Large-size cylinder coaxiality detection method and system

Abstract

The invention relates to the technical field of mechanical detection, in particular to a large-size cylinder coaxiality detection method and system.The method comprises the steps that images of the two ends of a large-size cylinder are collected, and a copper cylinder, a pole tube and a pole root are defined as key circular areas respectively; preprocessing the acquired image, and processing the acquired image by adopting a circle fitting algorithm to obtain the circle center coordinate and the radius of the key circle area; according to the circle center coordinates and the radiuses of the key circular areas, relative position deviations between the key circular areas are calculated; and according to a calculation result, if the relative position deviation of each key circular area is within a preset tolerance range, the coaxiality is judged to be qualified, otherwise, the coaxiality is judged to be unqualified, and a detection result is obtained. According to the invention, the problems of low precision, poor efficiency and dependence on artificial experience in traditional coaxiality detection are effectively solved, and the anti-interference capability and industrial efficiency of detection are significantly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of mechanical detection, and in particular to a method and system for detecting the coaxiality of a large-sized cylinder. Background Art

[0002] In the field of mechanical manufacturing and quality control, coaxiality is a key geometric tolerance parameter, which is mainly used to reflect the deviation between the axes that should theoretically remain coaxial in actual parts. For large-sized cylindrical structures, especially when complex components such as copper columns, pole tubes and pole roots are involved, ensuring the coaxiality of each component is crucial to ensuring the performance and quality of the product. Traditionally, coaxiality measurement usually relies on manual operations and simple mechanical detection methods such as the dial indicator rotation method, V-bracket method and lever dial indicator method. Although these methods are easy to operate and low in cost, they are usually limited by factors such as instrument accuracy, environmental interference, and operator experience, resulting in large measurement errors. Especially in the detection of large-sized or complex structural parts, the repeatability and accuracy of the measurement results are difficult to guarantee, and the error may reach several percent.

[0003] Copper pillars are common conductive parts in electrical components, and pole tubes and pole roots usually serve as their supporting and connecting parts. In actual production, their coaxiality is an important parameter for testing their assembly quality and performance. In traditional detection methods, it is difficult to achieve accurate measurement and efficient automated detection of the coaxiality of these important parts, especially when the relative position relationship between multiple key parts (such as copper pillars, pole tubes, pole roots, etc.) is complex. The accuracy and efficiency of traditional manual detection methods cannot meet the needs of modern industrial production. Summary of the invention

[0004] The present invention provides a method and system for detecting the coaxiality of a large-sized cylinder, thereby effectively solving the problems pointed out in the background technology.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] A method for detecting the coaxiality of a large-sized cylinder, comprising:

[0007] Collect images of both ends of a large-sized cylinder and define the copper column, pole tube, and pole root as key circular areas respectively;

[0008] Preprocessing the acquired image, and processing the acquired image using a circular fitting algorithm to obtain the center coordinates and radius of the key circular area;

[0009] Calculating relative position deviations between the key circular areas according to the center coordinates and radius of the key circular areas;

[0010] According to the calculation results, if the relative position deviation of each of the key circular areas is within a preset tolerance range, the coaxiality is determined to be qualified, otherwise the coaxiality is determined to be unqualified, and the test result is obtained.

[0011] Furthermore, the method also includes: in the fitting process, correcting the image that still has errors after preprocessing by using an optimization algorithm.

[0012] Furthermore, images of both ends of the large-sized cylinder are collected, including:

[0013] The cameras are installed at both ends of the large-sized cylinder, and the shooting angle after installation covers the measurement area of ​​the entire cylinder;

[0014] A strip light source and an annular light source combined on four sides are configured to illuminate the bottom reference surface of the large-sized cylinder and the pole tube respectively.

[0015] Furthermore, the pretreatment process includes:

[0016] Grayscale the acquired image;

[0017] Using a denoising algorithm to remove noise from the grayscale image;

[0018] The denoised image is processed using an edge detection algorithm to extract edge points of the key circular area.

[0019] Further, a circular fitting algorithm is used to process the acquired image to obtain the center coordinates and radius of the key circular area, including:

[0020] Mapping each edge point into a parameter space, and generating possible circle parameters according to different radii and circle center coordinates;

[0021] Voting for each edge point, and accumulating the corresponding circle center coordinates and radius values ​​in the parameter space;

[0022] The center coordinate and radius combination with the highest number of votes is selected as the fitting result to obtain the center coordinate and radius of the key circular area of ​​the large-sized cylinder.

[0023] Furthermore, before calculating the relative position deviation, the center coordinates of each of the key circular areas are checked for spatial consistency using multi-view fusion technology.

[0024] Furthermore, before calculating the relative position deviation, the center coordinates of each of the key circular areas are checked for spatial consistency by using a multi-view fusion technique, including:

[0025] Get the three-dimensional space coordinates of the two ends of the large-size cylinder;

[0026] Map the coordinates of the center of the circle fitted at different viewing angles to the same three-dimensional coordinate system;

[0027] Align the coordinates of the center of the circle from multiple perspectives and remove abnormal coordinate points;

[0028] If the coordinate error of the circle center after registration is less than the preset threshold, the spatial consistency check is considered to have passed, otherwise the image is recaptured.

[0029] Further, the relative position deviations between the key circular areas are calculated according to the center coordinates and radius of the key circular areas, including:

[0030] Taking the center coordinates of the copper column as the reference axis, respectively calculating the radial offsets of the center coordinates of the pole tube and the pole root relative to the reference axis;

[0031] According to the radial offset, the formula Calculate the relative position deviation of each circle center in the two-dimensional plane, where Δ represents the radial offset between two points, (x0, y0) is the reference circle center coordinate, (x i ,y i ) are the coordinates of the center of the circle to be measured.

[0032] A large-size cylindrical coaxiality detection system, the system comprising:

[0033] The image acquisition module collects images of both ends of the large-sized cylinder and defines the copper column, pole tube and pole root as key circular areas respectively;

[0034] The data acquisition module pre-processes the acquired image and processes the acquired image using a circular fitting algorithm to obtain the center coordinates and radius of the key circular area;

[0035] A deviation calculation module, which calculates the relative position deviation between the key circular areas according to the center coordinates and radius of each key circular area;

[0036] The result acquisition module determines that the coaxiality is qualified if the relative position deviation of each key circular area is within a preset tolerance range according to the calculation result, otherwise the coaxiality is judged to be unqualified, and the detection result is obtained.

[0037] Furthermore, the deviation calculation module includes:

[0038] An offset calculation unit, taking the center coordinates of the copper column as a reference axis, and respectively calculating the radial offsets of the center coordinates of the pole tube and the pole root relative to the reference axis;

[0039] The position deviation determination unit determines the radial offset according to the formula Calculate the relative position deviation of each circle center in the two-dimensional plane, where (x0, y0) is the reference circle center coordinate, (x i ,y i ) are the coordinates of the center of the circle to be measured.

[0040] The technical solution of the present invention can achieve the following technical effects:

[0041] The present invention effectively solves the problems of low accuracy, poor efficiency and reliance on manual experience in traditional coaxiality detection, and significantly improves the anti-interference ability and industrial efficiency of detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 It is a flow chart of a method for detecting the coaxiality of a large-sized cylinder;

[0044] Figure 2 It is a schematic diagram of the process of collecting images of both ends of a large-sized cylinder;

[0045] Figure 3 It is a schematic diagram of the process of pretreatment;

[0046] Figure 4 A schematic diagram of the process of obtaining the center coordinates and radius of the key circular area;

[0047] Figure 5 The figure is a flowchart for performing spatial consistency check on the center coordinates of each key circular area. DETAILED DESCRIPTION

[0048] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0049] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0050] Embodiment 1

[0051] like Figure 1 As shown, the present invention provides a method for detecting the coaxiality of a large-sized cylinder, the method comprising:

[0052] S1: Collect images of both ends of a large-sized cylinder and define the copper column, pole tube and pole root as key circular areas respectively;

[0053] Specifically, in the process of coaxiality detection of large-sized cylinders, it is first necessary to ensure that accurate image data is obtained, which provides a reliable basis for subsequent circular fitting and coaxiality calculation. The goal of this step is to collect images of both ends of the large-sized cylinder and define the copper column, pole tube and pole root in the image as key circular areas, so as to ensure that the position and geometric features of these key components can be accurately identified from the image. This process is not just a simple acquisition of images, but also to extract accurate center coordinates and radius data from the image, which will serve as an important basis for subsequent coaxiality analysis. By clarifying the position and size of these key circular areas, it can ensure that the coaxiality data calculated in the subsequent steps has high precision, and provide accurate input for the subsequent circular fitting algorithm.

[0054] S2: preprocessing the acquired image, and using a circular fitting algorithm to process the acquired image to obtain the center coordinates and radius of the key circular area;

[0055] Specifically, the captured images may contain noise, uneven lighting, blur and other factors, which will affect the accuracy of subsequent circular fitting. Therefore, the images need to be preprocessed first to remove unnecessary interference and improve image quality. The preprocessed images can provide more accurate input for the application of circular fitting algorithms. Through circular fitting algorithms (such as least squares fitting or Hough transform), the center coordinates and radius can be extracted from key circular areas in the image (such as copper columns, pole tubes and pole roots). These geometric features are the core data for subsequent coaxiality calculations, which are used to determine whether the coaxiality requirements are met between various components.

[0056] S3: Calculate the relative position deviation between the key circular areas according to the center coordinates and radius of each key circular area;

[0057] Specifically, the calculation of relative position deviation is achieved by comparing the distance and angle differences between different circle center coordinates. Through this process, the relative position errors of each key circular area can be quantified to determine whether they are on the same axis. The calculation of relative position deviation helps to evaluate whether the installation of each component is accurate and whether there is any offset or deformation, thereby ensuring the coaxiality of the entire large-size cylinder.

[0058] S4: According to the calculation results, if the relative position deviation of each key circular area is within the preset tolerance range, the coaxiality is determined to be qualified, otherwise the coaxiality is determined to be unqualified, and the test result is obtained.

[0059] Specifically, by comparing the calculated relative position deviation with the preset tolerance range, it is determined whether the coaxiality can be judged as qualified. If the relative position deviation of each key circular area is within the preset tolerance range, it means that the relative position of each component meets the specified tolerance requirements, and the coaxiality of the entire large-size cylinder can be judged as qualified; on the contrary, if the relative position deviation exceeds the tolerance range, it means that there is a deviation in the installation of the component and the coaxiality requirements cannot be met, and it is judged as unqualified.

[0060] The present invention effectively solves the problems of low accuracy, poor efficiency and reliance on manual experience in traditional coaxiality detection, and significantly improves the anti-interference ability and industrial efficiency of detection.

[0061] As a preferred embodiment of the above embodiment, it also includes: in the fitting process, correcting the image that still has errors after preprocessing through an optimization algorithm.

[0062] Specifically, in the fitting process, the preliminary image data is first obtained through image preprocessing, but due to factors such as noise and uneven illumination, there may be errors. At this time, the optimization algorithm will be introduced in the preferred step to correct the image that still has errors after preprocessing. In specific implementation, the optimization algorithm will further adjust the center coordinates and radius parameters based on the fitting results of the preprocessed image. First, the preliminary circular fitting results will be analyzed according to the errors of the image, and the error value will be calculated by comparing the difference between the fitted circle and the actual image; then, the optimization algorithm will iteratively adjust the values ​​of the center coordinates and radius to reduce these errors, thereby improving the accuracy of the fitting results. The optimization process will be repeated many times until the error falls within the preset tolerance range to ensure a more accurate circular fitting result, and finally provide high-precision geometric data for subsequent coaxiality detection.

[0063] As a preferred embodiment of the above, Figure 2 As shown, the images of both ends of a large-sized cylinder are collected, including:

[0064] A10: Install the cameras at both ends of a large cylinder, and the shooting angle after installation covers the entire measurement area of ​​the cylinder;

[0065] A20: It is equipped with a four-sided combination of strip light source and ring light source, which are used to illuminate the bottom reference surface of the large-sized cylinder and the pole tube respectively.

[0066] Specifically, to ensure that the camera can fully capture the key areas at both ends of the large-sized cylinder, the camera needs to be installed at both ends of the cylinder, and the shooting angle should be adjusted so that its viewing angle can cover the entire measurement area of ​​the cylinder to avoid shooting blind spots or image distortion. During installation, the angle of the camera should be precisely adjusted to ensure that high-quality, distortion-free image data can be obtained, providing sufficient information for subsequent analysis. At the same time, in order to ensure the clarity and uniformity of the image, it is necessary to improve the image quality through light source configuration. The configured four-dimensional strip light source and annular light source will be used to illuminate the bottom reference surface and the pole tube of the cylinder respectively, ensuring that during the shooting process, the light can be evenly irradiated on these key components to avoid shadows or overly bright areas, and ensure that the captured image has uniform brightness and clear details. This implementation uses two 12-megapixel CCD chip cameras, each camera is equipped with a 12mmFA high-resolution lens, and the light source selects LED as the lighting source as the preferred light source, which can maximize the technical effect to be achieved by this technical solution, but other scheme selections should also be within the scope of protection of this solution.

[0067] As a preferred embodiment of the above, Figure 3 As shown, the preprocessing process includes:

[0068] B10: grayscale the acquired image;

[0069] B20: Use denoising algorithm to remove noise from the grayscale image;

[0070] B30: Use edge detection algorithm to process the denoised image and extract edge points of key circular areas.

[0071] Specifically, first, the collected image is grayed. This process reduces the complexity of color information by converting the original color image into a gray image, which is convenient for subsequent processing and analysis. After graying, the color value of each pixel in the image is converted into a gray value, thereby reducing the amount of calculation; then, a denoising algorithm is used to remove noise from the grayed image. Noise removal is an important part of image preprocessing. It can eliminate random noise caused by sensors, uneven lighting or other environmental factors in the image. Commonly used denoising algorithms include median filtering, Gaussian filtering, etc. Through these methods, the image can be effectively smoothed, and unnecessary noise can be removed while retaining edge information; finally, the edge detection algorithm is used to process the denoised image and extract the edge points of the key circular area in the image. Edge detection algorithms, such as Canny edge detection or Sobel operator, can help determine the precise position of the edge of the object in the image, especially in morphological analysis. The extraction of circular areas is of great significance. Through edge detection, the outline of the circular area can be clearly outlined, providing an accurate basis for subsequent target recognition or measurement.

[0072] As a preferred embodiment of the above, Figure 4 As shown, a circular fitting algorithm is used to process the acquired image to obtain the center coordinates and radius of the key circular area, including:

[0073] C10: Map each edge point into the parameter space and generate possible circle parameters according to different radius and center coordinates;

[0074] C20: Vote for each edge point and accumulate the corresponding circle center coordinates and radius values ​​in the parameter space;

[0075] C30: Select the center coordinate and radius combination with the highest number of votes as the fitting result to obtain the center coordinate and radius of the key circular area of ​​the large-sized cylinder.

[0076] Specifically, first, each edge point is mapped to the parameter space. By considering different radii and center coordinates, the algorithm generates multiple possible circle parameters. In this process, the coordinate information of the edge point is used to generate candidate circles in the parameter space. This process is achieved by using technologies such as Hough transform, which can effectively parameterize the circle in the image and prepare for subsequent circle fitting. Next, each edge point is "voted" to accumulate the values ​​of the corresponding center coordinates and radius in the parameter space. This is the core step of the circle fitting algorithm. Through the vote of each edge point, it is determined which circle parameter is most likely to match the actual circular area. The voting mechanism will evaluate the rationality of each parameter according to the distribution of edge points, so that the correct circular area can be displayed in the parameter space. Finally, the combination of center coordinates and radius with the highest number of votes is selected as the fitting result. This process ensures that the selected circle parameters are most consistent with the actual situation in the image, that is, their cumulative value (number of votes) in the parameter space is the largest. Finally, the algorithm outputs the center coordinates and radius information of the key circular area of ​​the large-sized cylinder, which is crucial for further target recognition, size measurement or shape analysis.

[0077] As a preferred embodiment of the above embodiment, before calculating the relative position deviation, the center coordinates of each key circular area are checked for spatial consistency by using a multi-view fusion technology.

[0078] As a preferred embodiment of the above, Figure 5 As shown in the figure, before calculating the relative position deviation, the center coordinates of each key circular area are checked for spatial consistency through multi-view fusion technology, including:

[0079] D10: Get the three-dimensional coordinates of the two ends of the large-size cylinder;

[0080] D20: Map the coordinates of the center of the circle fitted at different viewing angles to the same three-dimensional coordinate system;

[0081] D30: align the coordinates of the center of the circle from multiple perspectives and remove abnormal coordinate points;

[0082] D40: If the coordinate error of the circle center after registration is less than the preset threshold, the spatial consistency check is considered to have passed, otherwise the image is re-collected.

[0083] Specifically, first, the three-dimensional spatial coordinates of the two ends of the large-size cylinder are obtained to provide a benchmark for subsequent coordinate alignment and consistency verification. These three-dimensional coordinates are usually obtained through precise three-dimensional scanning equipment and used as reference points to help ensure that the center coordinates at different viewing angles can be accurately aligned; next, the fitted center coordinates at different viewing angles are mapped to the same three-dimensional coordinate system. Since the image acquisition comes from different viewing angles, the center coordinates at each viewing angle may be in different coordinate systems. In order to unify these coordinates, camera calibration and coordinate transformation technology are used to convert the coordinates at each viewing angle into a common three-dimensional coordinate system through rotation matrices and translation matrices. This step ensures that the center coordinates from different viewing angles can be compared and processed in the same coordinate system; on this basis, the center coordinates of multiple viewing angles are mapped to the same three-dimensional coordinate system. The registration is performed and abnormal coordinate points are removed. The registration process is performed by the least squares method or the iterative closest point (ICP) algorithm, aiming to align the center coordinates of the circle under different perspectives as accurately as possible. During the registration process, some abnormal coordinate points caused by noise or errors may be encountered. These points will be removed to ensure the accuracy of the registration results. The outlier detection usually adopts a method based on the standard deviation. In this way, the abnormal points that do not conform to the rules can be effectively removed to ensure the quality of the data. Finally, if the error of the center coordinate after registration is less than the preset threshold, the spatial consistency check is judged to pass. At this time, the error of the center coordinate is already within the acceptable range, which means that the coordinates under different perspectives have good spatial consistency and the check is successful. If the error exceeds the threshold, the image needs to be re-collected and the above process is repeated. Through this series of steps, the consistency of the center coordinates of the circle after multi-perspective fusion in space is ensured, thereby improving the accuracy and reliability of the subsequent relative position deviation calculation.

[0084] As a preferred embodiment of the above, the relative position deviation between the key circular areas is calculated according to the center coordinates and radius of each key circular area, including:

[0085] Taking the center coordinates of the copper column as the reference axis, calculate the radial offsets of the center coordinates of the pole tube and the pole root relative to the reference axis respectively;

[0086] According to the radial offset, the formula Calculate the relative position deviation of each circle center in the two-dimensional plane, where Δ represents the radial offset between two points, (x0, y0) is the reference circle center coordinate, (x i ,y i ) are the coordinates of the center of the circle to be measured.

[0087] Specifically, first, the center coordinates of the copper column are used as the reference axis, and the radial offsets of the center coordinates of the pole tube and the pole root relative to the reference axis are calculated respectively. The center coordinates of the copper column are used as the reference and are set to (x0, y0). Then, the center coordinates of the pole tube and the pole root are recorded as (x i ,y i ), the core idea of ​​radial offset calculation is to find the plane straight line distance between the center coordinates of the circle to be measured and the center of the reference circle. In order to calculate this distance, the formula can be used: This process provides a basis for the subsequent relative position deviation calculation by calculating the distance of each circle center relative to the reference circle center; then, based on the calculated radial offset, we can further evaluate the relative position deviation of each circle center in the two-dimensional plane. This deviation value can quantify the position difference between the circle centers. Generally, the smaller the deviation, the closer the position between the circle centers is to the reference circle center. In this way, the relative position between different circular areas can be accurately evaluated.

[0088] Embodiment 2

[0089] Based on the same inventive concept as the method for detecting the coaxiality of a large-sized cylinder in the aforementioned embodiment, the present invention further provides a system for detecting the coaxiality of a large-sized cylinder, the system comprising:

[0090] The image acquisition module collects images of both ends of the large-sized cylinder and defines the copper column, pole tube and pole root as key circular areas respectively;

[0091] The data acquisition module pre-processes the acquired images and processes the acquired images using a circular fitting algorithm to obtain the center coordinates and radius of the key circular area;

[0092] The deviation calculation module calculates the relative position deviation between the key circular areas according to the center coordinates and radius of each key circular area;

[0093] The result acquisition module determines that the coaxiality is qualified if the relative position deviation of each key circular area is within a preset tolerance range according to the calculation result, otherwise the coaxiality is judged to be unqualified and the test result is obtained.

[0094] The above-mentioned detection system in the present invention can effectively implement the coaxiality detection method of large-size cylinders, and the technical effects that can be achieved are as described in the above-mentioned embodiments, which will not be repeated here.

[0095] As a preferred embodiment of the above, the deviation calculation module includes:

[0096] The offset calculation unit uses the center coordinates of the copper column as the reference axis and calculates the radial offsets of the center coordinates of the pole tube and the pole root relative to the reference axis.

[0097] The position deviation determination unit is based on the radial offset through the formula Calculate the relative position deviation of each circle center in the two-dimensional plane, where (x0, y0) is the coordinate of the reference circle center, (x i ,y i ) are the coordinates of the center of the circle to be measured.

[0098] Similarly, the above-mentioned optimization schemes for the system can also respectively achieve the optimization effects corresponding to the method in Example 1, which will not be repeated here.

[0099] Although the present application has been described in conjunction with specific features and embodiments thereof, it is apparent that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the accompanying drawings are merely exemplary illustrations of the present application as defined herein, and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application.

[0100] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these modifications and variations.

Claims

1. A method for detecting the coaxiality of a large-sized cylinder, characterized in that: include: Collect images of both ends of a large-sized cylinder and define the copper column, pole tube, and pole root as key circular areas respectively; Preprocessing the acquired image, and processing the acquired image using a circular fitting algorithm to obtain the center coordinates and radius of the key circular area; Calculating relative position deviations between the key circular areas according to the center coordinates and radius of the key circular areas; According to the calculation results, if the relative position deviation of each of the key circular areas is within a preset tolerance range, the coaxiality is determined to be qualified, otherwise the coaxiality is determined to be unqualified, and the test result is obtained.

2. The method for detecting the coaxiality of a large-size cylinder according to claim 1, characterized in that: Also includes: During the fitting process, the image that still has errors after preprocessing is corrected through the optimization algorithm.

3. The method for detecting the coaxiality of a large-size cylinder according to claim 1, characterized in that: Collect images of both ends of a large cylinder, including: The cameras are installed at both ends of the large-sized cylinder, and the shooting angle after installation covers the measurement area of ​​the entire cylinder; A strip light source and an annular light source combined on four sides are configured to illuminate the bottom reference surface of the large-sized cylinder and the pole tube respectively.

4. The method for detecting the coaxiality of a large-size cylinder according to claim 1, characterized in that: The pretreatment process comprises: Grayscale the acquired image; Using a denoising algorithm to remove noise from the grayscale image; The denoised image is processed using an edge detection algorithm to extract edge points of the key circular area.

5. The method for detecting the coaxiality of a large-sized cylinder according to claim 4, characterized in that: The collected image is processed using a circular fitting algorithm to obtain the center coordinates and radius of the key circular area, including: Mapping each edge point into a parameter space, and generating possible circle parameters according to different radii and circle center coordinates; Voting for each edge point, and accumulating the corresponding circle center coordinates and radius values ​​in the parameter space; The center coordinate and radius combination with the highest number of votes is selected as the fitting result to obtain the center coordinate and radius of the key circular area of ​​the large-sized cylinder.

6. The method for detecting the coaxiality of a large-sized cylinder according to claim 1, characterized in that: Before calculating the relative position deviation, the center coordinates of each of the key circular areas are checked for spatial consistency using multi-view fusion technology.

7. The method for detecting the coaxiality of a large-sized cylinder according to claim 6, characterized in that: Before calculating the relative position deviation, the center coordinates of each key circular area are checked for spatial consistency by using multi-view fusion technology, including: Get the three-dimensional space coordinates of the two ends of the large-size cylinder; Map the coordinates of the center of the circle fitted at different viewing angles to the same three-dimensional coordinate system; Align the coordinates of the center of the circle from multiple perspectives and remove abnormal coordinate points; If the coordinate error of the circle center after registration is less than the preset threshold, the spatial consistency check is considered to have passed, otherwise the image is recaptured.

8. The method for detecting the coaxiality of a large-sized cylinder according to claim 1, characterized in that: Calculating the relative position deviations between the key circular areas according to the center coordinates and radius of the key circular areas includes: Taking the center coordinates of the copper column as the reference axis, respectively calculating the radial offsets of the center coordinates of the pole tube and the pole root relative to the reference axis; According to the radial offset, the formula Calculate the relative position deviation of each circle center in the two-dimensional plane, where Δ represents the radial offset between two points, (x0, y0) is the reference circle center coordinate, (x i ,y i ) are the coordinates of the center of the circle to be measured.

9. A large-size cylindrical coaxiality detection system, characterized in that: The system comprises: The image acquisition module collects images of both ends of the large-sized cylinder and defines the copper column, pole tube and pole root as key circular areas respectively; The data acquisition module pre-processes the acquired image and processes the acquired image using a circular fitting algorithm to obtain the center coordinates and radius of the key circular area; A deviation calculation module, which calculates the relative position deviation between the key circular areas according to the center coordinates and radius of each key circular area; The result acquisition module determines that the coaxiality is qualified if the relative position deviation of each key circular area is within a preset tolerance range according to the calculation result, otherwise the coaxiality is judged to be unqualified, and the detection result is obtained.

10. The large-size cylindrical coaxiality detection system according to claim 9, characterized in that: The deviation calculation module includes: An offset calculation unit, taking the center coordinates of the copper column as a reference axis, and respectively calculating the radial offsets of the center coordinates of the pole tube and the pole root relative to the reference axis; The position deviation determination unit determines the radial offset according to the formula Calculate the relative position deviation of each circle center in the two-dimensional plane, where (x0, y00 is the reference circle center coordinate, (x i ,y i ) are the coordinates of the center of the circle to be measured.

Images