically summarizes and finds the number of pixels corresponding to the area to be detected and the size to be detected; (4) The image contour processing module fits the contour edge of the area to be detected to obtain the precise number of pixels. The size calculation module calculates the size to be detected based on the precise number of pixels after the camera pixel mapping distance value is calibrated; (5) The detection rotating mechanism drives the tube product to rotate a set angle for multiple times, and repeats steps (2) to (4) after each rotation to obtain the dimensions to be detected of multiple cross-sectional images, thereby obtaining the external dimensions of the tube product; (6) The axial moving mechanism drives the laser displacement sensor to extend into the specified position inside the tube product. The laser displacement sensor rotates circumferentially for one circle. The data acquisition module collects data of a circumferential section through the laser displacement sensor. The size calculation module uses the three-point method to calculate the inner diameter value of the specified position.

10. The measuring method according to claim 9, characterized in that: The following steps are also included: The data analysis and processing module obtains the product identification of the pipe product to be tested through a scanner, matches the external dimensions and inner diameter values ​​obtained in steps (5) and (6) with the product identification, and analyzes and classifies the external dimensions and inner diameter values ​​and stores them in a database.

Images