Method for detecting the position and size of a pipe on a laser cutting device and device therefor

Abstract

The present application relates to a kind of for detecting the posture and size of pipe material on laser cutting equipment method, namely its equipment, comprising: control chucking the chuck of pipe material to be cut is rotated multiple times, and multiple photographing is controlled in the process of rotation by pre-calibrated photographing equipment;Multiple images to be processed are superimposed, and complete pipe material cross-sectional profile image is obtained;Pipe material cross-sectional profile image and multiple types of pipe material cross-sectional profile image template are compared to determine target pipe material type information;Call the posture size identification algorithm corresponding to target pipe material type information to identify the posture size of pipe material to be cut;Including target pipe material type information and target size information the information to be matched is matched with drawing information, if it is determined by matching that target pipe material type information and pipe material type information in drawing information are same and target size information and size information in drawing information are same, then output target posture information, to reduce the cost.

Description

Technical Field

[0001] This invention relates to the field of laser cutting technology, and in particular to a method and apparatus for detecting the posture and dimensions of pipes on laser cutting equipment. Background Technology

[0002] Laser cutting equipment faces two main problems when cutting pipes: first, the uncertainty of the phase angle of the pipe to be processed in the rotation direction; and second, the mismatch between the cross-sectional dimensions of the pipe to be processed and the drawings. These problems can lead to cutting errors, waste of pipe resources, and even damage to the laser equipment, resulting in economic and property losses.

[0003] Currently available pipe size inspection equipment typically requires a specially designed structure to surround the pipe and inspect it from multiple angles to obtain the pipe's orientation and size information. This structural design does not reserve a specific installation area on the laser cutting equipment, and its assembly and implementation will bring a relatively high cost burden.

[0004] Therefore, how to provide a convenient, efficient, and stable method for recognizing the posture and size of pipes during laser cutting has become a problem that needs to be solved. 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 method and device for detecting the posture and size of pipes on a laser cutting equipment, so as to reduce costs while efficiently and stably identifying the posture and size of pipes.

[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 method for detecting the posture and size of a pipe on a laser cutting device, comprising: upon receiving drawing information of the pipe to be detected, controlling a chuck clamping the pipe to be cut to rotate multiple times, and controlling a pre-calibrated imaging device to take multiple pictures during the rotation process to obtain multiple images to be processed; superimposing the multiple images to be processed to obtain a complete pipe cross-sectional contour image; comparing the pipe cross-sectional contour image with multiple types of pipe cross-sectional contour image templates to determine the target pipe type information of the pipe to be cut; calling a posture and size recognition algorithm corresponding to the target pipe type information to identify the posture and size of the pipe to be cut, obtaining the target size information and target posture information of the pipe to be cut; matching the matching information including the target pipe type information and target size information with the drawing information; if the matching determines that the pipe type information and the drawing information are the same and the target size information and the drawing information are the same, then outputting the target posture information.

[0010] In one possible embodiment, the imaging device includes a line laser emitter and a camera located on the same side of the pipe to be cut, wherein the line laser emitted by the line laser emitter is parallel to the cross-section of the pipe to be cut, and the line laser emitted by the line laser emitter is within the shooting range of the camera.

[0011] In one possible embodiment, multiple images to be processed are superimposed to obtain a complete pipe cross-sectional contour image, including: transforming each of the multiple images to be processed based on a preset pixel precision of the transformed images to obtain multiple transformed images; wherein each of the multiple transformed images corresponds to one image to be processed, and each transformed image is obtained after eliminating pixel precision errors in the X and Y directions; rotating the remaining images of the multiple images to be processed, except for the first image to be processed, by corresponding angles based on multiple rotation angles, and calculating the homogeneous transformation matrix corresponding to the image after each rotation; calculating the matrix product of the coordinate matrix of each pixel of each transformed image and its corresponding homogeneous transformation matrix to obtain multiple images to be superimposed; and superimposing the multiple images to be superimposed using the maximum value method to obtain the pipe cross-sectional contour image.

[0012] In one possible embodiment, the pipe cross-sectional profile image is compared with multiple types of pipe cross-sectional profile image templates to determine the target pipe type information of the pipe to be cut, including: calculating the Euclidean distance between the pipe cross-sectional profile image and each pipe cross-sectional profile image template; sorting all Euclidean distances in descending order to obtain the sorting results of all Euclidean distances; and counting the pipe type information corresponding to the most frequent Euclidean distances in the sorting results as the target pipe type information.

[0013] In one possible embodiment, an attitude dimension recognition algorithm corresponding to the target pipe type information is invoked to identify the attitude dimension of the pipe to be cut, thereby obtaining the target size information and target attitude information of the pipe to be cut. This includes: when the target pipe type information includes rectangular pipe, racetrack pipe, angle steel and channel steel, calculating the distance between the straight part of the pipe cross-sectional contour image and the center of the pipe according to a preset pixel precision.

[0014] In one possible embodiment, an attitude dimension recognition algorithm corresponding to the target pipe type information is invoked to identify the attitude dimension of the pipe to be cut, thereby obtaining the target size information and target attitude information of the pipe to be cut. This includes: when the target pipe type information includes round pipe, racetrack pipe, and elliptical pipe, converting the arc portion of the pipe cross-sectional contour image into a set of discrete points; determining the discrete points in the set with a slope close to zero as vertices; and calculating the distance between the vertices and the center of the pipe according to a preset pixel precision.

[0015] In one possible embodiment, an attitude dimension recognition algorithm corresponding to the target pipe type information is invoked to identify the attitude dimension of the pipe to be cut, thereby obtaining the target size information and target attitude information of the pipe to be cut. This includes: when the target pipe type information includes angle steel and channel steel, acquiring and sorting the Y coordinate data of the pipe edge pixels in the pipe cross-sectional contour image to obtain a coordinate sorting result that satisfies a normal distribution; dividing all Y coordinate data in the coordinate sorting result into quartiles, removing outliers in all Y coordinate data based on the interquartile range, and determining the target attitude information based on the maximum value of the remaining data after removing outliers in all Y coordinate data.

[0016] In one possible embodiment, an attitude dimension recognition algorithm corresponding to the target pipe type information is invoked to identify the attitude dimensions of the pipe to be cut, obtaining the target dimension information and target attitude information of the pipe to be cut. This includes: when the target pipe type information includes rectangular pipe, racetrack circle, ellipse, and I-beam, and the target dimension information includes target width and target height, obtaining the size relationship between the target width and target height; when the size relationship between the target width and target height is the same as the size relationship between the width and height in the drawing information, the pipe angle is 0°; when the size relationship between the target width and target height is opposite to the size relationship between the width and height in the drawing information, the pipe angle is 90°.

[0017] In one possible embodiment, an attitude dimension recognition algorithm corresponding to the target pipe type information is invoked to identify the attitude dimension of the pipe to be cut, thereby obtaining the target size information and target attitude information of the pipe to be cut. This includes: if the target pipe type information includes angle steel and channel steel, generating multiple line segments based on the image size of the pipe cross-section contour image and the center coordinates of the pipe to be cut; and determining the pipe angle based on the intersection of the multiple line segments and the cross-section of the pipe to be cut.

[0018] In a second aspect, embodiments of this application provide a laser cutting device, characterized in that it includes a controller, the controller being used to execute the method for detecting the posture and size of a pipe on a laser cutting device as described in the first aspect.

[0019] (III) Beneficial Effects

[0020] The beneficial effects of this invention are:

[0021] This application provides a method and apparatus for detecting the posture and dimensions of pipes on a laser cutting machine. Upon receiving the drawing information of the pipe to be detected, the method controls the chuck clamping the pipe to be cut to rotate multiple times, and during the rotation, controls a pre-calibrated imaging device to take multiple pictures, obtaining multiple images to be processed. These multiple images are then superimposed to obtain a complete cross-sectional contour image of the pipe. The cross-sectional contour image is compared with multiple types of pipe cross-sectional contour image templates to determine the target pipe type information. A posture and dimension recognition algorithm corresponding to the target pipe type information is invoked to identify the posture and dimensions of the pipe to be cut, obtaining the target dimension information and target posture information of the pipe. Finally, the matching information, including the target pipe type information and target dimension information, is matched with the drawing information. If the matching determines that the target pipe type information and the pipe type information in the drawing information are the same, and the target dimension information and the dimension information in the drawing information are the same, then the target posture information is output, thereby ensuring the safety of the processing and avoiding potential economic losses.

[0022] 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

[0023] 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.

[0024] Figure 1 A flowchart illustrating a method for detecting the orientation and dimensions of a pipe on a laser cutting device, as provided in an embodiment of this application, is shown.

[0025] Figure 2 The present application provides a schematic diagram of the hardware structure of a photographing device according to an embodiment of the present application;

[0026] Figure 3 The present application provides a schematic diagram of the structure of an angle iron and a camera according to an embodiment. Detailed Implementation

[0027] 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.

[0028] To address the uncertainties in the phase angle of the pipe to be processed in the rotation direction and the mismatch between the cross-sectional dimensions of the pipe and the drawing in existing technologies, this application provides a method and apparatus for detecting the posture and dimensions of a pipe on a laser cutting device. Upon receiving the drawing information of the pipe to be detected, the method controls the chuck clamping the pipe to be cut to rotate multiple times, and during the rotation, controls a pre-calibrated imaging device to take multiple pictures, obtaining multiple images to be processed. These multiple images are then superimposed to obtain a complete cross-sectional contour image of the pipe. Finally, the method combines the cross-sectional contour image of the pipe with multiple... The system compares the cross-sectional contour image template of the pipe with the target pipe type to determine the target pipe type information. It also calls the posture dimension recognition algorithm corresponding to the target pipe type information to identify the posture dimension of the pipe to be cut, thereby obtaining the target dimension information and target posture information of the pipe to be cut. The matching information, including the target pipe type information and the target dimension information, is matched with the drawing information. If the matching determines that the pipe type information and the drawing information are the same, and the target dimension information and the drawing information are the same, the target posture information is output, thus ensuring the safety of processing and avoiding possible economic losses.

[0029] 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.

[0030] First Embodiment

[0031] Please see Figure 1, Figure 1 A flowchart illustrating a method for detecting the orientation and dimensions of a tube on a laser cutting device, according to an embodiment of this application, is shown. Specifically, the method can be executed by a controller in the laser cutting device, and the method includes:

[0032] Step S110: Upon receiving the drawing information of the pipe to be inspected, control the chuck clamping the pipe to be cut to rotate multiple times, and control the pre-calibrated imaging device to take multiple pictures during the rotation process to obtain multiple images to be processed.

[0033] It should be understood that the specific devices included in the photographing equipment can be configured according to actual needs, and the embodiments of this application are not limited thereto.

[0034] Optionally, such as Figure 2 As shown, the imaging device includes a line laser emitter and a camera located on the same side of the pipe to be cut. The line laser emitter and camera are connected by a fixed structure to ensure that their relative positions remain unchanged. The line laser emitted by the line laser emitter is parallel to the cross-section of the pipe to be cut, and the line laser emitted by the line laser emitter is within the camera's field of view. That is, the line laser emitted by the line laser emitter is perpendicular to the surface of the pipe to highlight the surface; the camera provides image data for this method to identify the pipe's orientation and dimensions.

[0035] It should also be understood that the specific steps of step S110 can be set according to actual needs, and the embodiments of this application are not limited thereto.

[0036] Specifically, the controller receives the drawing information of the pipe to be cut from the CNC system and the interface parameters input by the user. The interface parameters are the maximum allowable dimensional error parameters of the pipe set by the user based on the process parameters of the pipe to be cut. Among them, the interface parameters may include the camera's exposure parameters, camera gain parameters, the maximum allowable dimensional error for pipe size recognition, and the vertical distance between the camera and the center of the pipe clamping device; the calibration parameters may include the camera's intrinsic parameters and distortion coefficients, and the X-axis pixel accuracy and Y-axis pixel coordinates at different distances of the camera.

[0037] Furthermore, the interface parameters in this embodiment are input via the laser cutting equipment and transmitted to the controller via a configuration file. Additionally, the drawing information of the pipe to be cut mainly includes the type and size information of the pipe, which is transmitted to this embodiment by saving it to a local engineering information file.

[0038] Furthermore, after the controller receives the start command output by the user, it begins to take the first picture, at which point the pipe angle is 0°; then the laser system rotates the pipe to 90°, and in this embodiment, it takes the second picture; then the laser system rotates the pipe to 180°, and in this embodiment, it takes the third picture; then the laser system rotates the pipe to 270°, and in this embodiment, it takes the fourth picture, and then the laser system rotates the pipe back to the initial 0°.

[0039] It should be noted that although the above description uses four rotations as an example, those skilled in the art should understand that more or fewer rotations are possible, and the embodiments of this application are not limited to this.

[0040] Step S120: Superimpose multiple images to be processed to obtain a complete cross-sectional profile image of the pipe.

[0041] Specifically, based on the preset pixel precision of the transformed image, each of the multiple images to be processed is transformed to obtain multiple transformed images. Each transformed image corresponds to one image to be processed, and each transformed image is obtained after eliminating pixel precision errors in the X and Y directions. Based on multiple rotation angles, the remaining images (excluding the first image to be processed) are rotated by corresponding angles, and the homogeneous transformation matrix corresponding to each rotation is calculated. The matrix product of the coordinate matrix of each pixel in each transformed image and its corresponding homogeneous transformation matrix is ​​calculated to obtain multiple images to be superimposed, where each transformed image corresponds to a homogeneous transformation matrix. The multiple images to be superimposed are then superimposed using the maximum value method to obtain a pipe cross-sectional contour image.

[0042] For example, since the pixel precision ac_x in the X direction and the pixel precision ac_y in the Y direction of the image to be processed are different, direct rotation and superposition will produce errors. Therefore, the pixels in each image to be processed can be transformed using the following formula:

[0043]

[0044] In the formula, x represents the x-coordinate of the transformed pixel; x_c is the x-coordinate of the center of the image to be processed; (x_o, y_o) are the coordinates of the pixel to be transformed in the image to be processed; since each pixel in the image to be processed has a corresponding pixel precision in the X direction and pixel precision in the Y direction, fx(y_o) is the pixel precision ac_x in the X direction of the pixel to be transformed; AC represents the preset pixel precision of the transformed image; y represents the y-coordinate of the transformed pixel; y_c is the y-coordinate of the center of the image to be processed; fy(y_o) is the formula for converting the ordinate y_o of the pixel coordinates in the image to the camera physical coordinate system coordinates.

[0045] Furthermore, given that the center of the pipe in the image has coordinates (x_c, y_c) and a rotation angle of θ, the homogeneous transformation matrix of the rotation transformation is:

[0046]

[0047] Therefore, the second image to be processed, taken at a 90° rotation, can be rotated 90° around the center of the pipe, resulting in the homogeneous transformation matrix:

[0048]

[0049] Furthermore, the third image to be processed, taken at a 180° rotation, can be rotated 180° around the center of the pipe, resulting in the homogeneous transformation matrix:

[0050]

[0051] Furthermore, the fourth image to be processed, taken at a 270° rotation, can be rotated 270° around the center of the pipe. The resulting homogeneous transformation matrix is:

[0052]

[0053] Furthermore, all four transformed images can be grayscale images, with each pixel value ranging from [0.255]. Then, the product of the coordinate matrix (containing two coordinate data points) of each pixel in the first transformed image and the homogeneous transformation matrix at 0° can be calculated; the value remains unchanged, and the resulting image is still a single image to be superimposed. Subsequently, the product of the coordinate matrix of each pixel in the second transformed image and the homogeneous transformation matrix at 90° can be calculated, again resulting in a single image to be superimposed. The same process is repeated for the third transformed image, and the same process is repeated for the fourth transformed image. Finally, the four images are superimposed using the maximum value method to obtain a complete cross-sectional profile image of the pipe. For example, the maximum value of the pixels at the same position in the four images to be superimposed can be selected as the pixel value of the corresponding pixel in the cross-sectional profile image of the pipe.

[0054] It should be noted that the cross-sectional profile image of the pipe only includes the outer ring of the pipe and does not show the thickness of the pipe.

[0055] Step S130: Compare the pipe cross-sectional profile image with multiple types of pipe cross-sectional profile image templates to determine the target pipe type information of the pipe to be cut.

[0056] Specifically, the Euclidean distance between the pipe cross-sectional profile image and each pipe cross-sectional profile image template is calculated; all Euclidean distances are sorted in descending order to obtain the sorting results of all Euclidean distances; the pipe type information corresponding to the most frequent Euclidean distances in the sorting results is counted as the target pipe type information.

[0057] For example, the cross-sectional contour image of the pipe is compared with the collected cross-sectional contour image templates of various types of pipes. The Euclidean distance between the image to be detected and each template image is calculated. Based on the Euclidean distance of each type of pipe, the pipe type with the largest number of the n closest pipes is selected as the basis for the subsequent pipe posture and size algorithm. The specific implementation method is as follows:

[0058] When performing image comparison, the same image size is required. The saved templates for the cross-sectional contour images of various types of pipes all have a width and height of S. Furthermore, the bounding rectangle R of the pipe cross-sectional contour image is calculated, with vertices LT(x0,y0) and RB(x1,y1), and the pipe center's coordinates in the image are (x_c, y_c). The bounding rectangle R is then expanded to a square, with the center of the square coinciding with the pipe center. The vertices of the expanded rectangle R are:

[0059] LT:

[0060]

[0061] x_0=x0-max(x1+x0-2x_c, 0);

[0062] y_0=y0-max(y1+y0-2y_c,0);

[0063] In the formula for calculating LT, x_0 and y_0 represent intermediate parameters.

[0064] RB:

[0065]

[0066] x_0=x0+max(2x_c-x1-x0,0);

[0067] y_0=y0+max(2y_c-y1-y0,0);

[0068] In the formula for calculating RB, x_0 and y_0 also represent intermediate parameters.

[0069] Subsequently, the cross-sectional profile image of the pipe can be adjusted to an image of size S based on the vertices of the expanded rectangle R.

[0070] Subsequently, the Euclidean distance between the resized image and each pipe cross-sectional contour image template is calculated. All Euclidean distances are then sorted in descending order to obtain the sorted results. The pipe type corresponding to each of the top n sorted Euclidean distances is selected, and the pipe type with the most occurrences is chosen as the target pipe type information. Here, n is a positive integer.

[0071] Step S140: Call the attitude dimension recognition algorithm corresponding to the target pipe type information to identify the attitude dimension of the pipe to be cut, and obtain the target dimension information and target attitude information of the pipe to be cut.

[0072] It should be understood that the specific process of calling the posture and size recognition algorithm corresponding to the target pipe type information to identify the size of the pipe to be cut can be set according to actual needs, and the embodiments of this application are not limited thereto.

[0073] Optionally, if the target pipe type information includes rectangular pipe, racetrack pipe, angle steel and channel steel, the distance between the straight part of the pipe cross-sectional contour image and the center of the pipe is calculated according to the preset pixel accuracy.

[0074] For example, for the plane of rectangular tubes, runway tubes, and the outer surface of angle steel or channel steel, the image is displayed as a straight line. The distance between the average Y-coordinate of the straight portion of the tube's cross-sectional profile image and the center of the tube can be calculated.

[0075]

[0076] In the formula, L is the distance between the average Y-coordinate of the straight line portion of the pipe cross-section contour image and the center of the pipe; col is the horizontal resolution of the image, i.e., the width of the image; row is the vertical resolution of the image, i.e., the height of the image; f(i, j) indicates whether the pixel value at coordinate (i, j) is 0. If the pixel value at coordinate (i, j) is 0, then f(i, j) equals 0, indicating that there is no pipe at this location; if the pixel value at coordinate (i, j) is 1, then f(i, j) equals 1, indicating that there is a pipe at this location, and the Y-coordinate at this location is calculated; C row The number of pixels in the vertical pixel count that include the pipe section; To obtain the average Y-coordinate of the pipe location in the i-th column of the image; C col This represents the number of pixels containing pipe sections in the horizontal statistics. To obtain the average Y-coordinate of the location of the pipe cross-section in the image.

[0077] In other words, firstly, for each column in the cross-sectional contour image of the pipe, the average of the x-coordinates of all pixels with non-zero pixel values ​​is obtained. This average is used as the unique x-coordinate of this column. Then, the average of the unique x-coordinates of all columns in the image is calculated. Since this part of the contour image is a straight line, this average is the pixel value coordinate of this straight line. The pixel value is converted into a physical value, which is the distance from the outer surface of the pipe to the camera.

[0078] Optionally, if the target pipe type information includes round pipe, racetrack pipe, and elliptical pipe, the arc portion of the pipe cross-sectional contour image is converted into a set of discrete points; the discrete points in the set with a slope close to zero are identified as vertices; and the distance between the vertices and the center of the pipe is calculated according to a preset pixel precision.

[0079] For example, for the curved surfaces of circular pipes, racetrack pipes, elliptical pipes, etc., the image is displayed as an arc. Specifically, by solving for the distance between the Y-coordinate of the arc's vertex and the center of the pipe:

[0080] First, convert the curved portion of the image into a set of discrete points:

[0081] [(x0, y0), (x1, y1), ..., (x k-1 y k-1 ), (x k y k ), (x k+1 y k+1 ), ..., (x n y n )];

[0082] Furthermore, the slope m at a point on the arc can be obtained through numerical differentiation. k m k The larger the absolute value of m, the steeper the curve. k The smaller the absolute value of m, the flatter the curve, and the greater the slope m. k The calculation formula is:

[0083]

[0084] The Y-coordinate of a point to the left of the vertex of the arc increases, while the Y-coordinate of a point to the right of the vertex decreases. Therefore, the Y-coordinate of the vertex approaches 0 infinitely, and the slope is m. k The position approaching 0 is the vertex. The distance from the pipe cross-section to the pipe center can be calculated based on pixel precision and the pipe center coordinates. Specifically:

[0085]

[0086] In the formula, L is the distance between the vertex and the center of the pipe; y_k is the y-coordinate of the point where the slope approaches 0; y_c is the y-coordinate of the center of the pipe; and AC is the preset pixel precision.

[0087] Optionally, if the target pipe type information includes angle steel and channel steel, the Y coordinate data of the pipe edge pixels in the pipe cross-sectional contour image are obtained and sorted to obtain a coordinate sorting result that satisfies a normal distribution; all Y coordinate data in the coordinate sorting result are divided into quartiles, and outliers in all Y coordinate data are removed based on the interquartile range of the quartiles; and the target attitude information is determined based on the maximum value of the remaining data after removing outliers in all Y coordinate data.

[0088] For example, for the inner surfaces of open-shaped pipes such as angle steel and channel steel, the images of these pipes are irregular, and the only usable information is the specific edge pixels of the pipe. The Y coordinates of these pixels are obtained and sorted, and these data should satisfy a normal distribution. Here, specific edge pixels of the pipe refer to pixels on the surface facing the camera that are not part of the distance from the outer surface of the pipe to the camera. In practice, the line laser emitter and the camera can be considered as a whole; when the pipe is facing the line laser emitter, it can be considered facing the camera. See [link to relevant documentation] for details. Figure 3 As shown.

[0089] Furthermore, all Y-coordinate data in the coordinate sorting results are divided into quartiles, Q1, Q2, and Q3, with an interquartile range IQR = Q3 - Q1. Outliers (less than Q1 - 1.5 * IQR or greater than Q3 + 1.5 * IQR) are removed, corresponding to points outside the normal distribution with a σ value of 2.698. The maximum value among the remaining points is the position coordinate of the pipe, and the distance from the outer surface of the pipe to the camera can be calculated using the following formula:

[0090]

[0091] In the formula, L is the distance from the outer surface of the pipe to the camera; y_max is the maximum value of the remaining Y coordinates after removing outliers from all the Y coordinates mentioned above; y_c is the pixel coordinate of the center of the pipe; AC is the preset pixel precision.

[0092] Therefore, the distances L1, L2, L3 and L4 from the pipe surface to the pipe center can be obtained from the four images to be processed. The width and height of the pipe are the sum of the distances from the pipe surface to the pipe center in the two images on the opposite side of the pipe, W = L1 + L3 and H = L2 + L4.

[0093] It should be noted that the stitched contour image is used to detect the pipe type; the size and orientation recognition use the transformed image before stitching.

[0094] It should also be understood that the specific process of calling the posture size recognition algorithm corresponding to the target pipe type information to identify the posture size of the pipe to be cut and obtain the target posture information of the pipe to be cut can be set according to actual needs, and the embodiments of this application are not limited thereto.

[0095] Optionally, based on the pipe type and pipe size information obtained in the above steps, different algorithms are used to calculate the pipe angle. The pipe angle can be divided into four types: 0°, 90°, 180°, and 270°. Among them, round pipes and square pipes only have 0° and do not need to identify the posture. Rectangular pipes, racetrack circles, ellipses, I-beams, etc. only have 0° and 90°. Angle steel and channel steel have four angles.

[0096] For tubes such as rectangular tubes, track circles, ellipses, and I-beams that have only two angles, the angle is 0° when the width and height are the same as those in the drawing, and 90° when the width and height are opposite to those in the drawing.

[0097] In other words, when the target pipe type information includes rectangular pipe, racetrack circle, ellipse, and I-beam, and the target size information includes target width and target height, the size relationship between the target width and target height is obtained; when the size relationship between the target width and target height is the same as the size relationship between the width and height in the drawing information, the pipe angle is 0°; when the size relationship between the target width and target height is opposite to the size relationship between the width and height in the drawing information, the pipe angle is 90°.

[0098] For pipes with four angles, such as angle steel and channel steel, the above steps obtain a superimposed cross-sectional image I of the pipe, with image size S and pipe center coordinates (x_c, y_c), generating the following line segments:

[0099] [(0,y_c),(x_c-10,y_c)];

[0100] [(x_c+10,y_c),(S x ,y_c)];

[0101] [(x_c, 0), (x_c, y_c-10)];

[0102] [(x_c, y_c+10), (x_c, S) y )];

[0103] In the formula, S x S is the image width; y This represents the image height.

[0104] Subsequently, the intersection of the four lines (C1, C2, C3, and C4) with the cross-section of the pipe can be calculated. If an intersection exists, it is recorded as 1; otherwise, it is recorded as 0. Specifically:

[0105] Table 1: Relationship between Angle Steel Angles and Intersections

[0106] Angle\Intersection <![CDATA[C1]]> <![CDATA[C2]]> <![CDATA[C3]]> <![CDATA[C4]]> 0° 1 0 0 1 90° 1 1 0 0 180° 0 1 1 0 270° 0 0 1 1

[0107] Table 2: Relationship between the angle and intersection of channel steel

[0108] Angle\Intersection <![CDATA[C1]]> <![CDATA[C2]]> <![CDATA[C3]]> <![CDATA[C4]]> 0° 1 0 1 1 90° 0 1 1 1 180° 1 1 1 0 270° 1 1 0 1

[0109] Furthermore, for I-beams with equal width and height, the angle of the pipe cannot be determined by the relationship between the width and height. Instead, it can be calculated by the intersection of line segments. When C1 = C3 = 1 and C2 = C4 = 0, the angle of the I-beam is 90°, and when C1 = C3 = 0 and C2 = C4 = 1, the angle of the I-beam is 0°.

[0110] In other words, when the target pipe type information includes angle steel and channel steel, multiple line segments are generated based on the image size of the pipe cross-section contour image and the center coordinates of the pipe to be cut; the pipe angle is determined based on the intersection of the multiple line segments and the cross-section of the pipe to be cut.

[0111] Step S150: Match the information to be matched, including the target pipe type information and the target size information, with the drawing information. If the matching determines that the target pipe type information and the pipe type information in the drawing information are the same, and the target size information and the size information in the drawing information are the same, then output the target attitude information.

[0112] Additionally, if the target pipe type information and the pipe type information in the drawing information are different, or the target size information and the size information in the drawing information are different, an alarm signal will be output.

[0113] For example, the identification results are compared with the pipe type, pipe width, and pipe height information in the drawing. If the pipe type does not match or the pipe size error exceeds the user-set interface parameters, the pipe to be cut does not match the drawing, and feedback is sent to the laser cutting system to terminate the cutting. Otherwise, the pipe posture recognition result is output to the laser cutting system.

[0114] Therefore, by means of the above technical solution, the method of this application solves the problems of uncertainty of the phase angle of the pipe to be processed in the rotation direction and mismatch between the cross-sectional dimensions of the pipe to be processed and the drawing during the cutting process, thereby ensuring the safety of processing and avoiding possible economic losses.

[0115] Furthermore, the controller implementing this method is integrated into the laser cutting system, which enables the laser cutting system to operate more stably and efficiently, preventing processing errors or equipment damage caused by incorrect dimensions or orientation of the tube to be processed.

[0116] Furthermore, this application can accurately identify and measure the geometric features of various pipe materials, including but not limited to closed-type pipe materials (such as round pipes, square pipes, etc.) and open-type pipe materials (such as channel steel, angle steel, etc.). At the same time, for new pipe material types, they can be quickly adapted by adding templates.

[0117] It should be understood that the above-described method for detecting the posture and size of pipes on laser cutting equipment is merely exemplary. Those skilled in the art can make various modifications based on the above method, and the modified or altered content is also within the scope of protection of this application.

[0118] Second Embodiment

[0119] This application provides a laser cutting device, including a controller, which is used to perform the method for detecting the posture and size of a pipe on a laser cutting device as described in the first embodiment.

[0120] 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.

[0121] 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.

[0122] 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.

[0123] 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.

[0124] 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.

[0125] 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 method for detecting the posture and dimensions of pipes on a laser cutting device, characterized in that, include: Upon receiving the drawing information of the pipe to be inspected, the chuck clamping the pipe to be cut is controlled to rotate multiple times, and during the rotation, the pre-calibrated imaging device is controlled to take multiple pictures to obtain multiple images to be processed. The multiple images to be processed are superimposed to obtain a complete cross-sectional contour image of the pipe. The cross-sectional profile image of the pipe is compared with multiple types of cross-sectional profile image templates of pipes to determine the target pipe type information of the pipe to be cut; The attitude dimension recognition algorithm corresponding to the target pipe type information is invoked to identify the attitude dimension of the pipe to be cut, thereby obtaining the target size information and target attitude information of the pipe to be cut; The matching information, including the target pipe type information and the target size information, is matched with the drawing information. If the matching determines that the target pipe type information and the pipe type information in the drawing information are the same, and the target size information and the size information in the drawing information are the same, then the target posture information is output. The process of overlaying the multiple images to be processed to obtain a complete cross-sectional contour image of the pipe includes: Based on the preset pixel precision of the transformed image, each of the multiple images to be processed is transformed to obtain multiple transformed images; wherein, each of the multiple transformed images corresponds to one image to be processed, and each transformed image is obtained after eliminating the pixel precision errors in the X and Y directions; Based on the rotation angles of multiple rotations, the remaining images of the multiple images to be processed, excluding the first image to be processed, are rotated by the corresponding angles, and the homogeneous transformation matrix corresponding to the image after each rotation is calculated. Calculate the matrix product of the coordinate matrix of each pixel in each transformed image and its corresponding homogeneous transformation matrix to obtain multiple images to be superimposed; Multiple images to be superimposed are superimposed using the maximum value method to obtain the cross-sectional contour image of the pipe.

2. The method according to claim 1, characterized in that, The photographing device includes a line laser emitter and a camera located on the same side of the pipe to be cut. The line laser emitted by the line laser emitter is parallel to the cross-section of the pipe to be cut, and the line laser emitted by the line laser emitter is within the shooting range of the camera.

3. The method according to claim 1, characterized in that, The step of comparing the pipe cross-sectional profile image with multiple types of pipe cross-sectional profile image templates to determine the target pipe type information of the pipe to be cut includes: Calculate the Euclidean distance between the pipe cross-sectional profile image and each of the pipe cross-sectional profile image templates; Sort all the Euclidean distances in descending order to obtain the sorted results of all the Euclidean distances; The pipe type information with the most Euclidean distances among all the sorted results is taken as the target pipe type information.

4. The method according to claim 1, characterized in that, The step of calling the attitude dimension recognition algorithm corresponding to the target pipe type information to identify the attitude dimension of the pipe to be cut, and obtaining the target dimension information and target attitude information of the pipe to be cut, includes: When the target pipe type information includes rectangular pipe, racetrack pipe, angle steel and channel steel, the distance between the straight part of the pipe cross-sectional contour image and the center of the pipe is calculated according to the preset pixel accuracy.

5. The method according to claim 4, characterized in that, The step of calling the attitude dimension recognition algorithm corresponding to the target pipe type information to identify the attitude dimension of the pipe to be cut, and obtaining the target dimension information and target attitude information of the pipe to be cut, includes: When the target pipe type information includes round pipe, racetrack pipe and elliptical pipe, the arc portion of the pipe cross-sectional contour image is converted into a discrete point set; The discrete points in the set of discrete points whose slopes approach zero are identified as vertices; The distance between the vertex and the center of the pipe is calculated based on the preset pixel precision.

6. The method according to claim 1, characterized in that, The step of calling the attitude dimension recognition algorithm corresponding to the target pipe type information to identify the attitude dimension of the pipe to be cut, and obtaining the target dimension information and target attitude information of the pipe to be cut, includes: When the target pipe type information includes angle steel and channel steel, the Y coordinate data of the pipe edge pixels in the pipe cross-sectional contour image are obtained and sorted to obtain a coordinate sorting result that satisfies a normal distribution. All Y-coordinate data in the coordinate sorting result are divided into quartiles, and outliers in all Y-coordinate data are removed based on the interquartile range of the quartiles. The target pose information is determined based on the maximum value of the remaining data after removing the outliers in all Y-coordinate data.

7. The method according to claim 1, characterized in that, The step of calling the attitude dimension recognition algorithm corresponding to the target pipe type information to identify the attitude dimension of the pipe to be cut, and obtaining the target dimension information and target attitude information of the pipe to be cut, includes: When the target pipe type information includes rectangular pipe, racetrack circle, ellipse and I-beam and the target size information includes target width and target height, obtain the size relationship between the target width and target height; When the relationship between the target width and target height is the same as the relationship between the width and height in the drawing information, the pipe angle is 0°. When the relationship between the target width and target height is opposite to the relationship between the width and height in the drawing information, the pipe angle is 90°.

8. The method according to claim 1, characterized in that, The step of calling the attitude dimension recognition algorithm corresponding to the target pipe type information to identify the attitude dimension of the pipe to be cut, and obtaining the target dimension information and target attitude information of the pipe to be cut, includes: When the target pipe type information includes angle steel and channel steel, multiple line segments are generated based on the image size of the pipe cross-sectional contour image and the center coordinates of the pipe to be cut; The pipe angle is determined based on the intersection of the multiple line segments and the cross-section of the pipe to be cut.

9. A laser cutting device, characterized in that, Includes a controller for performing the method for detecting the posture and dimensions of a pipe on a laser cutting device as described in any one of claims 1 to 8.

Images