A pipe centering method and laser cutting method based on machine vision

Abstract

The present application relates to a kind of pipe centering method and laser cutting method based on machine vision, the pipe centering method includes: in the case where current pipe is clamped to the below laser cutting head, the pipe parameter of current pipe is acquired;Wherein, pipe parameter includes pipe type;Determine the target pipe centering method corresponding to pipe type;Based on target pipe centering method, the 3D line laser profile sensor that has been calibrated is controlled to take photo to pipe, to obtain pipe profile image;Machine vision detection is carried out to pipe profile image, to obtain machine vision detection result;Based on machine vision detection result, the center position of pipe is determined, so compared with existing capacitive centering, any pipe can be centered, including the special-shaped pipe that capacitive centering cannot center, and can improve centering efficiency.

Description

Technical Field

[0001] This invention relates to the field of laser cutting technology, and in particular to a machine vision-based pipe centering method and laser cutting method. Background Technology

[0002] In the field of laser processing, calculating the center of the tube is fundamental to ensuring processing accuracy; therefore, the accuracy of tube centering is crucial. Typically, most tube clamping chucks have a rectangular profile. Whether it's a rectangular tube, round tube, D-shaped steel, H-shaped steel, or other irregularly shaped tube, each corresponds to a unique minimum bounding rectangle, i.e., a directed bounding box, and the chuck clamping profile coincides with this directed bounding box. Therefore, tube centering essentially involves solving for the coordinates of the chuck center holding the tube and calculating the relative coordinate difference with the machine's rotation axis. The CNC system then calculates the processing path based on this relative positional relationship.

[0003] Existing solutions generally employ capacitive centering. First, distance sensing calibration is performed using a capacitor, with different capacitance values ​​corresponding to the height of the capacitor from the surface of the metal pipe. After capacitor calibration, mechanical rotation center calibration is performed using a rectangular profile with capacitor follow-up calibration to obtain the coordinates of the mechanical rotation center. If the physical structure of the machine does not change, the coordinates of the mechanical rotation center will not change. With the coordinates of the mechanical rotation center obtained, pipe centering is performed, which can be done using methods such as 4-point calibration, 5-point calibration, and 8-point calibration.

[0004] However, since the above-mentioned capacitive centering method takes a long time, it has the problem of slow centering efficiency. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a pipe centering method and a laser cutting method based on machine vision, which solves the technical problems of the inability to center some irregularly shaped pipes and the relatively slow centering efficiency in the prior art.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the main technical solutions adopted by the present invention include:

[0009] In a first aspect, embodiments of the present invention provide a pipe centering method based on machine vision, comprising: acquiring pipe parameters of the current pipe while clamping it under a laser cutting head; wherein the pipe parameters include pipe type; determining a target pipe centering method corresponding to the pipe type; controlling a calibrated 3D line laser contour sensor to take pictures of the pipe based on the target pipe centering method to obtain a pipe contour image; performing machine vision detection on the pipe contour image to obtain a machine vision detection result; and determining the center position of the pipe based on the machine vision detection result.

[0010] In one possible embodiment, the 3D line laser profile sensor includes a line laser emitter, a first bandpass filter, a second bandpass filter, a mirror, a camera, and a lens;

[0011] The line laser emitted by the line laser emitter is perpendicular to the first bandpass filter and is parallel to the cross-section of the pipe. The second bandpass filter is placed in the optical path of the reflected laser reflected from the pipe and is not parallel to the first bandpass filter. The reflected laser signal reaches the lens connected to the camera through a reflector.

[0012] In one possible embodiment, the pipe type includes a cylindrical pipe, and the pipe parameters also include the semi-axis of the pipe's cross-section;

[0013] When the pipe type is a cylindrical pipe, the center position of the pipe is determined based on the machine vision inspection results, including: determining the center position of the pipe based on the machine vision inspection results and the half-axis of the pipe's cross-section.

[0014] In one possible embodiment, performing machine vision detection on the pipe outline image to obtain machine vision detection results includes: performing machine vision detection on the pipe outline image to obtain the pixel coordinates of the first vertex of the pipe; wherein, the first vertex is the point on the arc surface of the circular pipe captured by the 3D line laser profile sensor that is closest to the camera's Z-axis.

[0015] Furthermore, based on the machine vision inspection results and the semi-axis of the pipe's cross-section, the center position of the pipe is determined, including: processing the pixel coordinates of the first vertex of the pipe using the ratio coefficient between the camera coordinates and the real coordinates determined during the calibration process of the 3D line laser profile sensor to obtain the real coordinates of the first vertex of the pipe; using the Z value of the real coordinates of the first vertex as the first coordinate value of the center position of the pipe; and using the sum of the X value of the real coordinates of the first vertex and the semi-axis perpendicular to the camera's Z-axis as the second coordinate value of the center position of the pipe.

[0016] In one possible embodiment, the pipe type includes a cuboid pipe, and the pipe parameters also include the length and width of the rectangular cross-section of the cuboid pipe;

[0017] When the pipe type is a cuboid pipe, the center position of the pipe is determined based on the machine vision inspection results, including: determining the center position of the pipe based on the machine vision inspection results and the length and width of the rectangular cross-section of the cuboid pipe.

[0018] In one possible embodiment, performing machine vision inspection on the pipe outline image to obtain machine vision inspection results includes: performing machine vision inspection on the pipe outline image to obtain the pixel coordinates of the two endpoints of the pipe; wherein the two endpoints are the two endpoints of the face of the cuboid pipe captured by the 3D line laser profile sensor.

[0019] Furthermore, based on the machine vision inspection results and the length and width of the rectangular cross-section of the cuboid tube, the center position of the tube is determined, including: processing the pixel coordinates of the two ends of the tube using the ratio coefficient between the camera coordinates and the real coordinates determined during the calibration process of the 3D line laser profile sensor to obtain the real coordinates of the two ends of the tube; determining the first coordinate value of the center position of the tube based on the pixel coordinates of the two ends and the value of the face perpendicular to the face of the rectangular cross-section of the cuboid tube captured by the 3D line laser profile sensor; and taking the average of the X values ​​of the real coordinates of the two ends as the second coordinate value of the center position of the tube.

[0020] In one possible embodiment, when the pipe type includes irregularly shaped pipes, a calibrated 3D line laser profile sensor is controlled to take pictures of the pipe based on a target pipe centering method to obtain pipe profile images. This includes: controlling the calibrated 3D line laser profile sensor to take pictures of each face of the pipe based on the target pipe centering method to obtain multiple pipe profile images; and performing machine vision inspection on the pipe profile images to obtain machine vision inspection results. This includes: performing machine vision inspection on each pipe profile image in the multiple pipe profile images to obtain the pixel distance from each face of the pipe to the 3D line laser profile sensor.

[0021] Furthermore, based on the machine vision inspection results, the center position of the pipe is determined, including: using the ratio coefficient between the camera coordinates and the real coordinates determined during the calibration process of the 3D line laser profile sensor to process the pixel distance from each face of the pipe to the 3D line laser profile sensor, so as to obtain the real distance from each face of the pipe to the 3D line laser profile sensor; and determining the center position of the pipe based on the real distance from each face of the pipe to the 3D line laser profile sensor.

[0022] In one possible embodiment, the calibration process of the 3D line laser profile sensor includes: controlling a camera to acquire a calibration image when the calibration block used in the calibration process is a frustum calibration block and the top surface of the frustum calibration block is parallel to the first bandpass filter and the line laser emitted by the 3D line laser profile sensor can illuminate the top surface; extracting the high and low platform contour lines of the frustum block based on the calibration image; fitting the high and low platform contour lines into two straight line segments using the least squares method and solving the angle between the two line segments and the X-axis of the image coordinate system; rotating the calibration image by an angle with the image center of the calibration image as the rotation reference point to obtain the rotated calibration image; determining the coordinate difference between the two straight line segments in the Y-axis direction of the image coordinate system in the rotated calibration image, and calculating the scaling factor between the camera coordinates and the real coordinates based on the true high and low coordinate difference of the high and low platforms of the frustum calibration block and the coordinate difference between the two straight line segments in the Y-axis direction of the image coordinate system in the rotated calibration image.

[0023] Secondly, embodiments of this application provide a laser cutting method, including: obtaining the coordinates of a mechanical center and the center position of a pipe; wherein the center position of the pipe is determined based on a machine vision-based pipe centering method as described in the first aspect; planning a cutting path based on the coordinates of the mechanical center and the center position of the pipe; and cutting the pipe based on the cutting path.

[0024] (III) Beneficial Effects

[0025] The beneficial effects of this invention are:

[0026] This application provides a machine vision-based pipe centering method and laser cutting method. By clamping the current pipe under the laser cutting head, the method acquires the pipe parameters of the current pipe and determines a target pipe centering method corresponding to the pipe type. Based on the target pipe centering method, it controls a calibrated 3D line laser contour sensor to take a picture of the pipe to obtain a pipe contour image. The method then performs machine vision inspection on the pipe contour image to obtain machine vision inspection results. Based on the machine vision inspection results, it determines the center position of the pipe. Therefore, compared to existing capacitive centering methods, this method can improve centering efficiency.

[0027] To make the above-mentioned objectives, features and advantages to be achieved by the embodiments of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0028] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 A flowchart of a pipe centering method based on machine vision provided in an embodiment of this application is shown;

[0030] Figure 2A and Figure 2B A schematic diagram showing the outline of the cross-section of a special-shaped tube provided in an embodiment of this application is shown;

[0031] Figure 3 A schematic diagram of a 3D line laser contour sensor provided in an embodiment of this application is shown;

[0032] Figure 4 This diagram illustrates the relative positions of a 3D line laser profile sensor and a tube, as provided in an embodiment of this application.

[0033] Figure 5 A schematic diagram of a frustum calibration block provided in an embodiment of this application is shown;

[0034] Figure 6 This illustration shows a logic diagram for identifying a corresponding circular tube using a single-sided centering method, according to an embodiment of this application.

[0035] Figure 7 This illustration shows a logic diagram for identifying a corresponding cube tube using a single-sided centering method, according to an embodiment of this application.

[0036] Figure 8 This illustration shows a logic diagram for identifying a corresponding rectangular tube in a quasi-searching process, provided in an embodiment of this application.

[0037] Figure 9 A flowchart of a laser cutting method provided in an embodiment of this application is shown. Detailed Implementation

[0038] To better explain and facilitate understanding of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0039] For product applications, capacitive centering can easily reduce the cycle time in automated processing and has low stability. For processing requirements with low fault tolerance and high precision, the tube must be stopped and capacitive centering performed after each cutting distance. Furthermore, tubes with rough surfaces or large radii cannot meet the precision requirements.

[0040] Furthermore, for some concave polygonal irregular-shaped tubes, follow-up operation is not possible (due to the risk of mechanical collision and potential safety hazards). Therefore, capacitive centering cannot be performed on this type of tube. In other words, the principle of capacitive centering is to probe close to the surface of the tube, so it cannot be used on some irregular-shaped tubes. Forcing its use would result in collisions with the tube or the laser head.

[0041] Based on this, the embodiments of this application provide a pipe centering method and a laser cutting method based on machine vision. Considering factors such as efficiency, accuracy, and versatility, the vision system and the mechanical rotation center are calibrated using standard profiles, and then the pipe center coordinates are calculated through the vision system, thus realizing a highly efficient, high-precision, and highly versatile pipe centering system based on machine vision.

[0042] To better understand the above technical solutions, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the present invention can be understood more clearly and thoroughly, and that the scope of the present invention can be fully conveyed to those skilled in the art.

[0043] Please see Figure 1 , Figure 1 A flowchart of a pipe centering method based on machine vision provided in an embodiment of this application is shown. It should be understood that the pipe centering method based on machine vision can be executed by a pipe centering device based on machine vision, and the specific device can be configured according to actual needs; this embodiment is not limited thereto. Specifically, the pipe centering method includes:

[0044] Step S110: With the current pipe clamped under the laser cutting head, obtain the pipe parameters of the current pipe. The pipe parameters include the pipe type.

[0045] It should be understood that the pipe parameters of the current pipe can be entered by the user.

[0046] It should also be understood that the type of pipe used can be set according to actual needs, and the embodiments of this application are not limited thereto.

[0047] For example, the current pipe types can be round pipes, square pipes, and special-shaped pipes. Round pipes can be round tubes, elliptical tubes, racetrack tubes, and D-shaped steel, etc., with at least one curved surface; square pipes can be rectangular tubes, square tubes, racetrack tubes, angle steel, channel steel, H-shaped steel, and D-shaped steel, etc., with at least one flat surface; special-shaped pipes include, but are not limited to, […]. Figure 2A and Figure 2B The irregularly shaped tube shown.

[0048] Step S120: Determine the target pipe centering method corresponding to the pipe type.

[0049] Specifically, when there is a corresponding pipe centering method for each type of pipe, including round pipes, square pipes, and special-shaped pipes, the target pipe centering method corresponding to the pipe type can be determined based on the pipe type.

[0050] Step S130: Based on the target pipe centering method, control the calibrated 3D line laser profile sensor to take pictures of the pipe to obtain the pipe profile image.

[0051] It should be understood that the specific device of the 3D line laser contour sensor can be set according to actual needs, and the embodiments of this application are not limited thereto.

[0052] Optionally, such as Figure 3 As shown, the 3D line laser profile sensor includes a line laser emitter (C), a first bandpass filter (E1), a second bandpass filter (E2), a reflector (D), a camera (A), and a lens (B).

[0053] The line laser emitted by the laser emitter (C) is perpendicular to the first bandpass filter (E1), and the line laser emitted by the line laser emitter (C) can be parallel to the cross-section of the tube (e.g., as shown in the image). Figure 4 As shown, the 3D line laser profile sensor can be fixed on one side of the pipe, and the line laser emitted by the 3D line laser profile sensor is parallel to the cross-section of the pipe. The second bandpass filter (E2) is set in the optical path of the reflected laser reflected by the pipe, and the second bandpass filter (E2) and the first bandpass filter (E1) are not set in parallel. The reflected laser signal reaches the lens (B) connected to the camera (A) through the reflector (D).

[0054] Furthermore, the camera (A) can be a single 2D camera, and it can be selected with different resolutions depending on the actual accuracy requirements. For example, if the working distance of the camera (A) is a far field of view of 300*400mm, and the unit pixel accuracy requirement is 0.1mm / pixel, then the camera's pixel count should be at least 3000*4000 pixels.

[0055] Furthermore, the lens (B) is connected to the camera (A) via a C-mount, CS-mount, or F-mount interface, etc. The lens (B) should have a larger aperture size than the camera (A), and should be a model with a resolution and aperture size larger than the camera, such as a 15-megapixel lens with a 2 / 3" aperture size. The lens's optical axis exits towards the reflex mirror at a 45° angle (this angle can be adjusted according to actual needs, and the embodiments in this application are not limited to this).

[0056] Furthermore, the line laser emitter (C) consists of a fixed-wavelength point light source emitter and an optical scribing prism, and this line laser emitter (C) can emit a line laser of a fixed wavelength. For example, a red point laser emitter with a wavelength of 650 nm can be used in conjunction with a Powell prism to achieve a high-precision fixed-wavelength laser line. The line laser emitter is perpendicular to the exit window and perpendicular to the 0° horizontal line; as shown in the attached diagram. Figure 4 As shown, the laser line is parallel to the exit window and perpendicular to the upper and lower surfaces of the protective structure, and parallel to the cross-section of the object being measured.

[0057] Furthermore, the first bandpass filter (E1) and the second bandpass filter (E2) are a pair of filters placed between the line laser emitter (C) and the camera (A), which only allow light of a fixed wavelength to pass through in order to shield light of other wavelengths from interference.

[0058] It should also be understood that the calibration process of the 3D line laser profile sensor described above can also be set according to actual needs, and the embodiments of this application are not limited thereto.

[0059] Optionally, the calibration block used in the calibration process is as follows: Figure 5 Under the condition that the top surface of the frustum calibration block is parallel to the first bandpass filter and the line laser emitted by the 3D line laser profile sensor can illuminate the top surface, the camera is controlled to acquire a calibration image; the contour lines of the high and low platforms of the frustum block are extracted based on the calibration image; the contour lines of the high and low platforms are fitted into two straight line segments using the least squares method, and the angle between the two line segments and the X-axis of the image coordinate system is solved; the calibration image is rotated by an angle with the image center of the calibration image as the rotation reference point to obtain the rotated calibration image; the coordinate difference between the two straight line segments in the Y-axis direction of the image coordinate system in the rotated calibration image is determined, and the scaling factor between the camera coordinates and the real coordinates is calculated based on the true high and low coordinate difference of the high and low platforms of the frustum calibration block and the coordinate difference between the two straight line segments in the Y-axis direction of the image coordinate system in the rotated calibration image.

[0060] For example, after fixing the 3D line laser profile sensor to the installation position, system coordinate matching and calibration are performed, using methods such as... Figure 5 The truncated pyramid calibration block shown is calibrated to correct for installation deviations and to establish the zero plane as the reference plane. The specific steps of the calibration process are as follows:

[0061] Design a calibration block, which may have a two-plane solid model with elevation differences (e.g., it may be like...). Figure 5The diagram shows a frustum calibration block, with its top surface as a high platform and the lower platforms on its left and right sides as low platforms. A frustum calibration block is installed opposite the 3D line laser contour sensor, with the laser line facing the top surface of the frustum and passing through the low planes on both sides and the central high plane, forming a height coordinate difference between the two planes. An image can be acquired via a camera, with the origin at the top left corner, the horizontal axis as the X-axis, and the vertical axis as the Y-axis. Various thresholding and histogram analysis methods can be used to extract the contour lines corresponding to the high and low platforms, and the least squares method can be used to fit the contour lines of the high and low platforms into two straight line segments, solving for the angle between the two line segments and the X-axis. Subsequently, using the center of the calibration image as the rotation reference point, the calibration image is rotated by an angle. This process is used to obtain the rotated calibration image. Then, the coordinate difference between the two line segments (corresponding to the high and low platforms) in the rotated calibration image along the Y-axis of the image coordinate system is calculated. Furthermore, the true difference between the high and low platforms of the frustum is determined to be 150mm. Therefore, the camera / true coordinate ratio is... pixel / mm, at this point the calibration of the 3D line laser profile sensor is complete.

[0062] It should be noted that the calibration of this 3D line laser profile sensor only needs to be performed once, and it can be used directly thereafter. There is no need to perform calibration every time the center position of the pipe is measured.

[0063] It should also be noted that, for ease of understanding, the entire process of the target pipe centering method will be described below in conjunction with step S140, and will not be described in detail here.

[0064] Step S140: Determine the center position of the pipe based on the machine vision detection results.

[0065] It should be understood that the specific process of determining the center position of the pipe based on the machine vision inspection results can also be set according to actual needs, and the embodiments of this application are not limited thereto.

[0066] Optionally, the pipe centering system corresponding to this pipe centering method can include a motion control module, an image processing module, and a data calculation module, with different processing methods corresponding to different pipe types. The motion control module primarily controls the pipe's rotation during image acquisition. First, the pipe is leveled to ensure the clamping angle is horizontal before image acquisition and pipe rotation. The image processing module mainly includes pipe contour extraction, noise reduction, and endpoint detection, converting key image information into digital coordinates for output to the data calculation module. The data calculation module first pre-reads the pipe type from the processing drawings, such as square pipe, rectangular pipe, round pipe, elliptical pipe, angle steel, channel steel, D-shaped steel, H-shaped steel, racetrack pipe, and other special-shaped pipes, with each pipe type corresponding to its own algorithm logic. Furthermore, the system can automatically call the corresponding algorithm logic based on the pipe type. By performing corresponding calculations on the digital coordinates imported from the image processing module, the coordinates of the pipe's center point can be calculated and transmitted to the CNC system, which can then calculate the processing route based on the relative positional relationship with the machine's rotation center.

[0067] Furthermore, for round and square tubes, a single-sided centering method is used; for irregularly shaped tubes, a precise centering method is used. Single-sided centering refers to the method of calculating the centering by capturing an image of only one side, and it is applicable to high-precision square, rectangular, and round tubes. Precise centering refers to the method of capturing and rotating images of all sides of the tube (e.g., four sides) separately, and then performing the centering calculation; it is applicable to all tubes whose images can be captured by a 3D camera.

[0068] For the cylindrical tube, only one photo is taken. The coordinates of the tube's center are obtained by calculating the horizontal and vertical coordinates of the vertices of the arc in the point cloud. With the system calibration already completed, the distance from the mechanical center to the camera's Z-axis is F_Z, and the distance to the X-axis is F_X. In the camera coordinate system, the mechanical center's coordinates are (M_Z, M_X). After the tube is in place, one photo is taken of any surface to perform vertex detection (e.g., the first vertex D can be the point closest to the Z-axis on the current arc surface). The camera coordinates of the first vertex D are output as (S_Z, S_X). The value S_Z of the first vertex D along the camera's Z-axis is the first coordinate of the tube's center position. The vertical height of the tube's center can be determined by the value S_X of the first vertex D along the camera's X-axis. For example, as... Figure 6 Taking the elliptical tube shown as an example, let the major and minor axes of the elliptical tube be A and B, respectively. Then, the coordinates of the center of the elliptical tube corresponding to the current camera coordinate system are (T_Z, T_X). Where T_Z = S_Z, T_X = S_X + B. In the current state, in the mechanical coordinate system, the difference in X-axis coordinates of the elliptical tube corresponding to the mechanical rotation center (i.e., the crosshair in the figure) is M_Z - T_Z, and the difference in Z-axis coordinates is M_X - T_X.

[0069] It should be understood that, in addition to obtaining it through the first vertex D, it can also be obtained through the vertex opposite the first vertex D, and the embodiments of this application are not limited to this.

[0070] In other words, performing machine vision inspection on the pipe outline image to obtain the machine vision inspection result includes: performing machine vision inspection on the pipe outline image to obtain the pixel coordinates of the first vertex of the pipe; wherein, the first vertex is the point on the arc surface of the circular pipe captured by the 3D line laser profile sensor that is closest to the camera's Z-axis.

[0071] Furthermore, based on the machine vision inspection results and the semi-axis of the pipe's cross-section, the center position of the pipe is determined, including: processing the pixel coordinates of the first vertex of the pipe using the ratio coefficient between the camera coordinates and the real coordinates determined during the calibration process of the 3D line laser profile sensor to obtain the real coordinates of the first vertex of the pipe; using the Z value of the real coordinates of the first vertex as the first coordinate value of the center position of the pipe; and using the sum of the X value of the real coordinates of the first vertex and the semi-axis perpendicular to the camera's Z-axis as the second coordinate value of the center position of the pipe.

[0072] Furthermore, for the cuboid tube, only one photo is triggered. The center coordinates of the cuboid tube are obtained by calculating the horizontal coordinates of the center of the straight line segment in the point cloud and the corresponding vertical coordinates. It is assumed that system calibration has been completed and the tube's posture has been leveled. At this time, the distance from the mechanical center to the camera's Z-axis is F_Z, and the distance to the X-axis is F_X. In the corresponding camera coordinate system, the mechanical center coordinates are (M_Z, M_X). After the tube is in place, one photo is triggered on the plane to perform endpoint detection (e.g., a straight line or a whole with a radius). The camera coordinate system coordinates (S_Z, S_X) of the center D of the line connecting the two endpoints E and F are output. The first coordinate of the tube center is determined by the value S_Z along the camera's Z-axis from this center D coordinate, and the second coordinate of the tube center is determined by the value S_X along the camera's X-axis from this center D coordinate. For example, as... Figure 7 Taking the rectangular tube shown as an example, with length and width A and B respectively, the coordinates of the center of the rectangular tube corresponding to the current camera coordinate system are (T_Z, T_X) to the center of the elliptical tube. Where T_Z = S_Z + A / 2, T_X = S_X. In the current state, in the mechanical coordinate system, the difference in X-axis coordinates of the rectangular tube corresponding to the mechanical rotation center is M_Z - T_Z, and the difference in Z-axis coordinates is M_X - T_X.

[0073] In other words, performing machine vision inspection on the pipe outline image to obtain the machine vision inspection result includes: performing machine vision inspection on the pipe outline image to obtain the pixel coordinates of the two endpoints of the pipe; wherein, the two endpoints are the two endpoints of the face of the cuboid pipe captured by the 3D line laser profile sensor.

[0074] Furthermore, based on the machine vision inspection results and the length and width of the rectangular cross-section of the cuboid tube, the center position of the tube is determined, including: processing the pixel coordinates of the two ends of the tube using the ratio coefficient between the camera coordinates and the real coordinates determined during the calibration process of the 3D line laser profile sensor to obtain the real coordinates of the two ends of the tube; determining the first coordinate value of the center position of the tube based on the pixel coordinates of the two ends and the value of the face perpendicular to the face of the rectangular cross-section of the cuboid tube captured by the 3D line laser profile sensor; and taking the average of the X values ​​of the real coordinates of the two ends as the second coordinate value of the center position of the tube.

[0075] Furthermore, for precise centering, when the pipe type includes irregular pipes, the calibrated 3D line laser profile sensor is controlled to take pictures of the pipe based on the target pipe centering method to obtain pipe profile images. This includes: the calibrated 3D line laser profile sensor is controlled to take pictures of each face of the pipe based on the target pipe centering method to obtain multiple pipe profile images.

[0076] In addition, machine vision inspection is performed on the pipe contour images to obtain machine vision inspection results, including: performing machine vision inspection on each pipe contour image in multiple pipe contour images to obtain the pixel distance from each face of the pipe to the 3D line laser contour sensor.

[0077] Furthermore, based on the machine vision inspection results, the center position of the pipe is determined, including: using the ratio coefficient between the camera coordinates and the real coordinates determined during the calibration process of the 3D line laser profile sensor to process the pixel distance from each face of the pipe to the 3D line laser profile sensor, so as to obtain the real distance from each face of the pipe to the 3D line laser profile sensor; and determining the center position of the pipe based on the real distance from each face of the pipe to the 3D line laser profile sensor.

[0078] For example, such as Figure 8 As shown, taking a rectangular tube as an example, the distance between the mechanical center and the camera is F_Z in the Z-axis direction and F_X in the X-axis direction. Correspondingly, in the camera coordinate system, the coordinates of the mechanical center are (M_Z, M_X). Also, obtain the relative distance H1 between the current plane and the camera; rotate the chuck holding the tube to 90° and obtain H2 using the same method; rotate the chuck holding the tube to 180° and obtain H3 using the same method; rotate the chuck holding the tube to 270° and obtain H4 using the same method.

[0079] And, given the center coordinates (T_Z, T_X) of the pipe, ; .

[0080] It should be noted here that, although the above is based on Figure 8This method is described using an example, but those skilled in the art should understand that it can also be applied to other non-standard tubes, which will not be described in detail here.

[0081] Therefore, compared with the prior art, the solution of this application has the following advantages: it can identify all tube shapes that can be captured by the 3D line laser contour sensor; it is efficient and has a compact operation, eliminating the need for capacitive tracking by lowering the laser head; in terms of accuracy, the high-precision 3D line laser contour sensor can surpass traditional capacitive centering; the vision kit has a modular structure with a vision guidance module to assist in installation, allowing for quick calibration, assembly, and disassembly; and it is safe, as centering of irregularly shaped tubes does not require physical contact, eliminating the risk of collisions. In order to avoid cutting debris and dust, an overall protective shell is designed.

[0082] It should be understood that the above-described machine vision-based pipe centering method is merely exemplary, and those skilled in the art can make various modifications based on the above method, and the modified solutions also fall within the protection scope of this application.

[0083] like Figure 9 As shown, Figure 9 A flowchart of a laser cutting method provided in an embodiment of this application is shown. Specifically, the laser cutting method includes:

[0084] Step S910: Obtain the coordinates of the mechanical center and the center position of the pipe.

[0085] It should be understood that the coordinates of the mechanical center can be achieved using existing methods.

[0086] For example, the coordinates of the mechanical center can be obtained using the method for determining the position of the mechanical center of the rotating shaft of a metal pipe cutting system disclosed in application number 201710004642.2.

[0087] For example, the center position of the pipe is determined based on the pipe centering method described above, which is based on machine vision.

[0088] Step S920: Based on the coordinates of the machine center and the center position of the pipe, plan the cutting path.

[0089] Specifically, the difference between the center coordinates of the pipe and the center coordinates of the machine rotation can be calculated and transmitted to the CNC system for planning the machining path.

[0090] Step S930: Cut the pipe based on the cutting path.

[0091] Therefore, by utilizing the aforementioned technical solution, machine vision methods based on 3D line laser contour sensors are employed to acquire surface distance information of the metal pipe. Image processing and data calculation are then performed to determine the difference between the pipe's center coordinates and the machine's rotation center coordinates, which is then transmitted to the CNC system for machining path planning. This method is applicable to various common pipe types, including rectangular, square, round, elliptical, racetrack, angle steel, channel steel, D-shaped steel, and H-shaped steel. Furthermore, the system is also suitable for other irregularly shaped pipes (such as...). Figure 2A and attached Figure 2B (As shown), but not limited to these two types of special-shaped tubes.

[0092] It should be understood that the above-described laser cutting method is merely exemplary, and those skilled in the art can make various modifications based on the above method, and the modified solutions also fall within the protection scope of this application.

[0093] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0094] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions.

[0095] It should be noted that any reference numerals placed between parentheses in the claims should not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention can be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In claims that enumerate several means, several of these means may be embodied by the same hardware. The use of the terms first, second, third, etc., is merely for convenience of expression and does not indicate any order. These terms can be understood as part of the component names.

[0096] Furthermore, it should be noted that in the description of this specification, the terms "one embodiment," "some embodiments," "embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0097] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the claims should be interpreted to include both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0098] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, then this invention should also include these modifications and variations.

Claims

1. A pipe centering method based on machine vision, characterized in that, include: With the current pipe clamped under the laser cutting head, the pipe parameters of the current pipe are obtained; wherein, the pipe parameters include the pipe type; Determine the target pipe centering method corresponding to the pipe type; Based on the target pipe centering method, the calibrated 3D line laser contour sensor is controlled to take pictures of the pipe to obtain the pipe contour image. The pipe outline image is subjected to machine vision inspection to obtain machine vision inspection results; Based on the machine vision detection results, the center position of the pipe is determined; The pipe type includes a cylindrical pipe, and the pipe parameters also include the semi-axis of the pipe's cross-section; When the pipe type is the cylindrical pipe, determining the center position of the pipe based on the machine vision detection results includes: Based on the machine vision detection results and the half-axis of the cross-section of the pipe, the center position of the pipe is determined; The step of performing machine vision inspection on the pipe outline image to obtain machine vision inspection results includes: Machine vision inspection is performed on the pipe outline image to obtain the pixel coordinates of the first vertex of the pipe; wherein, the first vertex is the point on the arc surface of the cylindrical pipe that is closest to the camera Z-axis as captured by the 3D line laser profile sensor. And, determining the center position of the pipe based on the machine vision detection results and the half-axis of the pipe's cross-section includes: The pixel coordinates of the first vertex of the pipe are processed using the ratio coefficient between the camera coordinates and the real coordinates determined during the calibration process of the 3D line laser contour sensor, so as to obtain the real coordinates of the first vertex of the pipe. The Z value of the actual coordinates of the first vertex is used as the first coordinate value of the center position of the pipe. The sum of the X value of the true coordinates of the first vertex and the half-axis perpendicular to the Z-axis of the camera is used as the second coordinate value of the center position of the pipe.

2. The pipe centering method according to claim 1, characterized in that, The 3D line laser profile sensor includes a line laser emitter, a first bandpass filter, a second bandpass filter, a reflector, a camera, and a lens; The line laser emitted by the line laser emitter is perpendicular to the first bandpass filter, and the line laser emitted by the line laser emitter can be parallel to the cross-section of the pipe. The second bandpass filter is disposed in the optical path of the reflected laser reflected by the pipe, and the second bandpass filter and the first bandpass filter are not disposed parallel to each other. The reflected laser signal reaches the lens connected to the camera through the reflector.

3. The pipe centering method according to claim 2, characterized in that, The pipe type includes a cuboid pipe, and the pipe parameters also include the length and width of the rectangular cross-section of the cuboid pipe; When the pipe type is the cuboid pipe, determining the center position of the pipe based on the machine vision detection results includes: Based on the machine vision detection results and the length and width of the rectangular cross-section of the cuboid tube, the center position of the tube is determined.

4. The pipe centering method according to claim 3, characterized in that, The step of performing machine vision inspection on the pipe outline image to obtain machine vision inspection results includes: Machine vision inspection is performed on the pipe outline image to obtain the pixel coordinates of the two endpoints of the pipe; wherein the two endpoints are the two endpoints of the face of the cuboid pipe captured by the 3D line laser profile sensor. And, determining the center position of the tube based on the machine vision detection results and the length and width of the rectangular cross-section of the cuboid tube includes: The pixel coordinates of the two ends of the pipe are processed using the ratio coefficient between the camera coordinates and the real coordinates determined during the calibration process of the 3D line laser contour sensor, so as to obtain the real coordinates of the two ends of the pipe. The first coordinate value of the center position of the tube is determined based on the pixel coordinates of the two endpoints and the value of the face perpendicular to the cross-section of the rectangular portion corresponding to the face of the cuboid tube captured by the 3D line laser profile sensor. The average of the X values ​​of the actual coordinates of the two endpoints is taken as the second coordinate value of the center position of the pipe.

5. The pipe centering method according to claim 1, characterized in that, When the pipe type includes irregularly shaped pipes, the step of controlling a calibrated 3D line laser contour sensor to photograph the pipe based on the target pipe centering method to obtain a pipe contour image includes: Based on the target pipe centering method, the calibrated 3D line laser contour sensor is controlled to take pictures of each face of the pipe to obtain multiple pipe contour images. And, the step of performing machine vision inspection on the pipe outline image to obtain machine vision inspection results includes: Machine vision detection is performed on each of the plurality of pipe contour images to obtain the pixel distance from each face of the pipe to the 3D line laser contour sensor; And, determining the center position of the pipe based on the machine vision detection results includes: The pixel distance from each face of the pipe to the 3D line laser profile sensor is processed using the ratio coefficient between the camera coordinates and the real coordinates determined during the calibration process of the 3D line laser profile sensor, so as to obtain the real distance from each face of the pipe to the 3D line laser profile sensor. The center position of the pipe is determined based on the actual distance from each face of the pipe to the 3D line laser profile sensor.

6. The pipe centering method according to claim 2, characterized in that, The calibration process of the 3D line laser contour sensor includes: When the calibration block used in the calibration process is a frustum calibration block and the top surface of the frustum calibration block is parallel to the first bandpass filter, and the line laser emitted by the 3D line laser profile sensor can illuminate the top surface, the camera is controlled to acquire calibration images. Extract the contour lines of the high and low platforms of the frustum calibration block based on the calibration image; The contour lines of the high and low platforms are fitted into two straight line segments using the least squares method, and the angles between the two straight line segments and the X-axis of the image coordinate system are solved. Using the image center of the calibration image as the rotation reference point, rotate the calibration image by the included angle to obtain the rotated calibration image; Determine the coordinate difference between the two straight line segments in the rotated calibration image along the Y-axis of the image coordinate system, and calculate the scaling factor between the camera coordinates and the true coordinates based on the true height difference of the height platform of the frustum calibration block and the coordinate difference between the two straight line segments in the rotated calibration image along the Y-axis of the image coordinate system.

7. A laser cutting method, characterized in that, include: Obtain the coordinates of the mechanical center and the center position of the pipe; wherein the center position of the pipe is determined based on the machine vision-based pipe centering method as described in any one of claims 1 to 6; Based on the coordinates of the machine center and the center position of the pipe, plan the cutting path; The pipe is cut based on the cutting path.

Images