Method and device for detecting wall thickness of transparent pipe

Abstract

The invention relates to the technical field of transparent pipe detection, and provides a transparent pipe wall thickness detection method and device, and the method comprises the steps: obtaining original images of a pipe in a first direction and a second direction; processing the original image to obtain a feature enhanced image; extracting edge points of the pipe in the image, and performing fitting to obtain contours of the inner wall and the outer wall of the pipe; on the basis of the outer wall coordinate points, corresponding inner wall coordinate points are obtained, and the wall thickness between the two coordinate points is calculated; and repeatedly calculating the wall thicknesses of multiple points, carrying out statistical analysis, and outputting an average value, a maximum value, a minimum value and a standard deviation. According to the invention, the original images are respectively collected in two directions, and the wall thickness detection is realized based on the original images, so that on one hand, the accuracy of wall thickness detection can be improved based on the collection of two point positions, and on the other hand, the overall image of the pipe can be obtained when the pipe moves by a small distance or rotates by a small angle. And the wall thickness detection efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of transparent pipe detection, and in particular to a transparent pipe wall thickness detection method and device. BACKGROUND

[0002] The wall thickness of a transparent circular hollow pipe has a key influence on its structural strength, weight and functional performance. In actual production, the wall thickness of a transparent pipe needs to be detected to meet design requirements.

[0003] In related technologies, due to the characteristics of the transparent pipe, the accuracy of conventional wall thickness detection systems and methods is low, and it is difficult to achieve high accuracy and high efficiency detection. For example, ultrasonic detection, laser ranging detection and other methods have low precision and limited efficiency improvement. Therefore, a detection method and device that can accurately and efficiently detect are urgently needed. SUMMARY

[0004] The present application provides a transparent pipe wall thickness detection method and device to solve the defects of low detection accuracy and efficiency of transparent pipe wall thickness in the prior art.

[0005] The first aspect of the present application provides a transparent pipe wall thickness detection method, comprising the following steps: Obtaining original images of the pipe in a first direction and a second direction; Processing the original images to obtain a processed feature-enhanced image; Extracting edge points of the pipe in the feature-enhanced image and performing linear fitting to obtain the profiles of the inner wall and the outer wall of the pipe; Based on the outer wall coordinate point of any outer wall point, obtaining the inner wall coordinate point of the nearest inner wall point, and calculating the Euclidean distance between the two coordinate points as the wall thickness of the point; Repeating the calculation of the wall thickness of multiple points, and statistically analyzing the wall thickness values of the multiple points to output the average value, the maximum value, the minimum value and the standard deviation to obtain the measured wall thickness of the pipe.

[0006] According to the transparent pipe wall thickness detection method provided by the present application, the original images of the pipe in the first direction and the second direction are obtained, comprising the following steps: A first acquisition camera is arranged in the first direction, and a second acquisition camera is arranged in the second direction. The pipe is controlled to rotate or move, and the first acquisition camera and the second acquisition camera are controlled to continuously capture the original images during rotation or movement.

[0007] According to the transparent pipe wall thickness detection method provided by the present application, the original images are processed to obtain a processed feature-enhanced image, which specifically comprises the following steps: preprocessing the original image to obtain a preprocessed image; enhancing image features of the preprocessed image to obtain the feature-enhanced image.

[0008] The transparent pipe wall thickness detection method provided by the present application includes the following steps of preprocessing the original image to obtain a preprocessed image: performing gray scale conversion on the original image to reduce the number of color channels; performing noise reduction and smoothing processing on the gray scale converted image, and performing edge preserving filtering on the image to maintain clear boundaries while reducing internal noise, to obtain the preprocessed image.

[0009] The transparent pipe wall thickness detection method provided by the present application includes the following steps of extracting edge points of the pipe in the image and performing linear fitting to obtain the accurate profile of the inner wall of the pipe and the outer wall of the pipe: performing contrast stretch on the preprocessed image to enhance the contrast of dark and bright areas, and then performing histogram equalization or adaptive histogram equalization to obtain a to-be-processed image; processing the to-be-processed image using a Canny edge detection algorithm to extract the edges of the inner wall of the pipe and the outer wall of the pipe.

[0010] The transparent pipe wall thickness detection method provided by the present application includes the following steps of repeating the calculation of the wall thickness of multiple points, statistically analyzing the wall thickness values of the multiple points, and outputting the average value, the maximum value, the minimum value, and the standard deviation to obtain the measured wall thickness of the pipe. using a calibration rod to calibrate the mapping relationship between the pixel size of the image and the actual physical size; adjusting the spatial positions of the first acquisition camera and the second acquisition camera to verify the spatial coordinate alignment accuracy and ensure that the alignment error is less than or equal to ±1 μm.

[0011] The transparent pipe wall thickness detection method provided by the present application includes the following steps of calculating the Euclidean distance between the two coordinate points as the wall thickness of the point based on the outer wall coordinate point of any outer wall point and the inner wall coordinate point of the nearest inner wall point, and detecting the collimation degree of the pipe. calculating the three-dimensional spatial coordinates of the pipe edge points based on the pixel coordinates of the images obtained by the first camera and the second camera, and combining the calibration parameters of the first camera and the second camera; calculating the spatial position of the center axis of the pipe using the principle of triangulation, and analyzing the bending degree.

[0012] The second aspect of the present application provides a detection device for performing the transparent pipe wall thickness detection method of any one of the above, comprising: a base and a connecting seat; the connecting seat is connected with the base, a through hole for pipe passing through is arranged in the middle of the connecting seat, an arc-shaped sliding groove hole is arranged in the connecting seat along the circumference of the through hole, and an image acquisition assembly is arranged in the sliding groove hole; the image acquisition assembly comprises: a moving mechanism, a first acquisition camera and a second acquisition camera; the moving mechanism is arranged in the sliding groove hole, the first acquisition camera and the second acquisition camera are connected with the moving mechanism, an included angle is arranged between the first acquisition camera and the second acquisition camera, and the lens ends of the first acquisition camera and the second acquisition camera are both arranged towards the through hole.

[0013] According to the detection device provided by the present application, a plurality of backlight sources are arranged on the side opposite to the sliding groove hole, and the plurality of backlight sources are arranged at intervals around a part of the through hole.

[0014] According to the detection device provided by the present application, the moving mechanism comprises a driving device, a traction chain and a plurality of toothed sprockets, the plurality of toothed sprockets are arranged in the sliding groove hole, the traction chain is arranged around the plurality of toothed sprockets to form a closed-loop moving chain, and the driving device is in transmission connection with one of the toothed sprockets to drive the traction chain to move; wherein, the traction chain is provided with a connecting plate, and the first acquisition camera and the second acquisition camera are arranged on the connecting plate.

[0015] The transparent pipe wall thickness detection method provided by the present application can acquire original images in two directions respectively, and realize wall thickness detection based on the original images, which can improve the accuracy of wall thickness detection based on the acquisition of two points, and can obtain the overall image of the pipe when the pipe moves a small distance or rotates a small angle, thereby improving the efficiency of wall thickness detection. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present application or prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0017] Figure 1 is a flowchart of the transparent pipe wall thickness detection method provided by the present application.

[0018] Figure 2 is a schematic diagram of the overall installation of the detection device provided by the present application.

[0019] Figure 3 is a schematic diagram of the three-dimensional structure of the detection device provided by the present application.

[0020] Figure 4 is a schematic diagram of the structure in the sliding groove hole of the detection device provided by the present application.

[0021] Figure 5 is a schematic diagram of the specific structure of the moving mechanism in the detection device provided by the present application.

[0022] Figure 6 is one of the schematic diagrams of the state of image acquisition of the two acquisition cameras in the detection device provided by the present application.

[0023] Figure 7 is the second schematic diagram of the state of image acquisition of the two acquisition cameras in the detection device provided by the present application.

[0024] Reference signs: 1, detection device; 11, base; 12, connecting seat; 121, through hole; 122, sliding groove hole; 13, image acquisition assembly; 131, first acquisition camera; 132, second acquisition camera; 133, moving mechanism; 1331, driving device; 1332, toothed chain wheel; 1333, traction chain; 1334, connecting block; 14, backlight source; 2, upper computer; 3, pipe. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical scheme and advantages of the present application clearer, the technical scheme of the present application will be described clearly and completely below in combination with the drawings in the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0026] In the description of the embodiments of the present application, it should be noted that the orientations or positional relationships indicated by the terms "center", "longitudinal", "transverse", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like are based on the orientations or positional relationships shown in the drawings, and are only for the purpose of clarifying the embodiments of the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the embodiments of the present application. In addition, the terms "first", "second", "third" are only for the purpose of description, and cannot be understood as indicating or implying relative importance.

[0027] In the description of the embodiments of the present application, it should be noted that, unless otherwise explicitly specified and limited, the terms "connected", "connected" should be understood in a broad sense, for example, it can be fixedly connected, or it can be detachably connected, or it can be integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium. For those skilled in the art, the specific meaning of the above terms in the embodiments of the present application can be understood according to the specific circumstances.

[0028] In the embodiments of the present application, unless otherwise explicitly specified and limited, the first feature is "on" or "under" the second feature, which can be direct contact between the first and second features, or indirect contact between the first and second features through an intermediate medium. Moreover, the first feature "above", "over" and "on" the second feature can be directly above or obliquely above the first feature, or only indicate that the horizontal height of the first feature is higher than that of the second feature. The first feature "below", "under" and "under" the second feature can be directly below or obliquely below the first feature, or only indicate that the horizontal height of the first feature is less than that of the second feature.

[0029] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine the different embodiments or examples described in the present specification and the features of the different embodiments or examples without contradiction.

[0030] In the related art, a single-dimensional wall thickness inspection is generally used, i.e. focusing on measurement in two-dimensional space, lacking the ability to obtain three-dimensional information of the internal structure of the product. This limits the comprehensive understanding of the integrity of complex structures such as hollow tubes, resulting in low detection accuracy of wall thickness and being unable to provide a stereoscopic view for in-depth geometric analysis and performance prediction.

[0031] In view of the problems in the related art, as shown in Figure 1 The present application provides a transparent pipe wall thickness detection method, comprising the following steps: Step S100, obtaining the original image obtained by the pipe along the first direction and the second direction. The transparent pipe 3 needs to collect images first when detecting the wall thickness. In the present embodiment, the image information of the pipe is efficiently and completely obtained by setting the collection device in two directions, and the detection quality of the wall thickness is improved by collecting in two directions.

[0032] Specifically, the first acquisition camera 131 is arranged in the first direction, and the second acquisition camera 132 is arranged in the second direction. The pipe is rotated or moved, and the first acquisition camera 131 and the second acquisition camera 132 are controlled to continuously capture original images in the process of rotation or movement. That is, in this embodiment, the image acquisition is realized by using a binocular camera. This method can improve the efficiency and quality of image acquisition, and realize efficient image acquisition.

[0033] In step S200, the original image is processed to obtain a processed feature-enhanced image. The original image needs to be processed after being captured. In this embodiment, the pipe feature is enhanced through processing, so as to avoid the influence of other parts in the image.

[0034] Specifically, the original image is preprocessed to obtain a preprocessed image, and the preprocessed image is subjected to feature-enhanced image processing to obtain a feature-enhanced image. In this way, the pipe feature can be enhanced, thereby facilitating the extraction of edge points in the subsequent process.

[0035] When the image is processed, the original image is first subjected to grayscale conversion to reduce the number of color channels. In this way, the complexity of calculation can be simplified, thereby facilitating the subsequent processing.

[0036] Next, the image subjected to grayscale conversion is subjected to noise reduction and smoothing processing. Specifically, unnecessary details or random fluctuations are removed by using a filter, such as Gaussian blur processing. The edges of the image are subjected to reserved filtering to maintain clear boundaries while reducing internal noise, thereby obtaining a preprocessed image.

[0037] In step S300, the edge points of the pipe in the feature-enhanced image are extracted, and straight line fitting is performed to obtain the accurate contour of the inner wall of the pipe and the outer wall of the pipe. After the image is processed, the pipe feature in the image is enhanced, and then the edge points of the pipe contour can be extracted. By fitting a plurality of edge points, a three-dimensional pipe image can be formed.

[0038] When the edge points are extracted and fitted, the contrast of the preprocessed image is first enhanced to enhance the contrast between the dark area and the bright area. Specifically, the pixel intensity distribution range is adjusted so that the dark area becomes darker and the bright area becomes brighter. Then, histogram equalization or adaptive histogram equalization is used to obtain a to-be-processed image.

[0039] Next, the to-be-processed image is processed by using a Canny edge detection algorithm to extract the edges of the inner wall of the pipe and the outer wall of the pipe. Specifically, the steps of the Canny edge detection algorithm are as follows: The gradient amplitude and direction of the image are calculated: ; Gx calculates its horizontal gradient component: ∂I / ∂x represents the limit of the rate of change of a function along the x direction at a certain point; Gy calculates its vertical gradient component: ∂I / ∂y represents the limit of the rate of change of a function along the y direction at a certain point.

[0040] Gradient magnitude: ; The greater the value of the gradient magnitude G(x, y), the more intense the gray scale change at the pixel point (x, y), and the more likely it is an edge point.

[0041] Gradient direction: ; The gradient direction θ(x, y) represents the direction in which the gray scale of the image changes most rapidly at the pixel point (x, y) (i.e., the normal direction of the edge). It is determined by the horizontal gradient Gx and the vertical gradient Gy, and is calculated using the arctangent function arctan.

[0042] Non-maximum suppression is performed on the gradient magnitude, the local maximum points are retained, and a double-threshold method is used to determine the edge points, connect the edge contours, and realize edge fitting.

[0043] In step S400, based on the outer wall coordinate point of any outer wall point, the inner wall coordinate point of the nearest inner wall point is obtained, and the Euclidean distance between the two coordinate points is calculated as the wall thickness of the point. Specifically, first, a point on the outer wall is determined, then the nearest point on the inner wall is obtained, and the coordinate values of the points on the outer wall and the inner wall are obtained. After obtaining the coordinate values, the Euclidean distance between the two can be calculated, and the calculated distance value is the wall thickness of the point.

[0044] It should be noted that since the pipe material is transparent, the inner wall and outer wall of the pipe material can be obtained in the above obtained image, so the coordinate points of the inner wall and outer wall will appear in the image, and the wall thickness can be obtained by calculation.

[0045] In a specific example, for each outer wall point (X out , Y out , Z out ), find the nearest inner wall point (X in , Y in , Z in ), and calculate the Euclidean distance between the two points: ; In the formula, (X out , Y out , Z out ) represents the coordinate value of the outer wall of the pipe material, (X in , Y in , Zin represents the coordinate value of the inner wall point closest to the selected point.

[0046] Step S500, the wall thickness of multiple points is repeatedly calculated, and statistical analysis is performed on the wall thickness values of the multiple points, and the average value, maximum value, minimum value and standard deviation are output to obtain the measured wall thickness of the pipe. In order to improve the accuracy of the wall thickness detection, multiple point detection is used in the embodiment, and different values are obtained through the statistics to meet the design requirements of different designs.

[0047] Specifically, generally, if the uniformity of the wall thickness is required to be high, the average value can be selected, but if the thinnest part of the wall thickness is concerned or the safety of the pipeline is required to be ensured, the minimum value can be used. For special requirements, for example, in high-strength occasions such as pressure vessels, the maximum value or higher safety standards can be considered. By outputting multiple wall thickness values, different design requirements can be met, and flexibility is improved.

[0048] According to the present application, in some embodiments, after repeatedly calculating the wall thickness of multiple points, and performing statistical analysis on the wall thickness values of the multiple points, the average value, maximum value, minimum value and standard deviation are output to obtain the measured wall thickness of the pipe, and a calibration step is further included, specifically comprising: It can be understood that, by detecting the center offset in two directions, the present application prevents the detection of wall thickness from being wrong due to inclination, improves the accuracy of online wall thickness detection, and improves the reliability of detection.

[0049] Step S600, the mapping relationship between the pixel size of the image and the actual physical size is calibrated by using the calibration rod; and the spatial positions of the first acquisition camera and the second acquisition camera are adjusted to verify the spatial coordinate alignment accuracy and ensure that the alignment error is less than or equal to ±10μm. In actual online detection, the accuracy of thickness measurement needs to be ensured, in the embodiment, the actual physical size is mapped by using the calibration rod, and the spatial positions of the first acquisition camera 131 and the second acquisition camera 132 are adjusted to verify, so that the accuracy of calibration can be realized.

[0050] Specifically, the spatial position relationship of the first acquisition camera 131 and the second acquisition camera 132 will affect the overall accuracy, in the embodiment, the calibration is realized by using the calibration rod, and the spatial positions of the two cameras are adjusted to realize the overall alignment accuracy and improve the detection accuracy.

[0051] The specific implementation process of step S600 includes: First calibration: the value obtained by the camera is pixel value. The corresponding pixel value can be obtained by a calibration rod (for example, a standard rod body with a fixed outer diameter, with high precision up to 1 um). The size corresponding to each pixel, i.e. the actual physical size, is obtained by the outer diameter ratio pixel value. Usually, the size corresponding to a single pixel is about 10 um, and after double camera calibration, the error can be ensured to be less than or equal to 10 um after averaging.

[0052] Secondly, the two cameras can separately measure the outer diameter and the inner diameter (the measurement method is written correctly), and the key is that the measurement of the double camera can determine the current position state of the pipe material, and the deviation amount of the X, Y axis from the center axis is determined (for example, the X axis deviation amount is the difference between the center line of the two outer diameter edges detected and the center line of the camera image, and the Y axis is the same). The deviation amount of the current pipe material can be dynamically adjusted by other ways (for example, adjusting the position of the mother pipe), so as to draw the pipe material with better bow degree.

[0053] Since the included angle of the two cameras is a fixed angle, there is an overlapping part in the vision. First, the fixed included angle and the fixed position are used to form a parallel optical axis relationship, and then the binocular overlapping part is used to form a depth map. The depth map can form a point cloud map, thereby constructing a high-precision 3D model.

[0054] According to some embodiments provided by the present application, when step S400 is performed, the pipe material collimation degree detection is also included, and the collimation degree detection method comprises the following steps: Step S401, based on the pixel coordinates of the images obtained by the first camera and the second camera, and in combination with the calibration parameters of the first camera and the second camera, the three-dimensional space coordinates of the pipe material edge points are calculated.

[0055] Step S402, the spatial position of the pipe material center axis is calculated by using the principle of triangulation, and the bending degree is analyzed. Through the acquisition of the spatial position of the pipe material center axis, the bending degree judgment can be realized based on the position change of the center axis, and then the collimation degree detection of the pipe material can be realized.

[0056] The center axis calculation method is as follows: First, the three-dimensional point cloud data of the inner and outer walls of the pipe material is fitted to obtain the spatial equation of the center axis. The least square method is used to fit the straight line equation: ; In the formula, (X0, Y0, Z0) is a point on the straight line, and (a, b, c) is a direction vector.

[0057] Specifically, in the continuous frame image, the bending degree of the transparent pipe material 3 is calculated by tracking the position change of the center axis; the center line is smoothed by using a polynomial fitting or spline interpolation method to eliminate noise interference, and the position change of the two-axis center axis is output synchronously, and the pipe material collimation degree is analyzed in real time.

[0058] As Figures 2-5 shown, the second aspect of the present application provides a detection device 1 for performing the detection method of the wall thickness of the transparent pipe 3 in any of the above embodiments, comprising a base 11 and a connecting seat 12; the connecting seat 12 is connected with the base 11, a through hole 121 for pipe passing is arranged in the middle of the connecting seat 12, an arc-shaped sliding groove hole 122 is arranged in the connecting seat 12 along the circumference of the through hole 121, and an image acquisition assembly 13 is arranged in the sliding groove hole 122; the image acquisition assembly 13 comprises a moving mechanism 133, a first acquisition camera 131 and a second acquisition camera 132; the moving mechanism 133 is arranged in the sliding groove hole 122, the first acquisition camera 131 and the second acquisition camera 132 are connected with the moving mechanism 133, an included angle is arranged between the first acquisition camera 131 and the second acquisition camera 132, and the lens end of the first acquisition camera 131 and the lens end of the second acquisition camera 132 are both arranged towards the through hole 121. When the wall thickness of the transparent pipe 3 is detected, images need to be acquired and analyzed based on the images. In the embodiment, a binocular camera group is formed by the arrangement of the first acquisition camera 131 and the second acquisition camera 132, the binocular camera group can acquire images in two directions of the pipe at the same time, so as to improve the efficiency of acquisition, and the accuracy of image acquisition can be improved through the acquisition of images in two directions.

[0059] Specifically, as Figure 2 shown, the entire device is arranged below the furnace opening, the pipe is detected immediately after being formed, and feedback is implemented. Of course, the detection device 1 also comprises an industrial computer for data processing, the first acquisition camera 131 and the second acquisition camera 132 are in communication connection with the industrial computer, so as to transmit the acquired image data and the like to the industrial computer, the detection method described above is built in the industrial computer, and the automatic detection of the pipe wall thickness is realized by executing the detection method.

[0060] Specifically, as Figure 3 , Figure 4 shown, the lens end of the first acquisition camera 131 is arranged along a first direction, the lens end of the second acquisition camera 132 is arranged along a second direction, and the included angle between the first acquisition camera 131 and the second acquisition camera 132 is 60°. After the included angle is fixed, the two acquisition cameras can detect the pipe wall thickness in two directions at the same time, and can realize the on-line two-axis axial offset detection, so as to real-time feedback the posture of the current pipe.

[0061] In the specific embodiment, in the production state, the transparent tube 3 in the drawing process passes through the through hole 121 uninterruptedly, the first acquisition camera 131 and the second acquisition camera 132 rotate continuously around the transparent tube, and the image frame data of two directions of the transparent tube are collected correspondingly and sent to the industrial computer (image and data processing system) at the same time. When the industrial computer (image and data processing system) acquires each frame of the transparent tube image according to one of the two acquisition cameras, the frame analysis of each acquired image is performed. Specifically, first, the transparent tube wall boundary is determined by using the brightness difference between the transparent tube wall boundary and the outer background and the hollow background of the transparent tube. According to the known pixel value of the whole image, the pixel value between the transparent tube wall boundaries can be obtained. According to the mm value corresponding to each pixel value, the size of the two transparent tube walls can be calculated. The detection results and the setting parameters are comprehensively analyzed to determine whether the thickness of the transparent tube wall is qualified in real time.

[0062] In combination with the above embodiment, three-dimensional reconstruction of the pipe can be realized by the two acquisition cameras. The three-dimensional reconstruction is realized by taking pictures of the transparent tube from different positions by the two cameras to obtain two images. The three-dimensional geometric information (i.e., the depth map) of the object can be recovered by calculating the position deviation (i.e., the parallax) between the corresponding points in the two images. The picture taken at this moment is fused with the depth map to obtain the three-dimensional reconstruction of the transparent tube at an angle.

[0063] In the specific setting, the arc of the arc-shaped sliding groove hole 122 is greater than or equal to 180 degrees. This enables the two acquisition cameras to rotate 180 degrees around the transparent tube and to collect multiple points to realize multiple reconstruction, thereby obtaining a more complete three-dimensional image of the transparent tube.

[0064] In addition, since the binocular camera is not on the same optical axis, a general model of binocular stereo vision must be established to analyze the actual system. As shown in Figure 6 , Figure 7 , the first acquisition camera 1331 is a left camera, and the second acquisition camera 1332 is a right camera. The left camera coordinate system O l -X l Y l Z l is fixed in the world coordinate system, and the image coordinate system o l -x l y l has a focal length f l . The right camera coordinate system Or-XrYrZr corresponds to the image coordinate system or-xryr, and the focal length is fr. The optical axis of the right camera forms an angle θ with the left camera. The projection transformation models of the left and right cameras (the image coordinate system is in physical unit scale) are as follows: ; wherein Sl / Zl=Sr / Zr=1.

[0065] Let the transformation matrix of the left camera coordinate system relative to the right camera coordinate system be: ; Therefore, the coordinates in the right coordinate system are expressed as the coordinates in the left coordinate system and the transformation matrix: ; By combining the above formula, the coordinates of the target point in the left coordinate system (world coordinate system) can be calculated: ; From the above derivation, it can be seen that the three-dimensional coordinates of the target point can be obtained by knowing the focal length f l , fr, the coordinates of the target point in the left and right cameras p l , pr, the transformation matrix R, and T. Thus, image correction and three-dimensional reconstruction are completed.

[0066] According to some embodiments provided by the present application, a plurality of backlight sources 14 are arranged on the side opposite to the sliding groove hole 122, and the plurality of backlight sources 14 are arranged at intervals around a part of the through hole 121. In order to obtain accurate and detailed images during image acquisition, the backlight sources 14 are arranged in the embodiment to provide sufficient illumination for camera acquisition, so that the pipe body is highlighted to a higher degree, and image acquisition can be better achieved.

[0067] Specifically, the first acquisition camera 131 and the second acquisition camera 132 are arranged on one side of the through hole 121, and the plurality of backlight sources 14 are arranged on the other side of the through hole 121, and the lens end faces of the first acquisition camera 131 and the second acquisition camera 132 face the through hole 121, so that the two cameras can acquire more illumination, so that the pipe material can be highlighted in the image, which is beneficial to subsequent image processing.

[0068] Specifically, three backlight sources 14 are arranged on the side of the through hole 121 opposite to the sliding groove hole 122, and the three backlight sources 14 are arranged side by side and form a light source array similar to 180°, so as to provide sufficient illumination intensity and realize high-quality image acquisition.

[0069] According to some embodiments provided by the present application, as Figure 5As shown, the moving mechanism 133 includes a driving device 1331, a traction chain 1333 and a plurality of toothed sprockets 1332 arranged in the sliding groove hole 122, the traction chain 1333 is wound on the plurality of toothed sprockets to form a closed loop moving chain, the driving device 1331 is in transmission connection with one of the toothed sprockets 1332 to drive the traction chain 1333 to move; wherein the traction chain 1333 is provided with a connecting plate, the first acquisition camera 131 and the second acquisition camera 132 are arranged on the connecting plate. When different specifications and different needs are required, the first acquisition camera 131 and the second acquisition camera 132 have different image acquisition positions, in the embodiment, the first acquisition camera 131 and the second acquisition camera 132 are connected with the traction chain 1333, so as to be driven by the traction chain 1333 to move in the sliding groove hole 122, thereby realizing image acquisition at different positions.

[0070] Specifically, the driving device 1331 includes a driving motor, the output end of the driving motor is engaged with one of the toothed sprockets 1332, so as to drive the toothed sprocket 1332 to rotate, the toothed sprocket 1332 is engaged with the traction chain 1333, so as to drive the traction chain 1333 to move, thereby realizing the movement of the first acquisition camera 131 and the second acquisition camera 132.

[0071] Among them, the first acquisition camera 131 and the second acquisition camera 132 are both CCD cameras, the lenses of which are directed to the side of the through hole 121 to realize the collection of the pipe image. The backlight source 14 is an LED light source, and a blue diffuse reflection LED highlight backlight source 14 can also be used.

[0072] Specifically, the connecting plate has two mounting holes, the first acquisition camera 131 and the second acquisition camera 132 are connected through the two mounting holes, so as to realize the installation of the two cameras. When setting, the two mounting holes can be opened in a 60° angle, so that the first acquisition camera 131 and the second acquisition camera 132 connected to the two mounting holes have a 60° angle between them.

[0073] In some specific embodiments, limit sensors are arranged at the limit positions on both sides of the sliding groove hole 122, the limit sensors can limit the movement of the first acquisition camera 131 and the second acquisition camera 132, so as to limit the movement of the two cameras in the sliding groove hole 122.

[0074] It should be pointed out finally that the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit the same; and although the present application has been described in detail with reference to the foregoing embodiments, it should be appreciated by those skilled in the art that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features thereof can be replaced equivalently; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method of detecting the wall thickness of a transparent pipe, characterized by, The method comprises the following steps: Obtaining original images of the pipe along a first direction and a second direction; Processing the original images to obtain a processed feature-enhanced image; Extracting edge points of the pipe in the feature-enhanced image and performing linear fitting to obtain the contours of the inner wall and the outer wall of the pipe; Based on the outer wall coordinate point of any outer wall point and the inner wall coordinate point of the nearest inner wall point, the Euclidean distance between the two coordinate points is calculated as the wall thickness of the point; Repeat the calculation of the wall thickness of multiple points, and statistically analyze the wall thickness values of multiple points to output the average value, maximum value, minimum value and standard deviation to obtain the measured wall thickness of the pipe.

2. The method of claim 1, wherein Obtaining original images of the pipe along a first direction and a second direction comprises the following steps: A first acquisition camera is arranged in the first direction, and a second acquisition camera is arranged in the second direction. The pipe is controlled to rotate or move, and the first acquisition camera and the second acquisition camera are controlled to continuously shoot during the rotation or movement to obtain the original images.

3. The method of claim 1, wherein Processing the original images to obtain a processed feature-enhanced image comprises the following steps: Pretreating the original images to obtain a pretreated image; Enhancing the features of the pretreated image to obtain the feature-enhanced image.

4. The method of claim 3, wherein Pretreating the original images to obtain a pretreated image comprises the following steps: Converting the original image to grayscale to reduce the number of color channels; Performing noise reduction and smoothing processing on the grayscale-converted image, and performing edge retention filtering on the image to maintain clear boundaries while reducing internal noise to obtain the pretreated image.

5. The method of claim 3, wherein Extracting edge points of the pipe in the image and performing linear fitting to obtain the accurate contours of the inner wall and the outer wall of the pipe comprises the following steps: Contrast stretching is performed on the pretreated image to enhance the contrast of dark and bright areas, and then histogram equalization or adaptive histogram equalization is performed to obtain a to-be-processed image; The to-be-processed image is processed using a Canny edge detection algorithm to extract the edges of the inner wall and the outer wall of the pipe.

6. The method of claim 2, wherein After repeatedly calculating the wall thickness of multiple points and statistically analyzing the wall thickness values of multiple points to output the average value, maximum value, minimum value and standard deviation to obtain the measured wall thickness of the pipe, a calibration step is further included, which comprises: Using a calibration rod to calibrate the mapping relationship between the pixel size and the actual physical size of the image; Adjusting the spatial positions of the first acquisition camera and the second acquisition camera to verify the alignment accuracy of the spatial coordinates and ensure that the alignment error is less than or equal to ±1 μm.

7. The method of claim 6, wherein While calculating the Euclidean distance between the two coordinate points as the wall thickness of the point based on the outer wall coordinate point of any outer wall point and the inner wall coordinate point of the nearest inner wall point, the collimation degree of the pipe is also detected, and the collimation degree detection method comprises the following steps: Based on the pixel coordinates of the images obtained by the first camera and the second camera, and combined with the calibration parameters of the first camera and the second camera, the three-dimensional spatial coordinates of the pipe edge points are calculated; Using the principle of triangulation, the spatial position of the center axis of the pipe is calculated, and the bending degree is analyzed.

8. An apparatus for performing the method of claim 1 to 6, characterized in that It comprises: a base; The connecting seat is connected with the base, a through hole for pipe passing through is arranged in the middle of the connecting seat, an arc-shaped sliding slot hole is arranged in the connecting seat along the circumference of the through hole, and an image acquisition assembly is arranged in the sliding slot hole; The image acquisition assembly comprises a moving mechanism, a first acquisition camera and a second acquisition camera; The moving mechanism is arranged in the sliding slot hole, the first acquisition camera and the second acquisition camera are connected with the moving mechanism, an included angle is arranged between the first acquisition camera and the second acquisition camera, and the lens ends of the first acquisition camera and the second acquisition camera are both arranged towards the through hole.

9. The detection device of claim 8, wherein, A plurality of backlight sources are arranged on the side opposite to the sliding slot hole, and the plurality of backlight sources are arranged at intervals around a part of the through hole.

10. The detection device of claim 8, wherein, The moving mechanism comprises a driving device, a traction chain and a plurality of toothed chain wheels, the plurality of toothed chain wheels are arranged in the sliding slot hole, the traction chain is wound on the plurality of toothed chain wheels to form a closed-loop moving chain, and the driving device is in transmission connection with one of the toothed chain wheels to drive the traction chain to move. The traction chain is provided with a connecting plate, and the first acquisition camera and the second acquisition camera are arranged on the connecting plate.

Images